Alcohol consumption - ntakd

Alcohol consumption - ntakd

The ESPAD Report 2003 Alcohol and Other Drug Use Among Students in 35 European Countries Björn Hibell, Barbro Andersson, Thoroddur Bjarnason, Salme A...

5MB Sizes 0 Downloads 0 Views

Recommend Documents

Unregistered alcohol consumption and alcohol-related harm in
Background. In Sweden, alcoholic beverages are sold in special shops run by the State Alcohol Retail. Monopoly (hereafte

Although alcohol consumption proved a - History Matters
on the Social Drinking Culture of the Colonial Chesapeake. Sarah Lynn Kenney ..... An analysis of South Carolina yielded

lithuania - ntakd
The Lithuanian National programme on Drug Control and Prevention of Drug Addiction 2010– ..... 2) prevention of drugs

alcohol consumption in greenland - Syddansk Universitet
Isabelle Mairey: Alcohol Consumption in Greenland – An Empirical Analysis of ... Alcohol is the most used intoxicant i

energy drink consumption (with and without alcohol - UKnowledge
Jul 23, 2012 - founding of Red Bull, Dietrich Mateschitz, its creator, has realized the launch of ... Mateschitz's compa

Alcohol Consumption and Harm: A Consideration of Legal Liability
Alcohol Consumption and Harm: A Consideration of Legal Liability. Relating to the Service and Promotion of Alcohol. Anna

Patterns and dynamics of alcohol consumption during - Superu
Alcohol consumption during pregnancy was not confined to any one .... Recent New Zealand studies have indicated that alc

The Dynamics of Alcohol Consumption in the Russian Federation
Part of the Economics Commons, and the Public Health Commons. This Dissertation is .... Appendix B: Conversion Rates of

Impact of Alcohol Consumption on Young People - Digital Education
Dorothy Newbury-Birch, Janet Walker, Leah Avery,. Fiona Beyer, Nicola Brown, Katherine Jackson,. Catherine A Lock, Ruth

The

ESPAD Report 2003 Alcohol and Other Drug Use Among Students in 35 European Countries Björn Hibell, Barbro Andersson, Thoroddur Bjarnason, Salme Ahlström, Olga Balakireva, Anna Kokkevi, Mark Morgan Contributing authors Airi-Alina Allaste, Thomas Andersen, Karl Bohrn, Anina Chileva, Marie Choquet, Ladislav Csèmy, Aleksandra G. Davidaviciène, Zsuzsanna Elekes, Fernanda Feijão, Silvia Florescu, Gerhard Gmel, Ilze Koroleva, Eugenia Koshkina, Ludwig Kraus, Marina Kuzman, Patrick Lambrecht, Vivienne Mallia, Fabio Mariani, Karin Monshouwer, Richard Muscat, Kamran Niaz, Alojz Nociar, Martin Plant, Svend Sabroe, Janusz Sieroslawski, Astrid Skretting, Eva Stergar, Andreea Steriu, Alfred Uhl, Pál Weihe, Kyriacos Veresies

The Swedish Council for Information on Alcohol and Other Drugs, CAN Council of Europe, Co-operation Group to Combat Drug Abuse and Illicit Trafficking in Drugs (Pompidou Group)

© The Swedish Council for Information on Alcohol and Other Drugs (CAN), The Pompidou Group at the Council of Europe and the authors Printed in Sweden by Modintryckoffset AB, Stockholm November 2004 Printing funded by The Swedish Government Cover design: Thomas Löwenberg Layout: Per-Erik Engström ISBN 91-7278-103-3

Preface

This is the third report published within the ESPAD project. It presents data on more than 100,000 European students in numerous diagrams and maps and around 150 tables. Independent researchers in 35 European countries have collaborated in planning, methodological discussions, the data collections and the reporting of the national results. The two earlier reports presented data from 1995 and 1999. The first report covered 26 European countries, the second included data from 30 countries. The project now covers most of the European continent and has become an important source of information on young people’s alcohol and drug use. Moreover, the body of articles with analyses published in international scientific journals is growing. The enormous data mass now kept in each individual

country will soon be gathered into a common database for further analyses. The work with this report would not have been possible without the economic support from the Swedish Government. We are also grateful for the support we have got from the Pompidou Group at the Council of Europe and the European Monitoring Centre for Drugs and Drug Addiction (EMCDDA) in Lisbon. We would like to take this opportunity to thank our colleagues in all ESPAD countries for the inspiring work, the good spirit and the always friendly and collaborative atmosphere that have characterised our meetings and seminars. We are also grateful to the teachers and huge number of students across Europe that participated in the 2003 data collection.

Stockholm in November 2004 Björn Hibell, Ph.D. Director, ESPAD Co-ordinator

Barbro Andersson Research Associate, ESPAD Co-ordinator

Table of Contents

Acknowledgements Introduction Summary of the 2003 findings Study design and procedures The target population The data collection instrument Sampling procedure Field procedure

Methodological considerations Introduction Representativeness Reliability Validity Comparisons with other survey data Conclusions

Changes in the use of alcohol and other drugs 1995–2003 Changes in cigarette smoking Changes in alcohol consumption Changes in illicit drug use prevalence Changes in perceived availability of drugs Summary

The alcohol and drug situation 2003 Tobacco use Alcohol consumption Expected personal consequences Illicit drugs Lifetime abstinence from various substances Attitudes towards drugs Purchase of alcoholic beverages Perceived cigarette, alcohol and drug use among friends Cigarette, alcohol and drug consumption among elder siblings

Correlates of adolescent substance use Introduction Parental education Family structure Economic situation Parental control Truancy

Table of Contents

9 13 19 27 27 27 28 28 31 31 33 41 45 55 60 63 64 65 89 107 124

129 129 133 160 165 177 182 187 188 189

191 191 191 193 196 197 198

Sibling substance use Summary

198 199

Key results country by country

201 201 202 202 203 203 204 204 205 205 206 206 207 207 208 208 209 209 210 210 211 211 212 212 213 213 214 214 215 215 216 216 217 217 218 218 219

Austria Belgium Bulgaria Croatia Cyprus The Czech Republic Denmark Estonia The Faroe Islands Finland France Germany (six Bundesländer) Greece Greenland Hungary Iceland Ireland Isle of Man Italy Latvia Lithuania Malta The Netherlands Norway Poland Portugal Romania Russia (Moscow) The Slovak Republic Slovenia Sweden Switzerland Turkey (six cities) Ukraine The United Kingdom

References Appendices I. Sampling and data collection I. in participating countries II. Tables III. Student questionnaire

223 293 437

5

Tables A.

Characteristics of the ESPAD surveys in participating countries.

34

B.

Not participating schools and classes, eliminated questionnaires and average time to complete the questionnaire.

37

C.

Participating students and response rates. Numbers and percentages among boys and girls.

40

D.

Some aspects of reliability. Two measures of inconsistency between two questions in a single administration. Percentages and quotients among all students.

43

E.

Opinions of the data collection leaders. Percentages.

47

F.

Proportions of unanswered questions. All students.

49

G.

Some aspects of validity: Inconsistent answers, unwillingness to admit drug use and reported knowledge and use of the dummy drug relevin. Percentages among all students.

51

Alcohol and drug use in Norway. Frequency of lifetime, last 12 months and last 30 days use. Data from ESPAD and a national survey in 2003. Percentages among all respondents.

55

Alcohol and drug use in Sweden. Frequency of lifetime and last 30 days use. Data from ESPAD and the annual Swedish school survey 2003 in grade 9. Percentages among boys and girls.

56

H.

I.

J.

Alcohol and drug use in the Netherlands. Frequency of lifetime, last 12 months and last 30 days use. Data from ESPAD and a parallell school survey (PEIL). Percentages among all respondents.

57

K.

Alcohol use in the ESPAD and HBSC surveys. Students answering 3 times or more often during the last 30 days (ESPAD) or at least weekly (HBSC). Percentages among boys and girls, rxy and Spearmans rangcorrelation cofficient (rrank).

58

L.

Drunkenness in the ESPAD and HBSC surveys. Students who have ever been drunk (ESPAD) and drunk at least twice (HBSC). Percentages among boys and girls, rxy and Spearmans rangcorrelation cofficient (rrank).

58

M.

Lifetime use of cannabis in the ESPAD and HBSC surveys. Percentages among boys and girls, rxy and Spearmans rangcorrelation cofficient (rrank).

59

N.

12 months prevalence of cannabis use in the ESPAD and HBSC surveys. Percentages among boys and girls, rxy and Spearmans rangcorrelation cofficient (rrank).

59

O.

Drunkenness measured with a new “softer” and an old translation in a splithalf test in Moscow schools in 2003.

274

Figures 1a–c.

Changes between 1995 and 2003 in lifetime use of cigarettes 40 times or more. Percentages among boys, girls and all students.

66

6a–c.

Changes between 1995 and 2003 in use of any alcoholic beverages 10 times or more during the last 30 days. Percentages among boys, girls and all students.

76

2a–c.

Changes between 1995 and 2003 in cigarette smoking during the last 30 days. Percentages among boys, girls and all students.

68

7a–c.

Changes between 1995 and 2003 in beer consumption 3 times or more during the last 30 days. Percentages among boys, girls and all students.

78

4a–c.

Changes between 1995 and 2003 in lifetime use of any alcoholic beverages 40 times or more. Percentages among boys, girls and all students.

72

8a–c.

Changes between 1999 and 2003 in wine consumption 3 times or more during the last 30 days. Percentages among boys, girls and all students 2003.

82

5a–c.

Changes between 1995 and 2003 in use of any alcoholic beverages 20 times or more during the last 12 months. Percentages among boys, girls and all students.

74

9a–c.

Changes between 1999 and 2003 in consumption of spirits 3 times or more during the last 30 days. Percentages among boys, girls and all students.

84

6

Table of Contents

10a–c. Changes between 1995 and 2003 in consumption of 101 cl beer or more on the last drinking occasion. Percentages among boys, girls and all students.

86

11a–b. Changes between 1995 and 2003 in consumption of 101 cl alcopops or more on the last drinking occasion. Percentages among boys, girls and all students.

24a–c. Changes between 1995 and 2003 in lifetime experience of alcohol together with pills.

116

25a–c. Changes between 1995 and 2003 in lifetime experience of inhalants.

118

88

120

12a–c. Changes between 1995 and 2003 in consumption of 15 cl wine or more on the last drinking occasion. Percentages among boys, girls and all students.

90

26a–c. Changes between 1995 and 2003 in the proportion of all students who perceive inhalants “very easy” or “fairly easy” to obtain.

122

13a–c. Changes between 1995 and 2003 in consumption of 11 cl spirits or more on the last drinking occasion. Percentages among boys, girls and all students.

92

27a–c. Changes between 1995 and 2003 in the proportion of all students who perceive marijuana or hashish “very easy” or “fairly easy” to obtain.

126

14a–c. Changes between 1995 and 2003 in the proportion who have been drunk 20 times or more in lifetime. Percentages among boys, girls and all students.

94

28a–c. Changes between 1995 and 2003 in the proportion of all students who perceive LSD or other hallucinogens “very easy” or “fairly easy” to obtain. 29a–b. Lifetime use of cigarettes 40 times or more. Percentages among boys, girls and all students. 2003.

130

15a–c. Changes between 1995 and 2003 in the proportion who have been drunk 10 times or more during last 12 months. Percentages among boys, girls and all students.

96

30a–b. Cigarette smoking during the last 30 days. Percentages among boys, girls and all students. 2003.

132

16a–c. Changes between 1995 and 2003 in the proportion who have been drunk 3 times or more during last 30 days. Percentages among boys, girls and all students.

98

31a–b. Daily smoking at the age of 13 or younger. Percentages among boys, girls anf all students. 2003.

134

135

17a–c. Changes between 1995 and 2003 in the proportion who have reported “binge drinking” 3 times or more during last 30 days. Percentages among boys, girls and all students. 2003.

100

32a–b. Lifetime use of any alcoholic beverage 40 times or more. Percentages among boys, girls and all students. 2003.

136

18a–c. Changes between 1995 and 2003 in the proportion of all students who have been drunk at the age of 13 or younger. Percentages among boys, girls and all students.

102

33a–b. Use of any alcoholic beverage 20 times or more during the last 12 months. Percentages among boys, girls and all students. 2003. 34a–b. Use of any alcoholic beverage 10 times or more during the last 30 days. Percentages among boys, girls all students. 2003.

138

19a–c. Changes between 1995 and 2003 in lifetime experience of any illicit drug.

104

139

20a–c. Changes between 1995 and 2003 in lifetime experience of marijuana or hashish.

108

35a–b. Beer consumption 3 times or more during the last 30 days. Percentages among boys, girls and all students. 2003.

141

21a–c. Changes between 1995 and 2003 in the proportion of all students who have used marijuana or hashish during the last 30 days.

110

36a–b. Wine consumption 3 times or more during the last 30 days. Percentages among boys, girls aand all students. 2003.

142

22a–c. Changes between 1995 and 2003 in lifetime experience of any illicit drug other than marijuana or hashish.

112

37a–b. Consumption of spirits 3 times or more during the last 30 days. Percentages among boys, girls and all students. 2003.

144

23a–c. Changes between 1995 and 2003 in lifetime experience of tranquillisers or sedatives without a doctor’s prescription.

114

38a–b. Consumption of 101 cl beer or more on the last drinking occasion. Percentages among boys, girls and all students. 2003. 39a–b. Consumption of 15 cl wine or more on the last drinking occasion. Percentages among boys, girls and all students. 2003.

146

Table of Contents

7

40a–b. Consumption of 11 cl of spirits or more on the last drinking occasion. Percentages among boys, girls and all students. 2003.

147

50a–b. Lifetime experience of any illicit drug. Percentages among boys, girls and all students. 2003.

164

41a–b. Estimated average consumption of beer, wine and spirits, in cl 100% alcohol, on the last drinking occasion. Boys, girls and all students. 2003.

150

51a–b. Lifetime experience of marijuana or hashish. Percentages among boys, girls and all students. 2003.

166

152

52a–b. Proportion of boys, girls and all students who have used marijuana or hashish during the last 30 days.2003.

167

42a–b. Proportion of boys and girls who have been drunk 20 times or more in lifetime. 2003.

153

53a–b. Lifetime experience of any illicit drug other than marijuana or hashish. Percentages among boys, girls and all students. 2003.

169

43a–b. Proportion of boys, girls and all students who have been drunk 10 times or more during last 12 months. 2003.

155

54a–b. Lifetime experience of tranquillisers or sedatives without a doctors prescription. Percentages among boys, girls and all students. 2003.

171

44a–b. Proportion of boys, girls and all students who have been drunk 3 times or more during the last 30 days. 2003. 45a–b. Proportion of boys, girls and all students who reported “binge drinking” 3 times or more during the last 30 days. 2003.

156

55a–b. Lifetime experience of alcohol together with pills. Percentages among boys, girls and all students. 2003.

172

46a–b. Proportion of boys, girls and all students who have been drunk at the age of 13 or younger. 2003.

159

56a–b. Lifetime experience of inhalants. Percentages among boys, girls and all students. 2003.

173

47.

Anticipated positive and negative consequences of alcohol consumption. Number of statements for which the percentage of all students answering “likely” or “very likely” exceeds the average of all countries.

160

57.

The proportion of students who reported that cannabis was easily available in schools and any other place respectively.

175

58a–b. Proportion of boys, girls and all students who perceive inhalants “very easy” or “fairly easy” to obtain. 2003.

179

Experienced problems caused by alcohol. The number of variables within each “problem group” for which a country’s percentage exceeds the average of all countries. All students.

162

59a–b. Proportion of boys, girls all students who perceive marijuana or hashish “very easy” or “fairly easy” to obtain. 2003.

180

181

Cross-national average of students who report having each of 14 problems because of their alcohol use and because of other reasons.

162

60a–b. Proportion of boys, girls all students who perceive LSD or other hallucinogen “very easy” or “fairly easy” to obtain. 2003.

48.

49.

8

Table of Contents

Acknowledgements

The planning and implementation of the ESPAD 2003 project has been a collaborative effort of the research teams in each of the 35 countries. The importance of the researchers and their supporting institutions cannot be over-estimated. The project has no money for the participating countries, but rely on each participant’s possibility to raise money within his or her country. The Swedish Government, however, has supported the co-ordination of the work. This grant has also covered some of the costs for the meetings and for the printing of the report. The Pompidou Group at the Council of Europe has supported the project from the first project meeting in 1994. In particular the support of the Pompidou Group has enabled researchers from eastern and central parts of Europe to participate in project meetings and regional seminars. Earlier the Pompidou Group also supported the meetings of the working group. A special thanks is due to Mr. Christopher Luckett and Mrs. Florence MabileauWhomsley at the Pompidou Group for their much appreciated assistance and support. We would also like to thank Richard Muscat at the Pompidou Group, for his valuable comments on this report. Some support has also been given by the European Monitoring Centre for Drugs and Drug Ad-

diction (EMCDDA) in Lisbon. The ESPAD project has been co-ordinated by Björn Hibell and Barbro Andersson, Sweden. A working group, nominated by the ESPAD members at project meetings, has been collaborating with the co-ordinators of this project. These persons have been planning, developing and discussing all details in relation both to the project meetings and to the planning of the 2003 data collection. The working group has also actively taken part in the development of this report. Besides the co-ordinators the members of the working group include Salme Ahlström, Finland, Olga Balakireva, Ukraine, Thoroddur Bjarnason, Iceland, Anna Kokkevi, Greece and Mark Morgan, Ireland. In addition to the results of the ESPAD 2003 survey, some comparable data from the Monitoring the Future Project in USA, and from a Spanish school survey has been included in this report. Lloyd Johnston, USA and Gregorio Barrio Anta, Spain have kindly provided these data. Each country has been represented in the project by a researcher, who is also a contributing author of this report (see title page). There are, however, a number of persons who have done important work with the 2003 ESPAD study. They are presented below in alphabetical order by country.

Collaborating Persons Austria: Gerhard Grimm, Regina Fenk, Margot Koller, Elke Lantschik, Conny Auer, Sabine Haas Belgium: Caroline Andries, Danielle Piette, Laurence Kohn, Christelle Senterre, Patrick de Smet, Tim Engels, Sylvia De Mey Bulgaria: Krassimir Kalinov, Momtchil Vassilev, Georgi Vasilev, Mirosslav Popov, Alexander Hodonovski, Julian Nakov, Rossitza Penkova Croatia: Ivana Pavic Simetin, Iva Pejnovic Franelic, Mario Hemen, Tanja Coric, Mario Hemen, Slavko Sakoman, Marina Políc-Vizintin, Vesna Jureša Cyprus: Andreas Pavlakis, Ioanna Yiasemi, George Poyiadjis, Panayiota Gregoriou

Acknowledgements

Czech Republic: Petr Sakilek, Hana Sovinova, Pavla Lejckova, Jitka Veralova, Zuzana Dvorakova Denmark: Kirsten Fonager Estonia: Niina Derman, Kati Tillemann, Aljona Kurbatova, Riin Undusk, Uno Saar, Eda Heinla Faroe Islands: Firouz Gaini, Elsa Olsen, Ronny Jacobsen, Mari Ann Ellendersen, Anna Sofia Veyhe, John Kjaerbaek Finland: Leena Metso, Eeva Liisa Tuovinen, Kari Törmäkangas, Eija Puhakka France: Francois Beck, Christine Hassler, Delphine Morin, Jean-Michel Costes, Stanislas Spilka, Stéphane Legleye

9

Germany: Rita Augustin, Boris Orth, Kathrin Heppekausen, Andrea E. Barrera Greece: Manina Terzidou, Clive Richardson, Anastasios Fotiou, Ioannis Tasioulas, Maria Spyropoulou, Fevos Zaferides Greenland: Charlotte Lycke Hungary: Borbála Paksi, Bálint Ferenczhalmy, Adrienn Nyirády Iceland: Stefan Hrafn Jonsson, Thorolfur Thorlindsson, Inga Dora Sigfusdottir Ireland: Paul Downes Isle of Man: Jane Powell, Patrick Miller, Martin Plant, Jan Green, Jim Green, Moira Plant, John Cain, Penny Creighton, PA Burnett, ME Tomaszewska, DJ Trace, SM Moore, JG Quayle, P John, JR Smith, Sue Barry, Marilyn Keary, Lyz Moore, Marilyn Payne, Lindsay Riordan, Margaret O’Reilly, Ann Easthope, Ian MacLean, Cheryl McKeon, Sheila Sharpe Italy: S. Molinaro, F. Denoth, S. Salvatori, C. Doveri, R. Panini, C. Imiotti, M. Rial, L. Pellegrino, S. Pardini, I. Masoni, C. Gatti, L. Bastiani, C. Sbrana, F. Cini, S. Gazzetti Latvia: Ilona Adamaite, Ritma Rungule, Maris Goldmanis, Marcis Trapencieris Lithuania: Rasa Juozapaityte, Algimantas Simaitis Malta: Sina Bugeja, Albert Bell, Sharon Arpa, Josephine Baldacchino, Edmund Pace

The Netherlands: W.A.M. Vollebergh, K. Monshouwer, S. van Dorsselaer, A. Gorter Norway: Elin K Bye Poland: Antoni Zielinski, Katarzyna Dabrowska Portugal: Elsa Lavado Romania: Martian Bogdan, Iuliana Robu, Marian Lupan, Marius Ciutan, Cassandra Butu Russia: Konstantin Vyshinsky, N N Ivantes, L P Kezina, E Yu Pavlova, V V Yakushev The Slovak Republic: Ján Luha, Jana Hamade, Lubica Bíziková, Marcela Bieliková, Eva Tomková Slovenia: Vesna Pucelj Switzerland: Emmanuel Kuntsche, Esther Grichting, Jürgen Rehm, Edith Bacher, Marina Delgrande Jordan, Elisabeth Grisel, Holger Schmid, Dominik Bolliger, Heidi Bolliger, Irene Bosshard, Dora Hamann Turkey: Zeynep Basrankut, Ferdi Tanir, Nesrin Dilbaz, Ali Ceylan, Sevil Atasoy, Yildiz Akvardar, Yildis Peksen Ukraine: Alexander Yaremenko, Oksan Artukh, Tetyana Bondar, Pavel Duplenko, Maryana Ryabova, Lidia Romanovska The United Kingdom: Patrick Miller, Moira Plant, Jan Green, Jim Green, George Carcy Spain: Gregorio Barrio Anta USA: Lloyd D Johnston

Funding agencies and supportive organisations Austria: Ludwig-Boltzmann-Institut für Suchtforschung (LBI Sucht), Institute for Social and Health psychology (ISG), Federal Ministry for Health and Women, Federal Ministry for Education, Science and Culture, Austraian Reitox Focal Point (OEBIG), Landesschulräte für Niederösterreich, Oberösterreich, Salzburg, Steiermark, Kärnten, Tirol, Vorarlberg, Burgenland, Stadtschulrat für Wien Belgium: The Scientific Institute of Public Health, the Ministry of Welfare, Public Health and Culture of the Government of Flanders – Flemish Community, the Ministry of Health of the French-speaking Community of Belgium, the Research Council of the Vrije Universiteit Brussel Bulgaria: National Programme for Prevention, Treatment and Rehabilitation of Drug Addictions in Republic of Bulgaria 2001–2005, Ministry of Health, National Centre of Public Health, National Centre for Drug Addictions, Ministry of Education, National Institute of Education Croatia: Croatian National Institute of Public Health, The Government of the City of Zagreb, The Governmental Commission for Drug Prevention Cyprus: KENTHEA, the Ministry of Education

10

The Czech Republic: Prague Psychiatric Centre, INRES-SONES Agency, National Institute of Public Health, the National Drug Commission, the Czech National Focal Point for Drugs and Drug Addiction (NFP) Denmark: Department of Epidemiology and Social Medicine at the University of Aarhus, Department of Social Medicine, Unit of Social Medicine at the Department of Public Health Estonia: Institute of International and Social Studies, Foundation of Public Health Study and Development, Ministry of Social Affairs, Ministry of Education The Faroe Islands: The Department of Occupational and Public Health, the Health Promotion Advisory Committee of the Faroes Finland: The National Research and Development Centre for Welfare and Health (STAKES) France: Ministry of Education, Secrétariat Général del’Enseignement Catholique, Institut National de la Santé et de la recherche Médicale (INSERM), Observatoire Français des Drogues et des Toxicomanies (OFDT), Institut de Recherche sur les boissons (IREB), Ministry of Youth and Education

Acknowledgements

(MJEN), Conseil Supérieur de l’Audiovisuel (CSA) Germany: The Federal Ministry of Health, the Ministries of Education of six German Bundeslaender, the Institute for Therapy Research (IFT) Greece: Greek Organisation Against Drugs (OKANA), Secondary Education Directorate, Ministry of Education, Pedagogical Institute (Research, Documentation and Educational Technology Unit), University of Thessaloniki, University Mental Health Research Institute Greenland: PAARISA Hungary: Behaviour Research Institute at the Budapest University of Economic Sciences and Public Administration, National Foundation of Scientific Researches, Ministry of Children, Youth and Sport Iceland: The Icelandic Alcohol and Drug Abuse Prevention Council, the Centre for Social Research and Analysis, City of Reykjavik, the University at Albany Ireland: St. Patrick’s College in Dublin, the Department of Health and Children Isle of Man: The Department of Education, the Alcohol & Health Research Trust at Centre for Research in Public Health & Primary Care Development of University of the West of England, Bristol, the Chief Secretary’s Drug and Alcohol Co-ordinator’s Office, the Department of Health and Social Security, the Directorate of Public Health, Dunedin Data Services, Edinburgh Italy: National Research Council (CNR), Institute of Clinical Physiology, Research Centre for the Supply of Health Services (IFC-CREAS), Unit of Epidemiology and Health Service Research Latvia: The Latvian Drug Abuse Prevention and Treatment Centre (Narcology Centre), the Institute of Philosophy and Sociology of Latvia (IPHS) Lithuania: Ministry of Education and Science, Public Health Services of the Ministry of Health. Malta: The Policy and Services Development Division at sedqa, Agency Against Drug and Alcohol Abuse, the Guidance and Counselling Services and the Anti-Substance Abuse Unit both pertaining to the Education Division of the Ministry of Education and Human Resources Netherlands: Trimbos Institute, Regional Health Services, the Dutch Ministry of Health, Welfare and Sport

Acknowledgements

Norway: The Norwegian Institute for Alcohol and Drug Research, Norsk Gallup A/S Poland: Institute of Psychiatry and Neurology in Warsaw, Bureau for Drug Addiction, State Agency for Prevention of Alcohol Problems Portugal: Instituto da Droga e da Toxicodependencia (IDT) (Portuguese Institute for Drug and Drug Addiction) Romania: Ministry of Education, Research and Youth, Ministry of Health, the Public Health Districts through the Health Promotion and Programmes Departments, the National Institute for Research and Development in Health Russia: Pompidou Group at the Council of Europe, National Research Centre on Addictions, Russian Federation Ministry of Health, Moscow Education Committee, Moscow City Government, Committee on Telecommunications and Mass Media The Slovak Republic: National Monitoring Centre for Drugs at the Office of the Government, Institute for Public Opinion Research at the Statistical Office of the Slovak Republic, State Health Institute, Institute of Information and Educational Prognoses of School Sector, Ministry of Education Slovenia: The Institute of Public Health Sweden: The Alcohol Research Fund of the Swedish Alcohol Retailing Monopoly, the Swedish National Drug Policy Coordinator, the Swedish Council for Information on Alcohol and other Drugs (CAN) Switzerland: Swiss Institute for the Prevention of Alcohol and Druguse (SIPA) in Lausanne, Addiction Research Institute (ARI) in Zurich Turkey: United Nations Office for Drugs and Crime (UNODC), Ministry of Foreign Affairs, Directorate General of Treatment Services at the Ministry of Health, Ministry of National Education Ukraine: Ukrainian Institute for Social Research, State Institute for Family and Youth, Social Monitoring Centre The United Kingdom: Alcohol and Health Research Trust at the University of the West of England, the Wates Foundation, the Joseph Rowntree Foundation, the Oakdale Trust, the North British Distillery Company, Butcombe Brewery Ltd, the Jack Goldhill Charitable Trust, Medicash and R & J Lass Charities

11

Introduction

Health effects of tobacco, alcohol and drug consumption are apparent on the individual as well as the societal level as a whole. The negative aspects are of great concern in municipalities and countries and for that matter the international community. Governments and major international bodies as the United Nations and the European Union are constantly looking for policy measures to reduce the negative impact of the use of different substances. The wellbeing of young people is of special concern in all societies and ongoing efforts are made to reduce all types of dangerous behaviour. These include many aspects of the consumption of tobacco, alcohol and different kinds of illegal drugs. Most countries have laws in place that restrict the availability of these substances. The legal regulations may vary between countries but many of them include limitations especially targeted to young people. The wellbeing of young people is visible in the Action plans of the European Union. The first covered the years from 1995 to 1999 and the second, the period from 2000 to 2004. A new plan from 2005 is in the preparative stage. The plan for 2000– 2004 included the following six targets: • To reduce significantly over five years the prevalence of illicit drug use, as well as new recruitment to it, particularly among young people under 18 years of age. • To reduce substantially over five years the incidence of drug-related health damage (HIV, hepatitis B and C, TBC, etc.) and the number of drug-related deaths. • To increase substantially the number of successfully treated addicts. • To reduce substantially over five years the availability of illicit drugs. • To reduce substantially over five years the number of drug related crimes. • To reduce substantially over five years moneylaundering and illicit trafficking of precursors. The European Union established the European Monitoring Centre for Drugs and Drug Addiction (EMCDDA) in Lisbon. The centre is responsible

Introduction

for supplying objective, reliable and comparable data to provide the Community and member states with an overall view of drugs, drug addiction and their consequences. The tasks of EMCDDA include; to collect and analyse existing data, to improve datacomparison methods, to disseminate data and to cooperate with European and international organisations and third countries. WHO formulated a European Alcohol action plan for the years 2000 to 2005 with the aim to reduce the harm caused by alcohol. To complement this broad plan a declaration on young people and alcohol was released in 2001. The declaration includes the following targets: • To substantially reduce the number of young people who start consuming alcohol. • To delay the age of onset of drinking by young people. • To substantially reduce the occurrence and frequency of high-risk drinking among young people, especially adolescents and young adults. • To increase education for young people on alcohol. • To substantially reduce alcohol-related harm, especially accidents, assaults and violence, and particularly as experienced by young people. The Pompidou Group at the Council of Europe provides a forum for European ministers, officials and other professionals to co-operate and exchange information about drugs. The main mission is the facilitation of the triangulation between policy, practice and research with the aim to promote evidence-based policy with focus on day-to-day practice as well as local level policy and practice. Platforms are the main instruments through which the mission of the Pompidou Group has been implemented. The functions of the research platform includes to signal developments in the use of data and research as a basis for policy and practice. In relation to the ESPAD project this includes examination of the impact of the ESPAD project on policy and practice and to better understand risk factors and communicate this information to policymakers and practitioners to elaborate evidence-

13

based prevention policies and programmes. The ESPAD project can play a key role in relation to the actions proposed by all these actors. One of the goals of the ESPAD project is to provide data that can be used as a part of the evaluation of the EU action plan on drugs as well as the WHO Europe declaration on young people and alcohol. In relation to the evaluation of the EU action plan co-operation with EMCDDA is essential. The same is true in relation to the Pompidou Group and its role to promote evidence-based drug policy measures.

There is a growing concern from policy makers and other decision makers about the negative effects of young peoples’ consumption of different substances. Informed and well supported decisions demand comprehensive information, which is a key mission for the ESPAD project. With three data collections in 1995, 1999 and 2003 the ESPAD project provides a reliable overview of trends in licit and illicit drug use among European adolescents between 1995–2003 as well as a comprehensive picture of young peoples’ use of tobacco, alcohol, cannabis and other drugs in Europe.

Background The use of tobacco, alcohol and other drugs among young people is of great concern in most countries and many studies have been conducted to better understand consumption patterns. Traditionally, in spite of the significant number of studies conducted in many countries, it was rather difficult to obtain a comprehensive picture and more to the point compare the levels of alcohol and drug use prevalence in different countries. The main reason for this was that the studies involved different age groups with different questionnaires and at different times, i.e. too many disparate factors that made comparisons difficult. During the 1980’s a subgroup of collaborating investigators was formed within the Pompidou Expert Committee on Drug Epidemiology, Council of Europe, to develop a standardised school survey questionnaire and methodology. The purpose and rationale for the work was to produce a standard survey instrument, which would permit different countries to compare alcohol and drug use in student populations. The common questionnaire was used by eight countries in a pilot study. Unfortunately the studies differed in sample size, representativeness and range of ages studied and they were not performed simultaneously. Due to these differences data were not directly comparable. However, the survey instrument proved to be valid and reliable (Johnston et al. 1994). Another study, who’s primary objective is the

14

health behaviour of children in Europe (aged 11, 13 and 15), was initiated by a small group of researchers in the beginning of the 1980s. The project was adopted by WHO and now has an increasing number of countries involved in it. Surveys have been conducted since 1983/84 and to date total some six, the last one in 2001/02. However, the focus of these studies is mainly health issues, although in later studies a few questions were asked on smoking, alcohol consumption and cannabis use (Currie et. al. 2004). Some few countries conduct school surveys on a more or less regularl basis. However, the long series of annual school surveys in Sweden since 1971 is unique. Over the years however there has been a growing interest to compare the results from the Swedish school surveys with comparable data from other countries. In the light of the experiences described above, the Swedish Council for Information on Alcohol and Other Drugs (CAN) initiated a collaborative project in 1993 by contacting researchers in most European countries, to explore the possibility of simultaneously performed school surveys on tobacco, alcohol and drugs in co-operation with the Pompidou Group. These contacts resulted in the first ESPAD study involving 26 European countries in 1995. The second study was conducted in 1999.

Introduction

Purpose of the project A main purpose of the ESPAD project is to collect comparable data on alcohol, tobacco and drug use among 15–16 year old students in European countries. The studies are conducted as school surveys by researchers in each participating country, during the same period of time and with a common methodology. By adopting this ESPAD format, comprehensive and comparable data on alcohol, tobacco and drug use among European students are produced. The most important goal of this project is to monitor trends in alcohol and drug habits among students in Europe and to compare trends between countries and between groups of countries. The knowledge thus gained will be important in the future when changes in one part of Europe may serve as a possible forecast for other countries where changes have not yet appeared. Such trends

may also function as the basis for future prevention initiatives. In relations to the EU action plan on drugs and the WHO Europe declaration on young people and alcohol, a third goal of the ESPAD project is to provide data that can be used as a part of the evaluation of these charters. The surveys are planned to be repeated every fourth year, thus providing long-term data on changes in alcohol and drug consumption among young people. The collected data should also be analysed in depth for a better understanding of young peoples’ alcohol and drug behaviour. European countries which are not yet involved in the ESPAD project are welcome to join the next wave in 2007, to further the coverage across Europe as completely as possible.

The use of surveys Knowledge pertaining to the levels of alcohol and drug use can be derived in different ways depending on which part of the phenomenon one wants to address. In many countries household surveys are conducted with the aim of measuring alcohol and drug habits in general populations. School surveys are also often performed, either complementary to other investigations or as the only measure. A problem with surveys is that they usually do not reach some segments of the population, including heavy abuser populations, homeless or dropouts from school. The latter is a group of young persons known to be vulnerable to alcohol and drug use. There are, however, other techniques available to measure drug use among these populations, e.g. snowball sampling, first treatment demand rates or estimates based on capture-recapture methods. The rationale for school surveys is that students represent age-groups when onset of different substance use is likely to occur and therefore important to monitor. Another reason is ease of accessibility, students are as such within the school system, which

Introduction

also reduces the costs. With student studies, it is a well accepted method to use group administrated questionnaires in a classroom setting where data are collected under the same conditions as a written test. The experience of using school surveys to collect information on alcohol and drug use certainly differs between countries. However, when students are the selected population for study, there are usually no other realistic ways of collecting data other than using group administrated questionnaires in the schools (usually in the classrooms). A handbook on the methods usually required in the conduct of school surveys on drug abuse has recently been published by United Nations Office on Drugs and Crime (Hibell et al 2003). It includes information on the planning of school surveys, methodological issues, sampling issues, questionnaire development, data collection procedure as well as report writing.

15

National project plans and regional seminars Prior to the survey each country produced a national project plan, following a standardised outline, describing the target population’s distribution over the grades in school and the proportion of students expected to be enrolled in school (Hibell and Andersson 2002). The plans for sampling and field procedures were also described in detail. In an effort to standardise the methodology regional seminars were held with small groups of

investigators. The purpose of the seminars was to maximise the standardisation of the data collection procedure and to discuss and suggest which of the sampling procedures were most appropriate for the different countries with different conditions in terms of available school statistics. The seminars per se also functioned as training courses for the less experienced participants.

Participants and ownership Each researcher raised funds in his or her own country and participated in the project and at project meetings independently and at own costs. Data collected in the project are owned by each country

independently. The co-ordination of the project is financed by a mutual agreement between the Swedish Council for Information on Alcohol and Other Drugs (CAN) and the Swedish Government.

Participating countries About 30 countries were involved in the planning process of the 1995 ESPAD study. Unfortunately a few of them were unable to raise the funding needed for data collection and thus the 1995 ESPAD Report included information gathered from 26 countries (Hibell et al 1997). In the second round of data collection held in 1999 data was collected from 30 countries. For the 2003 survey, new countries have joined and this report includes data from 35 participating countries including Austria, Belgium, Bulgaria, Croatia, Cyprus, the Czech Republic, Denmark, Estonia, the Faroe Islands, Finland, France, Germany, Greece, Greenland, Hungary, Iceland, Ireland, Isle of Man, Italy, Latvia, Lithuania, Malta, the Netherlands,

Norway, Poland, Portugal, Romania, Russia (Moscow), the Slovak Republic, Slovenia, Sweden, Switzerland, Turkey, Ukraine and United Kingdom. Five of these countries participated in the ESPAD project for the first time in 2003. They are Austria, Belgium, Germany, Isle of Man and Switzerland. Turkey collected data in 1995, but not in 1999, and re-joined for the 2003 survey. One country (FYROM – Former Yugoslav Republic of Macedonia) that participated in the 1999 study did not take part in the 2003 data collection exercise. Besides the 35 ESPAD countries the report also includes data from Spain and USA.

The structure of the 2003 ESPAD report The structure of this report follows to a large extent the structure of previous ESPAD reports. A major difference is a new more analytical chapter about the relationship between some background variables and the consumption of alcohol and other drugs.

16

Moreover, one of the first chapters includes an overview of the study design and procedures. As mentioned earlier, a goal of the ESPAD project has of course been to standardise the procedures as much as possible, including the target population, the questionnaire, the sampling procedure as well

Introduction

as the way in which data are collected. A complement to this overview can be found in Appendix I in which the sampling and field procedures are presented and commented on country by country. Changes between the three data collections in 1995, 1999 and 2003 are presented in the first of the result chapters. This is the only part of the report that includes data from previous data collections. (An exception is the last of the tables in the table section, where recalculated data on estimates for alcohol consumption from the 1999 study are presented.) To give an overview of major changes from 1999 to 2003 in the countries that participated in both studies the chapter is made more explicit by the significant use of a number of diagrams. In addition to this, a new type of diagram has been introduced that provides information on the trends between all the three data collections country by country. Major results from the 2003 data collection are presented in a separate chapter. As in previous reports, it includes maps that illustrate the differences between high and low prevalence countries

Introduction

for a large number of variables. The maps are complemented by bar graphs that _rank" all countries with available information. The key results for individual countries are gathered in a separate chapter. It includes a country by country overview in which the findings of each country are compared with the averages of all 35 ESPAD countries. Some of the most relevant variables describing the alcohol and drug situation among students across Europe are summarised in a short chapter. The overview includes information on cigarette smoking, alcohol consumption, drunkenness as well as the use of cannabis and other illicit drugs. The last chapter includes correlates of adolescent substance use. The use of cigarettes, alcohol and cannabis use correlated to parental education, family structure, economic situation, parental control, truancy and sibling substance use. The tables of the methodological chapter are presented in the text. However, the tables that include data related to the consumption of alcohol and other drugs are to be found in Appendix II.

17

Summary of the 2003 findings

Data on young people’s alcohol and drug habits have been collected in three waves of the European School Survey Project on Alcohol and Other Drugs, ESPAD. The first study was conducted in 26 countries in 1995. The second survey was done in 1999 and reached 30 participating countries. The focus of this chapter is on the findings from the surveys that were performed in 35 countries in 2003. The participating countries include Austria, Belgium, Bulgaria, Croatia, Cyprus, the Czech Republic, Denmark, Estonia, the Faroe Islands, Finland, France, Germany (6 Bundesländer), Greece, Greenland, Hungary, Iceland, Ireland, Isle of Man, Italy, Latvia, Lithuania, Malta, the Netherlands, Norway, Poland, Portugal, Romania, Russia (Moscow), the Slovak Republic, Slovenia, Sweden, Switzerland, Turkey (6 cities), Ukraine and the United Kingdom. The project is a collaborative project between independent research teams in the participating countries. More than 100,000 students participated in the 2003 data collection. In this chapter a short version of the 2003 findings is presented. Key data on important variables are presented in summary tables 1–3. The behaviours included are cigarette smoking, alcohol consumption, drunkenness and use of illicit drugs.

Methodology As in earlier studies, the surveys were conducted with a standardised methodology and a common questionnaire to provide as comparable data as possible. Data were mainly collected during Spring 2003 and the target population was students born in 1987. Thus, the age group studied turned 16 during the year of data collection. At the time of the data collections the average age was 15.8 years. Data were collected by group-administered questionnaires in schools on nationally representative samples of classes. Exceptions include Russia, where the study was restricted to Moscow only, Germany, where the study was performed in six Bundesländer and Turkey, where the study was restricted to six major cities in the six main regions in Turkey. Teachers or research assistants collected the data.

Summary of the 2003 findings

The students answered the questionnaires anonymously in the classroom under conditions similar to a written test. The sample sizes in participating countries ranges between 555 in Greenland to almost 6,000 in Poland. However, small study groups are only found in small countries where no sampling was done. In all remaining countries, the sample size was close to or above the recommended number of 2,400. The results of the survey were reported in a standardised format. These country reports form the basis of the content of this report.

Data quality Every effort was made to standardise the methodology of the ESPAD project across countries. Nevertheless, some methodological issues inevitably arise in a comparative survey of 35 countries. The validity is deemed to be high in most ESPAD countries. The cultural context in which the students have answered the questions has most probably differed between countries. However, this does not necessarily indicate large differences in the willingness to give honest answers. A few countries have experienced modest validity problems, but such problems are not of the magnitude necessary to seriously threaten the comparability of results. For various reasons it was not possible to give precise levels of statistical significance in this report. Small differences in point estimates between countries or over time should therefore be interpreted with caution. However, given the size of the national samples and the sampling methods employed, differences of more than a few percentage points can with considerable confidence be considered significant.

Tobacco he use of cigarettes 40 times or more in lifetime and the 30 days prevalence rates are presented in the summary tables. In nearly all ESPAD countries 50–80% of the students had smoked cigarettes at least once in their lifetime, and those who had smoked 40 times or more are mainly found in countries where the lifetime prevalence is high. In

19

Austria, the Czech Republic, the Faroe Islands, Greenland, Germany, Lithuania and Russia (Moscow) about 40% had smoked 40 times or more in their lifetime. The lowest prevalence rates are found in Turkey (13%), Malta (16%), Iceland and Portugal (18% each). In eight of the 35 ESPAD countries more boys than girls had smoked 40 times or more in their lifetime. These countries are mainly found in the eastern parts of Europe such as Estonia, Latvia, Lithuania, Poland, Romania and Ukraine, but also in Cyprus and Turkey. Large differences in the other direction with more girls reporting this behaviour are mainly found in two northern islands, Greenland and the Isle of Man. The highest percentage of students, which reported smoking during the last 30 days is found in Greenland, which stands apart from other countries on this variable (60%). High rates are also found in Austria, Bulgaria, Germany, Russia (Moscow) and the Czech Republic (43–49%). Particularly low proportions are found in Cyprus, Iceland, Sweden and Turkey with figures ranging between 18 and 25%. Countries with substantially higher rates of last month smoking among boys include Cyprus, Latvia, Lithuania, Turkey and Ukraine. Considerably higher rates among girls are found in Greenland, Ireland, Isle of Man and the United Kingdom.

Alcohol consumption Prevalence of alcohol consumption 40 times or more in lifetime is presented in the summary tables. They also contain the 30 days prevalence of alcohol consumption 10 times or more, as well as the 30 days prevalence of consuming beer, wine and spirits 3 times or more. In two thirds of the ESPAD countries the vast majority (90% or more) of the students have drank alcohol at least once in their lifetime. However, these students do not all drink on a regular basis. A student who has been drinking at least 40 times can be labelled as more of a regular consumer. The prevalence rates of this frequency of drinking are much lower than the total lifetime prevalence. The highest rates reporting use of alcohol 40 times or more in lifetime are primarily found in the same countries as reported the highest lifetime figures. They include Denmark, Austria, the Czech Republic, Isle of Man, the Netherlands and the United Kingdom (43–50%). The lowest proportion is reported from Turkey (7%) followed by Greenland, Iceland, Norway and Portugal (13–15%).

20

More boys than girls report this level of alcohol consumption. In a few countries, Isle of Man, Fin land and Norway, the gender distribution is about equal. However, no country reports prevalence rates among girls that exceed those of the boys. A higher frequency of alcohol use is revealed among students who had consumed alcohol 10 times or more during the last 30 days, i.e. at least every third day on average. About one quarter of the students in the Netherlands (25%) and about one fifth of the respondents in Austria, Belgium, Malta and the United Kingdom (17–21%) reported this frequency of alcohol use. In some countries, this drinking frequency is hardly reported at all. Proportions of 3% or less were found in Finland, Greenland, Iceland, Norway and Sweden. Thus, the very low prevalence rates are mainly concentrated to the Nordic countries. Many students report rather frequent beer consumption. The percentages of students who had consumed beer 3 times or more during the last 30 days varies between 10 and 44%. The highest figures are found in Denmark, Bulgaria, the Netherlands and Poland (40–44%). The smallest proportions were reported from Norway and Turkey (10 and 14% respectively). Other countries where less than 20% had consumed beer that often include Finland, Hungary, Iceland and Portugal. Drinking beer is a predominantly male behaviour in most ESPAD countries. The only exceptions are two countries in the North Atlantic, Greenland and Iceland, where almost equal proportions of girls and boys report frequent beer drinking. A smaller number of students had been drinking wine than beer during the last 30 days. The proportions of students reporting a wine consumption frequency of 3 times or more during last 30 days are in most cases lower than 20%. However, one country stands out in this respect, as one third (35%) of the students in Malta reported this frequency of wine drinking. Other high prevalence countries include Austria, the Czech Republic, Greece, Italy and Slovenia (21–23%). The lowest proportions that reported this frequency of wine consumption are found in Finland, Iceland, Norway and Turkey (5% or less). The number of students who had been drinking spirits during the last 30 days vary considerably between the ESPAD countries. This also holds true also when looking at the number of students who had been drinking 3 times or more during last month. The British Isles are at the top but also two

Summary of the 2003 findings

Mediterranean countries. The highest proportion is found in Malta, where 43% of the students reported this frequency of spirits consumption. The countries that come next include the Faroe Islands, Greece, Ireland, Isle of Man and the United Kingdom (37–39%). In about half of the countries, more boys than girls report such frequent consumption of spirits. However, almost the same number of countries report prevalence rates that are equal or almost equal between the sexes. Only three countries report proportions among the girls that exceed those of the boys. These countries are all high frequency countries and they are all parts of the British Isles, i.e. Ireland, Isle of Man and the United Kingdom.

Drunkenness Lifetime prevalence of having been drunk 20 times or more and the 30 days prevalence of being drunk 3 times or more are presented in the summary tables. Some students have a rather limited experience of getting drunk, while others get intoxicated more frequently. However, in 30 of the 35 countries studied a majority of the students have been drunk at least once. The countries with the highest percentages indicating that they had been drunk 20 times or more in lifetime include Denmark, Ireland, Isle of Man, the United Kingdom, Estonia and Finland (26–36%). In other countries only a few report this frequency of drunkenness. In Turkey only 1% had been drunk 20 times or more and in Cyprus, France, Greece and Portugal this was reported by about 3% of the students. In a majority of the countries there are more boys than girls that report this frequency of intoxication. In no country are the girls in majority. However, in relatively many countries the gender distribution is rather even. These countries include both the British Isles and most of the Nordic countries (Finland, the Faroe Islands, Iceland, Ireland, Isle of Man, Norway, Sweden and the United Kingdom). The number of students who have been drunk 3 times or more during the last 30 days is of course much smaller, but the highest ranked countries are in most cases the same. Thus, in Denmark and Ireland about one fourth of the students had been drunk that often. Other countries with high prevalence rates include Isle of Man and the United Kingdom. However, in about half of the ESPAD countries the number of students reporting this frequency of intoxication is 10% or less. The lowest figures are

Summary of the 2003 findings

reported from Cyprus, France, Greece, Portugal and Turkey (1–4%).

Binge drinking The frequency of having 5 or more drinks in a row, sometimes referred to as “binge drinking”, provides an alternative measure of heavy alcohol use. The proportion indicating such consumption 3 times or more during the last 30 days vary considerably over the ESPAD countries. This is reported by one fifth to one third of the students in about half of the ESPAD countries. The highest number of students reporting this behaviour is found in Denmark, Ireland, Isle of Man, Malta, the Netherlands, Norway, Poland, Sweden and the United Kingdom (24–32%). Thus, there is a concentration of countries to the northern and western parts of Europe with Malta as the only exception. Countries with the lowest binge drinking figures are Cyprus, France, Greece, Hungary, Iceland, Romania and Turkey (5–11%).

Illicit drugs Lifetime use of various illicit drugs are presented in the summary tables, including cannabis, amphetamines, LSD, Ecstasy, tranquillisers or sedatives without a doctor’s prescription and the use of inhalants. In addition the 30 days prevalence of cannabis is included. The vast majority of students in all ESPAD countries that have tried any illicit drug have used marijuana or hashish. Thus, the number of students reporting cannabis use is almost identical with the total illicit drug prevalence. The top country in this respect is the Czech Republic where 44% of the students have used marijuana or hashish. High prevalence rates are also reported in France, Ireland, Isle of Man, Switzerland and the United Kingdom (38–40%). Other countries where more than one fourth have used cannabis include Belgium, Germany, Greenland, Italy, the Netherlands, the Slovak Republic and Slovenia (27–32%). The lowest levels are reported in Cyprus, Greece, Sweden, Romania and Turkey (3–7%), but also in the Faroe Islands, Finland and Norway (around 10%). The use of cannabis during the last 30 days may indicate regular use. In some countries about one fifth of the students report this, in others much lower prevalence rates are noted. The countries with the highest 30 days prevalence include the Czech Republic, France, Isle of Man, Switzerland and the United Kingdom (19–22%).

21

In most ESPAD countries there are more boys than girls who have used cannabis. However, the gender differences are small in Bulgaria, Croatia, Greenland, Hungary, Iceland, Ireland, Russia (Moscow) the Slovak Republic and Slovenia. The countries with the highest percentages of students reporting use of amphetamines are Estonia, Germany, Iceland, Lithuania and Poland (5– 7%). In 13 countries 1% or less reported use of amphetamines. The ESPAD students do not use LSD very frequently. The highest percentages are found in the Czech Republic and Isle of Man where 5–6% reported such use. Ecstasy is the most used drug of those included in the questionnaire apart from cannabis. In the Czech Republic 8% had used it, followed by Croatia, Estonia, Ireland, Isle of Man, the Netherlands and the United Kingdom (5–7%). Tranquillisers or sedatives can be used both as a legally prescribed medicine and as an illicit drug. The use of such substances without prescription is most common in Poland (17%) followed by Lithuania (14%), France and the Czech Republic (11– 13%). The lowest prevalence rates are found in Austria, Bulgaria, Germany, Ireland, Ukraine and the United Kingdom (2% each). The highest prevalence of inhalants is reported in Greenland, where 22% had ever used them. Other countries with high levels of inhalant use include Cyprus, Greece, Ireland, Isle of Man, Malta and Slovenia (15–19%). Very small gender differences are found in relation to the use of inhalants. In a majority of the countries there are no gender differences, but in Belgium, Cyprus, Greece, Portugal and Ukraine more boys than girls reported this behaviour. Girls

22

only reported more use than boys in one country, Ireland.

Conclusions In summary, the pattern of alcohol consumption reveals that frequent drinking is most prevalent among students in the western parts of Europe, such as the British Isles, the Netherlands, Belgium but also in Austria, the Czech Republic and Malta. Very few students in the northern parts of Europe drink that often. Beer consumption is most prevalent in Bulgaria, Denmark, the Netherlands and Poland, while wine consumption is most prevalent in typical wine producing countries such as Austria, the Czech Republic, Greece, Italy, Malta and Slovenia. The consumption of spirits is less uniform, with high prevalence rates in as disparate countries as the Faroe Islands, Greece, Ireland, Isle of Man, Malta and the United Kingdom. The prevalence of drunkenness seem to be most concentrated to countries in the western parts of Europe, such as Denmark, Ireland, Isle of Man and the United Kingdom. Very few students report frequent drunkenness in Mediterranean countries such as Cyprus, France, Greece, Portugal, Romania and Turkey. The illicit drug use is dominated by use of marijuana or hashish. Frequent use is mainly reported from countries in the central and western parts of Europe, where more than one third of the students have used it. The high prevalence countries include the Czech Republic, France, Ireland, Isle of Man, Switzerland and the United Kingdom. The low prevalence countries are found in the north as well as the south of Europe.

Summary of the 2003 findings

Summary of the 2003 findings

23

41 28 32 32 28 39 26 41 39 32 .. 40 19 34 33 19 25 15 25 39 49 17 28 23 32 19 26 42 35 26 20 24 17 38 19

Lifetime use 40 times or more

48 32 42 36 30 43 27 40 42 35 31 43 27 56 39 20 28 23 35 46 49 28 32 24 35 28 32 44 39 35 20 33 22 49 25

Smoked during the last 30 days

Cigarette smoking

a) Binge drinking: 5 drinks or more in a row. b) Without a doctor’s prescription.

Austria Belgium Bulgaria Croatia Cyprus Czech Rep. Denmark Estonia Faroe Isl. Finland France Germany Greece Greenland Hungary Iceland Ireland Isle of Man Italy Latvia Lithuania Malta Netherlands Norway Poland Portugal Romania Russia Slovak Rep. Slovenia Sweden Switzerland Turkey Ukraine United Kingdom

Country

53 46 33 38 34 54 57 38 34 20 30 43 43 17 27 16 42 45 33 30 45 41 55 17 36 20 26 44 42 32 21 33 10 24 47

Lifetime use 40 times or more 27 28 13 15 18 17 18 8 6 3 10 15 17 5 8 2 17 19 17 7 13 25 34 3 13 11 9 16 12 10 2 18 4 6 18

Any alcohol 10 times or more 50 49 55 42 40 52 54 35 38 22 26 45 39 33 23 19 47 41 45 42 38 45 55 16 50 27 47 47 32 31 26 39 14 45 42

Beer 3 times or more

Last 30 days

Alcohol consumption

20 21 16 23 15 18 6 12 7 4 11 12 27 5 20 5 6 15 29 12 12 42 7 5 9 8 20 16 21 28 9 12 4 17 15

Wine 3 times or more 36 33 27 20 39 30 34 27 42 10 21 28 41 33 23 15 29 32 30 13 15 44 36 17 18 29 10 17 27 19 19 37 7 18 33

Spirits 3 times or more 37 11 15 14 2 25 41 33 26 25 4 16 4 24 16 16 32 28 8 19 29 7 9 14 15 5 7 18 20 20 18 14 3 24 27

Lifetime 20 times or more 22 12 17 12 6 17 30 23 20 15 5 11 3 20 11 9 27 20 9 12 17 7 10 12 13 6 6 13 14 16 12 12 4 18 22

Last 30 days 3 times or more

Drunkenness

Summary table 1. Selected variables on tobacco, alcohol and drug consumption. Boys.

28 26 19 17 24 31 26 21 18 13 31 14 23 12 13 31 26 19 24 19 32 37 25 17 20 19 22 20 23 27 21 9 28 26

..

Binge drinking a) last 30 days 3 times or more

23 37 23 24 7 48 27 28 9 11 42 31 7 29 18 14 38 41 31 20 18 13 32 9 23 18 4 26 32 31 9 44 6 29 41

Lifetime

Cannabis

12 20 10 9 4 21 10 8 2 3 26 14 2 12 7 4 16 24 19 5 8 5 17 3 10 11 1 7 10 14 2 23 3 8 23

Last 30 days

4 3 2 2 1 3 5 7 1 1 3 5 0 0 3 5 1 4 3 3 6 1 2 2 6 3 1 1 2 0 1 3 3 1 2

Amphetamines

2 4 2 2 2 6 1 3 0 1 1 3 1 1 2 2 2 6 4 1 3 1 3 1 3 3 0 2 2 1 2 1 2 1 3

LSD

3 5 3 5 2 8 3 5 0 1 4 3 2 2 3 2 4 7 4 3 3 1 6 2 3 5 1 3 3 3 2 2 3 2 5

Ecstasy

Lifetime use of other illicit drugs

1 9 2 4 7 8 4 5 5 4 10 1 3 3 7 8 2 6 5 2 10 2 7 3 12 4 3 2 3 3 5 4 3 3 2

14 9 4 14 19 9 9 9 10 8 12 12 17 23 6 12 14 18 8 8 6 16 7 6 10 10 2 7 10 15 8 9 5 9 12

Lifetime Lifetime use of use of tranquillinhalants isers or b) sedatives

24

Summary of the 2003 findings

44 26 37 29 12 38 28 29 42 32 .. 39 21 49 30 17 29 28 25 25 28 16 26 29 21 17 15 38 30 28 24 24 7 19 24

Lifetime use 40 times or more

56 33 50 37 14 43 32 33 41 41 36 46 30 65 40 20 37 36 40 36 33 26 31 32 27 27 26 44 36 38 27 34 12 28 34

Smoked during the last 30 days

Cigarette smoking

a) Binge drinking: 5 drinks or more in a row. b) Without a doctor’s prescription..

Austria Belgium Bulgaria Croatia Cyprus Czech Rep. Denmark Estonia Faroe Isl. Finland France Germany Greece Greenland Hungary Iceland Ireland Isle of Man Italy Latvia Lithuania Malta Netherlands Norway Poland Portugal Romania Russia Slovak Rep. Slovenia Sweden Switzerland Turkey Ukraine United Kingdom

Country

41 27 21 16 12 40 42 26 30 20 15 31 28 9 14 12 36 44 16 23 31 27 35 14 18 8 12 34 28 18 14 20 4 19 39

Lifetime use 40 times or more 15 13 7 11 6 10 10 5 4 2 5 9 9 4 4 1 14 13 6 4 5 16 17 2 6 4 3 10 6 4 1 7 1 4 15

Any alcohol 10 times or more 18 24 33 14 16 28 35 16 21 13 16 18 18 32 9 17 25 13 22 22 20 16 23 11 32 10 22 30 14 12 14 18 5 21 17

Beer 3 times or more

Last 30 days

Alcohol consumption

23 15 9 15 8 24 10 15 4 6 4 21 15 6 16 5 13 20 16 14 16 32 12 4 7 3 9 19 17 17 8 8 3 18 22

Wine 3 times or more 23 26 25 18 23 27 31 21 34 10 16 23 32 20 21 13 46 46 21 10 12 42 30 18 11 24 5 14 22 21 17 33 1 12 43

Spirits 3 times or more 13 3 7 5 0 13 31 19 23 28 2 8 3 18 5 15 29 29 3 10 12 3 4 13 5 2 2 13 10 10 15 6 0 11 27

Lifetime 20 times or more 11 4 8 5 1 10 21 13 16 17 2 8 3 19 5 9 25 25 3 7 8 4 4 10 5 2 2 9 8 8 9 6 1 10 25

Last 30 days 3 times or more

Drunkenness

Summary table 2. Selected variables on tobacco, alcohol and drug consumption. Girls

14 16 10 6 13 18 15 17 15 7 24 8 16 5 9 33 30 8 18 7 19 20 24 5 10 5 12 12 18 22 11 2 15 29

..

Binge drinking a) last 30 days 3 times or more

18 28 19 20 2 40 18 18 10 11 35 24 5 26 13 11 39 38 23 12 9 8 24 9 13 12 2 18 22 26 6 36 2 12 35

Lifetime

Cannabis

7 13 7 7 1 17 5 4 1 2 18 9 2 11 5 4 17 19 12 2 3 3 9 2 5 5 0 6 9 14 1 17 1 2 16

Last 30 days

5 2 2 3 1 5 3 8 0 1 2 6 0 0 3 5 2 2 2 3 4 1 1 2 4 3 0 1 2 1 1 3 1 1 3

Amphetamines

2 1 2 1 0 5 1 2 2 1 1 4 1 0 2 1 2 3 2 1 1 0 2 1 1 1 0 1 2 1 1 1 1 1 1

LSD

3 4 2 4 1 8 2 5 2 2 3 4 1 2 4 3 5 6 2 3 1 1 3 1 2 3 0 2 3 4 1 2 1 0 5

Ecstasy

Lifetime use of other illicit drugs

2 10 2 9 5 14 5 13 5 9 15 2 5 4 13 10 2 4 7 4 18 3 10 3 22 7 7 3 5 8 7 7 3 1 1

14 5 3 14 15 9 7 7 13 8 10 11 13 22 4 11 21 20 5 7 4 15 5 4 8 6 1 6 7 15 8 6 3 4 13

Lifetime Lifetime use of use of tranquillinhalants isers or b) sedatives

Summary of the 2003 findings

25

42 27 35 30 20 39 27 35 41 32 .. 40 20 42 31 18 27 22 25 32 39 16 27 26 26 18 20 40 32 27 22 24 13 28 22

Lifetime use 40 times or more

49 32 46 36 22 43 30 37 41 38 33 45 28 60 39 20 33 30 38 40 41 27 31 28 31 28 29 44 37 36 23 34 18 39 29

Smoked during the last 30 days

Cigarette smoking

a) Binge drinking: 5 drinks or more in a row. b) Without a doctor’s prescription.

Austria Belgium Bulgaria Croatia Cyprus Czech Rep. Denmark Estonia Faroe Isl. Finland France Germany Greece Greenland Hungary Iceland Ireland Isle of Man Italy Latvia Lithuania Malta Netherlands Norway Poland Portugal Romania Russia Slovak Rep. Slovenia Sweden Switzerland Turkey Ukraine United Kingdom

Country

48 36 27 27 21 46 50 32 32 20 22 37 35 13 21 14 39 45 24 26 38 33 45 15 27 14 18 39 34 25 17 27 7 22 43

Lifetime use 40 times or more 21 20 9 13 11 13 13 6 4 2 7 11 13 3 6 1 16 15 12 6 8 20 25 3 10 7 5 12 9 7 1 13 4 5 17

Any alcohol 10 times or more 36 36 43 28 28 39 44 25 31 18 20 30 28 32 17 19 36 25 34 32 28 29 40 14 41 18 33 38 38 21 20 28 10 34 31

Beer 3 times or more

Last 30 days

Alcohol consumption

22 18 14 19 13 21 9 15 7 5 8 17 21 6 19 5 10 18 23 12 13 35 11 3 8 6 13 17 19 21 8 10 4 17 18

Wine 3 times or more 30 29 25 17 31 28 31 24 37 10 19 24 37 26 22 13 38 38 25 12 14 43 34 17 14 26 6 16 25 20 18 35 5 16 39

Spirits 3 times or more 21 7 10 9 2 18 36 26 24 26 3 12 3 21 11 16 30 29 5 14 21 4 6 14 10 3 3 15 14 15 17 10 1 18 27

Lifetime 20 times or more 17 8 10 8 2 13 26 17 18 16 3 10 3 19 9 10 26 23 7 8 12 5 7 12 10 3 3 11 11 12 9 9 1 16 23

Last 30 days 3 times or more

Drunkenness

22 21 15 11 18 24 20 19 15 9 28 11 19 8 11 32 27 13 22 13 25 28 24 11 16 11 17 15 22 25 15 5 22 27

..

Binge drinking a) last 30 days 3 times or more

Summary table 3. Selected variables on tobacco, alcohol and drug consumption. All students.

21 32 21 22 4 44 23 23 9 11 38 27 6 27 16 13 39 39 27 16 13 10 28 9 18 15 3 22 27 28 7 40 4 21 38

Lifetime

Cannabis

10 17 8 8 2 19 8 6 1 3 22 12 2 11 6 4 17 21 15 4 6 4 13 3 8 8 0 7 10 14 1 20 2 5 20

Last 30 days

4 2 2 2 1 4 4 7 1 1 2 5 0 0 3 5 1 3 3 3 5 1 1 2 5 3 0 1 2 1 1 3 2 1 3

Amphetamines

2 3 2 1 1 6 1 2 1 1 1 3 1 0 2 1 2 5 3 1 2 1 2 1 2 2 0 1 2 1 1 1 2 1 2

LSD

3 4 3 5 2 8 2 5 1 1 3 3 2 2 3 3 5 7 3 3 2 1 5 2 3 4 1 3 3 3 2 2 2 1 5

Ecstasy

Lifetime use of other illicit drugs

2 9 2 6 6 11 4 9 5 7 13 2 4 3 10 9 2 5 6 3 14 3 8 3 17 5 5 3 4 5 6 6 3 2 2

14 7 3 14 17 9 8 8 11 8 11 11 15 22 5 12 18 19 6 7 5 16 6 5 9 8 1 7 9 15 8 7 4 6 12

Lifetime Lifetime use of use of tranquillinhalants isers or b) sedatives

Study design and procedures

The target population The target population for the ESPAD project is students that will become 16 years old during the year of the data collection i.e. they should all be born a specific year. The 1995 study focussed on students born in 1979 and in the second data collection in 1999 they were born in 1983. The third survey in 2003 targeted students born in 1987. The main idea behind the choice of this agegroup for the study is that the students should still be available in schools, but not too young to have had any experience of alcohol or drug use. The mean age among surveyed students have been about the same in all three data collections. In 2003 the approximate mean age was 15.8 years with a range of 15.6–15.9 years (Table A in the chapter “Methodological considerations”). There are, however, differences between coun-

tries in how well the samples represent the agegroup. In some countries schooling is compulsory until the age of 15–16 years, while in others the students begin secondary school at this age. Furthermore, many students do not continue to secondary school, but leave for other training or for work. Table A shows the approximate proportion of the age cohort expected to be enrolled in school in different countries. Available information about the proportion of the actual age cohort still in school shows that there are some differences between countries in this respect. However, with a few exceptions 85% or more of the 1987 age cohort was to be found at school at the time of the data collection. The lower this proportion, the less representative are the results for the 1987 birth cohort.

The data collection instrument The work of the Pompidou School Survey Subgroup in the 1980’s resulted in a battery of questions to be used by researchers in different countries that were interested in performing school surveys. The content was very much influenced by the questionnaire already developed and used within the Monitoring the Future project in Michigan. Dr Lloyd Johnston, who was the chair of the School Survey Subgroup, is also head of the group of researchers engaged in the Monitoring the Future project. The first ESPAD questionnaire was developed from the battery of questions that was tested by the Pompidou School Survey Subgroup. However, every question was discussed and agreed upon by the large group of collaborating investigators. A very large part of the first questionnaire was kept also in the 1999 and 2003 surveys. The main part of the questionnaire constitutes of core questions to be used in all countries. In addition a number of module and optional questions

Study design and procedures

were included to be used at the choice of each country. The questionnaire is presented in Appendix III. It was also decided that each country might add questions of special interest provided that those questions were not of a nature that would affect the students’ willingness to respond, or that their number would overload the questionnaire. It was decided that each country should translate the questionnaire into its own language and thereby adjust the wordings to make the questions as appropriate as possible in the cultural context. Drug streetnames etc. should be adjusted to what was common in the country. Once the translation was ready, it should be back translated into English again. By doing this, discrepancies from the original might be discovered and corrected. It was also recommended that each country should test the questionnaire in a small pilot study in order to discover any faults or difficulties while answering it. A test would also indicate how long time the students needed to complete the question-

27

naire. In the 2003 survey a little more than half of the countries did a pilot study (Table A). However, some of the countries that did not do so this time had tested the questionnaire in relation to earlier surveys. Table A shows the number of core, optional and own questions included in different countries’ questionnaires. For each question every single subquestion is counted as one variable. All countries but one asked all, or nearly all, core questions. The main exception is France that only used 174 of the 309 core questions (56%). However, only a few own questions were put within the core questions. Hence, the context of the French core questions have most probably not affected the possibilities to compare with data from other countries.

The Swiss questionnaire includes a battery of questions in the midst of the ESPAD questions because they belonged thematically to this section. However, before doing so two versions of the questionnaire were piloted and no effects on the response pattern were identified. Despite all efforts to standardise the data collection instrument, some discrepancies were inevitable. However, it may not be too optimistic to think that the discrepancies in the questionnaires only have had a very limited negative effect on the comparability of the findings from different countries. In the few cases when discrepancies are important enough to make a question less comparable, this will be commented in the result chapters.

Sampling procedure The sample size and sampling procedures have been discussed at some ESPAD project meetings. It soon became clear that the ESPAD countries were very different in terms of what kind of school statistics are available. Some countries had detailed information about the number of schools, classes and students, while in others only e.g. the total number of schools, but not the size of them, was known. The sample should consist of randomly selected classes. As mentioned in an earlier part of this report, regional seminars were organised aimed at discussing the project plans in detail, including problems and opportunities for the sampling procedure in each country. It was recommended that each country, with some minor exceptions, should draw a sample of about 2,800 students as a minimum, regardless of the size of the country (Bjarnason and Morgan,

2002). This was calculated to give about 2400 answered questionnaires, which would allow for breakdowns by sex plus another variable. However, in a few countries a lesser number of students participated, simply because the study population was smaller. The target population of students born in 1987 was very differently distributed over schooltypes (academic, vocational etc.) and grades in different countries. At the regional seminars solutions to the sampling problems were discussed and suggested. In some countries the vast majority of the agegroup was found in one grade only. In others there were two or more grades where this agegroup was taught. Whenever possible it was recommended to include all grades with students born in 1987. However, in some countries the grade with the highest proportion of students born in 1987 was the only chosen.

Field procedure In line with what was decided about the sampling and the data collection instrument, also the field procedures should be standardised as much as possible (Hibell and Andersson, 2002a). Due to cultural differences there are of course many factors, which make it difficult to follow exactly the same schedule in every country.

28

The recommended data collection period was March–April 2003. Most countries adhered to these dates, but the length of the period varied quite a lot, from one day only to about 2–3 months in some countries. For practical reasons the time of the data collection was different from the planned period in a few countries, including Malta (January), the

Study design and procedures

Netherlands (October–November), Poland (May– June), Portugal (May), Romania (June), Switzerland (May–June) and Turkey (May). The data collection in a country was planned to take place during a certain week, which should not be proceeded by any holiday, ensuring that the students referred to a "normal" week when answering the questions, i.e. no extraordinary alcohol or drug consumption due to any celebration should be reflected in the answers. Schools unable to perform the survey during the assigned week were allowed to do so in the preceding week instead. The headmaster of the participating schools were contacted and informed of the planned study. He or she was asked to inform the teacher(s) of the chosen class(es), but not to inform the students in order to avoid discussions among them, which could lead to biased data. The class teacher was asked to schedule the survey for one lecture following the same procedure as for a written test. Data were collected by group administered questionnaires, under the supervision of a teacher or a research assistant. At some ESPAD project meetings much discussion have been directed towards this issue. It was thought that in many countries teachers would not be trusted by the students and therefore cause biased data. The solution to this problem was that in countries where it was judged to be possible to use teachers this ought to be done, while in others research assistants were used. It was

Study design and procedures

considered crucial not whether a teacher or a research assistant was present, but whether they were trusted by the students or not. In a methodological study by Bjarnason (1995) no significant differences were found between teachers’ or research assistants’ modes of questionnaire administration. These findings suggest that, at least in some countries, the effect of administration mode is negligible. It was recommended that each student should get an (unmarked) envelope to put his or her completed questionnaire in, before it was sealed by him- or herself. When the data collection was over the teacher/research assistant had to collect the sealed envelopes and send them back to the research institute. The information to the survey leader included a written instruction, which described how to perform the data collection. The anonymous character of the study was stressed and the survey leader should refrain from walking around in the classroom while the forms were completed. A standardised classroom report was used. On this form the survey leader gave information about the average time needed to complete the questionnaires, the number of absent and present students, the reasons for absence and other important information about the situation in the classroom. The classroom report also contained information about whether the students were interested in the study and worked seriously.

29

Methodological considerations

Introduction All surveys encounter methodological problems which have to be considered when analysing the results. The 2003 ESPAD project is based on 35 national surveys united by a single project plan. The methodological issues that have been identified and resolved could fill several thick volumes such as this report. This chapter provides a brief overview of the issues of representativeness, reliability and validity in the ESPAD project. The chapter ends with a short summary of the most important conclusions. In the first ESPAD survey in 1995 it was apparent that several of the participating countries were also conducting a school survey on alcohol and drug use for the first time. In this third ESPAD study, increased experience and a long co-operation have contributed to a more robust and standardises methodology. There are still some discrepancies and areas of concern that need to be addressed, but it should be stressed that overall the ESPAD project has accomplished a high degree of representativeness, reliability and validity. In 1988 the Pompidou group of the Council of Europe initiated a pilot study of adolescent substance use. One of the main goals of the pilot study was to test the methodology, which resulted in a rather detailed discussion about the methodological results (Johnston et al. 1994). The discussion was a critical part of the report and has been very useful for the ESPAD project. The experiences of the pilot study were positive and implied that valid international research on substance use among students is feasible. The ESPAD project relies on experiences from more than 30 years of school surveys in Sweden, the Pompidou pilot project as well as knowledge gained by individual researchers from all over Europe in earlier ESPAD data collections. Many of the questions in the ESPAD questionnaire originate from the Pompidou pilot study that, in turn, to a large extent was based on the questionnaire used in the Monitoring the Future Project in the USA. The standardisation of survey methodology is one of the most important issues in the ESPAD

Methodological considerations

project. However, it should be stressed that standardisation alone does not ensure that data are directly comparable between countries. It is not possible to control for everything and some influences are not even possible to measure. The cultural contexts in which the students have given their answers varies and formally identical measures may have very different meanings in different contexts. In addition, one can never be certain of whether results from one country are more or less valid than those from another. This is one reason why the long-term goal, and one of the most characteristic features of the ESPAD project, is to compare trends in participating countries. In the figures two dots (..) symbolise that data does not exist or is not available. A zero (0) means that the information is related to at least one person but to less than 0.5%. A short line (–) signifies that no one has given that answer. To better ascertain the role of cultural context in different countries, and how it may impact on validity, a methodological study was conducted as one of the preparative measures prior to the ESPAD 99 data collection (Hibell et al. 2000). The methodology study was conducted in 1998 and included aspects of reliability as well as validity. Data were collected in countries from different parts of Europe. Two countries hailed from northern/western parts of Europe (Denmark and Sweden), two from the Mediterranean (Cyprus and Malta) while three were situated in the central and eastern parts of Europe (Lithuania, the Slovak Republic and Ukraine). The study indicated that the reliability as well as the validity was high in all seven countries. With a few modifications, the survey leader questionnaire (the classroom report) of the methodology study was used in the 1999 and 2003 data collections.

Changes over time One of the important long-term goals of the ESPAD project is to track changes in adolescent substance use over time. While cultural context may affect the meaning of responses to formally stand -

31

ESPAD Annual Swedish school survey

% 10 8 6 4 2 0 95

96

97

98

99

00

01

02

03 Year

Figur A. Lifetime prevalence of any illicit drug among girls in the ESPAD studies and in the annual Swedish school surveys. Source: Hvitfeldt et al. (2004)

ardised measures, changes in such responses over time may be relatively less affected by context. In other words, even if the percentages using a particular drug were not directly comparable between two countries, the increase or decrease in those two countries could still be compared. It should be noted that the ESPAD survey is repeated every four years. In the next chapter changes between 1995 and 1999 as well as between 1999 and 2003 are shown country by country in simple graphs in which a straight line is drawn between the dots of each of the three data collections. However, four years is a relatively long period during which many changes might have occurred. In other words, the straight lines may mask considerable annual fluctuation. An example of this can be seen in figure A. Data from the annual Swedish school surveys show that there was an increasing trend from 1998 to 2001 in the proportion of girls that tried any illicit drug. After that there is a downward trend. However, the figures from the three ESPAD data collections are indicative of a weak increasing trend.

32

A note on statistical significance As will be discussed in detail below, the sampling procedures in the ESPAD survey differ considerably between countries. This affects the precision of the estimates in each country but should in principle not bias the point estimate itself (Bjarna son and Morgan 2002). The calculation of standard errors is therefore rather complicated in many countries and the necessary software and resources to calculate them were in many cases unavailable. As a result, confidence intervals are not calculated for this report. This issue is an ongoing concern in the ESPAD project and will hopefully be resolved in future reports In the current report figures are compared between countries and over time in terms of substantive rather than statistical significance. In general it can be assumed that differences that are large enough to have policy implications far exceed the limit of statistically significance differences. However, considerably caution should be exercised in comparing small differences in percentages. Leena Metso (2000) has examined these issues in some detail using the Finnish ESPAD data collected in 1995 and 1999. As she points out, cluster sampling does not affect the estimates of percentages. However, she found a moderate level of intracluster correlation in the Finnish data. This implies that standard errors calculated for these data under the assumptions of simple random sampling would be too small and the precision of the results is therefore less than standard significance tests would suggest. This further underscores the importance of resolving the problems surrounding the calculation of standard errors in the future. It is important to note that a certain difference in a particular variable between 1999 and 2003 maybe significant in one country but note so in another. Differences have to be tested separately from each country’s results to make it possible to decide whether a difference is significant or not. However, to be able to do so it is necessary to have access to the whole data set and to use a statistical programme that accounts for cluster effects.

Methodological considerations

Representativeness The target population of the ESPAD study is defined as the national population of students whose sixteenth birthday is in the calendar year of the survey (Bjarnason and Morgan 2002). In 2003 the goal of a national survey was reached in 32 of the 35 countries. In Russia the ESPAD survey targeted only students living in Moscow, the capital of the Russian Federation with about 8.5 million inhabitants. In Germany the data collection was limited to the six out of 16 federal states (Bundesländer) that agreed to participate. They were Bavaria, Brandenburg, Berlin, Hesse, Mecklenburg-Western Pomerania and Thuringia. The population in these Bundesländer are about 28.6 million out of 82.5 million in the whole of Germany. Finally, in Turkey data were collected in one major city in each of six different regions in the country. Participating cities were Adana, Ankara, Diyarbakir, Istanbul, Izmir and Samsun. While the results in these countries may to some degree reflect the situation in the country as a whole, they can only be representative of the population from which they are drawn.

Average age and time of the data collection With the exception of the Netherlands, data were collected during the first half of 2003, with a majority conducted between the period March to May (Table A). The Dutch ESPAD researchers did not find it possible to collect data during springtime since this would most probably have resulted in substantially more refusals from schools and classes. Instead the questionnaires were administrated in October and November. Based on the time of data collection, an approximate average age of the students has been estimated for each country (Table A). In all but one of the 35 ESPAD countries the average age varies between 15.7 and 15.9 years, which is the same range in average age as in 19991. The only minor exception is Malta with the average age of 15.6 years. In the Netherlands the target population was redefined to be students born from August 1987 through July 1998, which gives an average age of 15.7 years. (A further discussion of this redefinition can be found in Appendix 1). In 1999 data in Greece were collected in October which gave an average age of about 16.3 years, 1

while the corresponding figure in 2003 is 15.8 years. This age difference of seven months must be kept in mind when interpreting changes in the substance use figures between 1999 and 2003.

Representativeness of the samples Sampling in the ESPAD project is based on classes as the final sampling unit (Bjarnason and Morgan 2002). This procedure is vastly more economical than sampling individual students and also has some desirable methodological properties. In particular, sampling entire classes can be expected to increase student perceptions of anonymity. Sampling individual students and asking them to fill out a questionnaire individually could affect the truthfulness of their answers and therefore bias the results of this study. If students born in 1987 were in two or more grades it was recommended that it was advisable to sample classes from all those grades and then screen the target population by using a question on the year of birth. If it was not possible to sample more than one grade, the grade chosen should include the majority of students born in 1987. In countries where sampling was not so straightforward it was recommended that one seek co-operation of an experienced sociologist or statistician. An overview of the sampling procedure in each country is provided in Table A. Further information can be found in chapter 2 and Appendix 1. The number of students born in 1987 in Faroe Islands, Greenland, Iceland, Isle of Man and Malta was similar to the number of students to be sampled according to the ESPAD guidelines (Bjarnason and Morgan 2002). In these countries all students were therefore targeted for sampling. In all other countries but one, classes were the sampling units. The only exception was Denmark where a small part of the sample was composed of schools (see Appendix 1). In some countries classes were the only sampling units, i.e. they were drawn from comprehensive lists of classes. In other countries school classes were the last units in a multistage stratified sampling process. In these countries schools were sampled before the final sampling of classes was done. In many countries sampled schools were asked to provide lists of classes before the final sample of classes could be effectively drawn.

The calculated averages ages in the ESPAD 99 report were systematically 0.5 years too low.

Methodological considerations

33

Table A. Characteristics of the ESPAD surveys in participating countries. Continues... Country

Born in 1983 still in school (approx. %)

Sampling unit(s)

Sample type

Grade level(s) included

Approx. mean age a)

Representativeness b)

Austria

90

class

stratified random

grades 9–10

15.8

national (86%)

Belgium

99

school, class

systematic random

grades 8–10c)

15.8

national (95%)

Bulgaria

72

school

stratified random

grades 9–10

15.9

national (100%)

Croatia

class

stratified random

grades 1–2

15.8

national (97%)

Cyprus

95 ..

school

stratified random

grades 1–2

15.8

national (74%)

Czech Republic

95

school

stratified random

grade 1

15.7

national (~68%)

98

school, class

stratified random

grade 9

15.8

national (85%)

~80

school, class

systematic random

grades 8–10

15.7

national (~80%) national (92%)

Denmark Estonia Faroe Islands

95

no sampling

total

grade 9

15.7

school, class

systematic random

grade 9

15.7

national (93%)

98

school

stratified random

grades 8–11

15.8

national (93%)

92

class

systematic random

grades 9–10

15.7

6 Bundesl. (84%)

~100

class

stratified random

gymn 3rd, lycee A, B, C

15.8

national (93%)

Finland

~100

France Germany Greece Greenland

88

no sampling

total

grades 9–11

15.7

national (~100%)

Hungary

91

class

stratified random

grades 8–10

15.7

national (91%) national (99%)

Iceland

99

no sampling

total

grade 10

15.7

Ireland

93

school, class

stratified random

grade 5

15.8

national (67%)

no sampling

total

grades 10–11

15.8

national (100%)

Isle of Man

≥ 80

Italy

~93

school

stratified random

grades 1–4

15.8

national (100%)

Latvia

87

classes

stratified random

grades 8–10, grade 1 vocational

15.8

national (89%)

Lithuania

96

school, class

systematic random

grades 8–10

15.7

national (97%)

Malta

95

no sampling

total

grade 5

15.6

national (75%)

Netherlands

~92

school, class

stratified random

grades 3–4 secondary school

15.7

national (92%)

Norway

100

classes

stratified random

grade 10

15.7

national (~100%)

Poland

95

class

systematic random

gymn. grade 3

15.9

national (92%)

Portugal

81

class

stratified random

grades 7–10

15.9

national (99%)

Romania Russia (Moscow) Slovak Republic

93

school, class

stratified random

grades 9–10

15.9

national (79%)

~95

school, class

systematic random l)

9–10th secondary, 1st techn., profess., nurses

15.7

Moscow (98%)

school

stratified random

grades 1–4

15.7

national (~67%)

Slovenia

90–95

98

class

systematic random

grade 1

15.8

national (84%)

Sweden

95

class

systematic random

grade 9

15.7

national (95%) national (85%)

Switzerland

98

class

strat syst random

grades 8–10

15.9

Turkey

60

school

stratified random

grades 9–10

15.9

six cities (90%)

Ukraine

90

school, class

stratified random

9–10th secondary, 1st vocat., techn., colleges

15.9

national (97%)

>90

school, class

proportionate random grades 4–6

15.8

national (100%)

United Kingdom

a) A calculated figure based on the time of the data collection. In the 1999 report the calculated mean averages were systematically 0.5 years too low. b) Representativiness in relation to the target population, i.e. students (not persons) born in 1987. The figure within brackets show the approximate population b) of students born in 1987 that attended participating grades. c) Grade 8 was included only in the French speaking part. d) Teachers in French and research assistants in Dutch speaking areas. e) Individual envelopes were used in the French speaking parts. In the Flemish speaking parts where research assistants collected data the questionnaires b) were put in a class envelope. f) Flemish and French speaking respectively. g) Staff members from Department of Occupational and Public Health. h) The students put their questionnaire in a locked letter box. i) Class envelopes were used. j) Two questionnaires were used. Form A contained 27 own questions and form B 43. k) Staff members from Regional Health Services, research assistants and researchers. l) 40 out of 208 classes were sampled via a two step random sample. m) Only a small questionnaire test among data collection leaders.

34

Methodological considerations

Table A. Continued. Country

Data collection leader

Data collection period

Individ- Pilot ual en- study velopes

Number of questions (variables) Core

Module

Optional

Austria

teacher

March 31–April 4

no

yes

294

36

Belgium

teacher, research assistant d)

March–May

yes e)

yes

309

57/0 f)



Data weighted

Own 13 35/120f)

no no

Bulgaria

research assistant

May 15–26

yes

no

300

147





no

Croatia

school councellor

April 1–15

yes

no

308

62





no

Cyprus

research assistant

March–April

no

yes

308

36





no

Czech Republic

research assistant

April 3–16

yes

no

309

25



36

no

Denmark

teacher

March 6–May 2

yes

no

307

24



8

no

Estonia

research assistant

March

yes

yes

309

54



2

no

Faroe Islands

staff from.. g)

March 10–21

no h)

yes

309

82

9

149

no

Finland

teacher

March–April

yes

no

306

16

3

6

no

France

doctor, nurse

March 17–May 18

no

yes

174

14



122

no

Germany

teacher

March–April

no i)

no

308

17



8

yes

Greece

research assistant

March 1– April 30

no

yes

308

36



77

no

Greenland

teacher

March

yes

no

306

24



8

no

Hungary

research assistant

March 5–20

no

yes

308

5





yes

27/43j)

Iceland

teacher, research assistant

March 8–28

yes

yes

309

67

7

Ireland

teacher

April

yes

no

309

16





no

Isle of Man

teacher

March 31–May 3

yes

no

309

71



26

no

Italy

teacher

March/April

yes

no

309

147

10



no

Latvia

research assistant

March–May

yes

no

309

57



38

yes

Lithuania

teacher

March–April

yes

no

309

41





no

Malta

teacher

January 22

no

no

303

74





no

Netherlands

research assistant k) October–November no i)

yes

309





4

yes

Norway

teacher

no

309

12



6

yes

March–April

yes

no

Poland

research assistant

May–June

yes

yes

309

22



32

yes

Portugal

teacher

May 28

yes

yes

294





117

no

Romania

research assistant

June 3–12

yes

yes

309

66



2

yes

Russia (Moscow)

research assistant

March–April

yes

no m)

309

36





no

Slovak Republic

health staff

March 24–28

yes

yes

307

62



23

no

Slovenia

health staff

April 7–18

yes

yes

308

62



14

no

Sweden

teacher

March 17–21

yes

yes

309

38

10

3

no

Switzerland

teacher

May–June

yes

yes

309

59



96

no

Turkey

research assistant

May

yes

yes

308

36





no

Ukraine

research assistant

May 10–24

yes

yes

309

71

10



yes

United Kingdom

school staff

March–May

yes

yes

301

71



26

no

a) A calculated figure based on the time of the data collection. In the 1999 report the calculated mean averages were systematically 0.5 years too low. b) Representativiness in relation to the target population, i.e. students (not persons) born in 1987. The figure within brackets show the approximate population b) of students born in 1987 that attended participating grades. c) Grade 8 was included only in the French speaking part. d) Teachers in French and research assistants in Dutch speaking areas. e) Individual envelopes were used in the French speaking parts. In the Flemish speaking parts where research assistants collected data the questionnaires b) were put in a class envelope. f) Flemish and French speaking respectively. g) Staff members from Department of Occupational and Public Health. h) The students put their questionnaire in a locked letter box. i) Class envelopes were used. j) Two questionnaires were used. Form A contained 27 own questions and form B 43. k) Staff members from Regional Health Services, research assistants and researchers. l) 40 out of 208 classes were sampled via a two step random sample. m) Only a small questionnaire test among data collection leaders.

Methodological considerations

35

Some countries have not considered what might be called “the problem of small and large classes”. In some countries all schools/classes have had the same probability to be sampled, independent of the size of the class and the school. In practice this means that students in small classes and schools are overrepresented in the samples. If students in these classes and schools have different alcohol and/or drug habits compared to students in large classes or schools, data are not entirely representative of the population. However, in many countries where this might be the case a stratified sample has been used and it seems reasonable to assume that the sizes of schools and classes are rather similar within strata. Furthermore, class size is rather standardised in many countries. As a whole the “problem of small and large classes” is not considered a major problem in the context of the entire ESPAD project.

Representativeness of participating grades The target population of the ESPAD project is students who’s 16th birthday falls during the year of data collection. For the 2003 study that they should be born in 1987. If possible, data were to be collected in March or April, which occurred in a large majority of the countries (Table A). The definition of the ESPAD target population excludes individuals who are no longer in school. Thus, it should be kept in mind that the student populations are not coextensive with the birth cohorts, and those who have left school are more likely to have used different substances and are likely to use them at higher rates than students. However, in about three fourths of the countries with available information 90% or more of the birth cohort was enrolled in school (Table A). Important exceptions include Turkey, where only 60% of the cohort was enrolled in school, and Bulgaria, where 72% of the cohort was enrolled. In some countries nearly all students born in 1987 were assigned to one grade only, while in other countries it was in two or more grades. When this was the case, it was recommended, if necessary resources were available, to include as many grades as possible that catered for students born in 1987. If only one of these grades could be included it should be the grade with the largest proportion of students born in 1987. In countries where not all grades with students in the target age group were included in the data collection the sample is only representative of the students found in the grades targeted. In more than half of the countries 90% or more

36

of the students born in 1987 were in the grades studied (Table A). In addition, the proportion was also rather high (85–89%) in some other countries. However, in some few countries the corresponding figure was considerably lower, including the Czech Republic, Ireland and the Slovak Republic (about 67% each), Cyprus (74%), Malta (75%) and Romania (79%). Due to changes in the Slovak school system the proportion of the 15–16 year old cohort diminished from 99% in 1999 to 67% in 2003. It is of course not possible to know how the results in countries with the smallest proportion of the 1987 cohort would have been affected if all relevant grades/school types had been included. This uncertainty should be kept in mind when reading the results and comparing countries. In nearly all countries students born in other years than 1987 have usually also answered the questionnaire. However, the results in this report only reflect the answers of students born in 1987. It should be noted that the results from the USA are based on students in tenth grade, not students born in 1987. However, a large majority of the tenth graders in the USA were born in 1987, which yields some modest degree of non-comparability with the ESPAD countries. In addition, data from the Spanish school survey are included in some tables and are based only on students born in 1987.

School co-operation The number of non-participating schools and classes are shown in Table B. As already mentioned, classes were the (final) sampling units in all countries but one. However, in most countries a multistage sample was drawn, which means that schools usually were sampled in the step before classes. Denmark had two samples. One was a sample of classes in public schools and the other a small sample of private and boarding schools. In the second sample schools were the final sampling unit since most private and boarding schools were rather small and did not have a class system. Consequently, all students born in 1987 in schools in the second sample were supposed to participate in the study. With some exceptions the number of refusing schools and refusing classes was low or very low. The highest proportion were found in Belgium (54%), Denmark (47%), the United Kingdom (45%) and the Netherlands (28%). The number of non-participating classes was usually low. However, it was above 20% in four countries, including Denmark (35%), Austria (24%), Norway (23%) and Estonia (20%).

Methodological considerations

Table B. Not participating schools and classes, eliminated questionnaires and average time to complete the questionnaire. Country

Austria

Non-participating

Eliminated questionnaires (%)a)

Schools

Classes

..

79/331 b)

0.9

c)

1.5

Average time to complete the questionnaire (minutes)

41 40/50d)

Belgium

153/284

52/442

Bulgaria

1/278

1/278

0.8

Croatia

1/113

45

1/43

2/238 ..

0.6

Cyprus

5.0

57

Czech Republic

0/180

0.7

47

0.3

37

0.1

35

0/180 e)

Denmark

35/74

Estonia

10/119

74/214

e)

66/324

51

Faroe Islands

1/19

1/38



55

Finland

7/200 f)

7/200 f)

0.6

31

France

50/450 ..

127/900

1.8

45

49/557 g)

0.7

40

Greenland

5/221h) ..

13/448 ..

2.3 ..

69

Hungary

6/407

8/432 i)

0.1

48

Iceland

3/132

4/250

0.8

55

Ireland

12/120

20/216

0.7

37

Isle of Man

0/7

..

3.6

60

Italy

12/336 j)

1.5

40

Latvia

12/336 ..

14/436

1.2

49

Lithuania

1/277

1/316

0.0

44

Malta

4/65

3/245

0.4

50

Netherlands

5/194

0.5

31

Norway

76/268 ..

60/265

0.3

36

Poland

6/390

6/390

0.9

37

Portugal

25/554

16/658

2.3

50

Romania

1/208

0/414

0.5

60

Russia (Moscow)

16/208

16/210

0.5

33

Slovak Republic

1/109

3/118 k)

0.4

47

Slovenia

0/150

0/150

1.2

40

Sweden

27/200 ..

27/200

1.4

35

Switzerland

65/473

0.6

42

Turkey

0/88

0/167

0.3

30

Ukraine

6/243

0.1

United Kingdom

64/141

6/243 ..

60 ..

Germany Greece

0.8

52

a) Proportion of all answered questionnaires judged not to be seriosly answered when the questionnaires were scrutinised. b) 28 classes were replaced. c) In addition to this 17 classes were replaced. d) Flemish and French speaking respectively. e) Two samples were drawn in Denmark. One sample of 74 private and boarding schools and another of 214 classes i public schools. f) The seven classes in the seven schools were replaced by substitutive schools/classes. g) 15 classes were replaced. h) 5 schools were replaced. i) 16 classes were replaced. j) 13 schools/classes were replaced. k) 3 classes were replaced.

Methodological considerations

37

Information about non-participating schools and classes is not available from Greenland, which was cause for some concern since Greenland was one of the countries with highest school dropout rate in the ESPAD 99 data collection (24%). In some countries, including Austria, Belgium, Finland, Germany, Greece, Hungary, Italy, Portugal and Slovak Republic non-participating schools or classes were replaced by other randomly selected schools/classes. The same was also done in the Monitoring the Future Survey in the United States. This procedure assumes that the replaced schools and classes are equivalent to those refusing. However, some of the schools/classes might have refused due to supposed “bad drug habits” among the students. In nearly all countries school co-operation is reported to have been very good. In countries with few non-participating schools or classes the main reasons for not doing so were usually different kinds of schoolwork, examinations or other reasons that can be considered random occurrences. Hence for countries with few schools or classes that did not take part in the data collection there is reason to assume that non-participating schools and classes have not influenced the representativeness of the samples drawn. Altogether seven countries reported a loss of schools and/or classes that represented at least 20% of the original sample. A recurring reason provided in these countries has been that schools are asked to take part in so many school surveys that they simply don’t have the time to participate in all of them. Austria used a particular technique that involved random replacement of refusing or non-responding schools. Despite this, a relatively large number of classes (24%) did not participate in the end. There is no information available on the drop-outs and whether the loss was systematic or not. However, the assumption adopted was that the non-participating classes were randomly distributed. About 20% of the sampled classes in Estonia did not take part in the data collection exercise. However, in most of these classes no or only a few students born in 1987 were to be found. The proportion of missing students is much lower than the 20% indicate. Hence, there is reason to assume that the rather high proportion of non-participating Estonian schools and classes has not caused any important problems about the representativity. The proportion of classes that did not participate in the Norwegian study increased from 14% in

38

1999 to 23% in 2003. A major reason was the impossibility of schools to accede to every request to participate in school surveys. The non-participating classes were spread all over the country and there were no indications that students in these classes have different alcohol and drug habits. However, since this conclusion is not based on any a systematic follow up, the high proportion of nonparticipating classes remains an uncertainty. About 28% of all sampled schools in the Netherlands did not participate. Participating and nonparticipating classes were compared for school size and proportion of immigrant students. No significant differences were found. Compared to similar school surveys in the Netherlands the response rate was high. Even if there are reasons to assume that the non-participating schools did not bias the results to any degree that the comparability with other ESPAD countries was jeopardised, the rather high proportion of schools that did not participate should be noted. In the United Kingdom 45% of the sampled schools did not participate in the data collection. The most common reason given for school refusals was that the school had taken part in other research projects. There were no discernible differences in the types of schools co-operating and not co-operating. Hence, there is reason to believe that the high proportion of non-participating schools has not biased the sample to any degree and hence it should be representative. However, the fact that relatively many schools did not want to participate should be borne in mind. In Denmark two samples were drawn. One consisted of private and boarding schools in which 47% of the schools did not participate. In the other, and larger, sample of classes in public schools 35% of the schools did not take part in the survey. Non-participating schools were contacted and the most common explanation was that the schools did not have the time and that they had received too many inquiries to participate in lifestyle surveys. A comparison between participating and non-participating schools did not show any systematic differ ences. Taken together this would suggest that the relatively large number of non-participating schools and classes may not have caused major problems as far as representativeness is concerned. However, some uncertainty still remains. The large proportion of school refusals in Belgium (54%) was in line with what was expected from earlier experiences. The major reason for nonparticipation was that Belgian schools were asked

Methodological considerations

to take part in so many school surveys that many of them simply did not have the time to accede to all requests. A comparison between participating and refusing Flemish speaking schools did not reveal any notable differences. If this was also so for French speaking schools then the problem of the large number of non-participating schools is not sufficient to jeopardise the possibilities for comparisons with data from other ESPAD countries. In summary, the rather high drop-out rate of schools and classes in some countries raises questions about representativeness. The refusals nevertheless do not appear to be linked to any particular characteristics of the students but rather the attitudes and working conditions of the school staff. It should be noted that the problem is mainly related to countries from the western parts of Europe where the use of school surveys is most widespread.

Participating students In order to obtain satisfactory precision of estimates for various subgroups of the population the ESPAD guidelines recommend a net sample of 2,400 participating students in each country (Bjar nason and Morgan 2002). Assuming that 10% of students would be absent and that some selected classes would be unable to participate, a sample size of 2,800 students was recommended. However, for countries where the target cohort was less than about 30,000, it could be advisable to reduce the sample size by a factor of (1-sf), where the sampling fraction (sf) equals sample size divided by cohort size. In small countries with fewer than 2,800 students in each cohort, the total population was targeted. This was the case in the three countries with the smallest sample sizes; Greenland (555), the Faroe Islands (640) and Isle of Man (721) (Table C). In other ESPAD countries the figure varies from 1,906 (Greece), 1,925 (Russia/Moscow), 2,068 (United Kingdom) and 2,095 (the Netherlands) to 5,964 (Poland). (In USA 16,244 students took part in the study.) Thus, the number of participating students is satisfactory for international comparisons between countries. In this report the results for all students are not weighted by gender. In other words, in countries where the proportion of boys in girls is not equal, the results are slightly skewed toward the patterns among the majority gender. However, in a large majority of the countries the distribution by sex was close to even. In three countries the difference between the sexes was more than 10 percentage

Methodological considerations

points (i.e. 45–55%). In Austria 56% of the sample were boys, in Malta 44% and in Romania 42%. The uneven gender distribution in Austria, with 56% boys in the data set is due to an uneven sex distribution in grade 10. The proportion of participating boys in Romania (42%) is most probably too low compared to the proportion of boys in the target population. For certain purposes it may be advisable to calculate a weighted proportion for these countries by taking the average of the numbers for boys and for girls. The target population of Malta consisted of 47% boys, which is close to the 44% among those who participated. Thus, in practice Malta is within “the margin” of ±5%.

Response rates The response rates in each country are shown in Table C. With the exception of Greenland the response rates are calculated as the proportion of students who completed the questionnaire out of all students in participating classes. Thus, the difference consists of students in participating classes who were ill or absent for other reasons on the day of the survey. Students in non-participating schools or classes are not included among the non-respondents. They are shown separately in Table B and discussed in the section above about school co-operation. The response rates in participating classes are good or very good in nearly all countries. In 24 of the 35 countries 85% or more of the students in participating classes answered the questionnaire. The only country with a response rate below 80% is Greenland with 68%. However, this is not calculated in the same way as the response rate in the other countries. Due to a lack of information the response rate for Greenland is calculated as the proportion of participating students out of all individuals born in 1987 in the country. In other words, the figure includes young people in the birth cohort that were not enrolled in school as well as students in possible schools and classes that did not take part in the survey. Hence, the response rate in Greenland would have been substantially higher if it had been possible to calculate in the same way as in other countries. In all countries that provided information on non-participation, the main reason to emerge was that students were ill or absent for other apparently random reasons. No country reported any major methodological problems in connection with absent students. Student refusal to participate was

39

Table C. Participating students and response rates. Numbers and percentages among boys and girls. Country

Austria

Number of participating students

Response rates (%) a)

Boys

Girls

Total

Boys

Girls

1,340

1,062

2,402

..

..

90

..

81b, c)

Total

Belgium

1,112

1,208

2,320

..

Bulgaria

1,291

1,449

2,740

84

86

85

Croatia

1,446

1,438

2,884

Cyprus

999

1,153

2,152

88 ..

88 ..

88

Czech Republic

1,472

1,723

3,195

96

94

95

Denmark

1,504

1,474

2,978

90

88

89

Estonia

1,246

1,217

2,463

87

86

86

Faroe Islands

88

322

318

640

85

87

86

1,739

1,804

3,543

1,087

1,112

2,199

91 ..

91

France

92 ..

Germany

2,402

2,685

5,110

..

..

..

.. 69d) ..

Finland

Greece

886

1,020

1,906

Greenland

281

274

555

91 89b) 83

Hungary

1,398

1,279

2,677

68d) ..

Iceland

1,728

1,604

3,348

82

80

81

Ireland

1,219

1,188

2,407

381

721

97 ..

96

340

96 ..

85b)

2,300

2,571

4,871

99

98

98

85b)

84b)

85

88

Isle of Man Italy

68d) 82

Latvia

1,372

1,469

2,841

83b)

Lithuania

2,517

2,519

5,036

90

Malta

1,557

1,943

3,500

79

88

83

Netherlands

1,061

1,034

2,095

1,945

1,888

3,833

93b) ..

93b)

Norway

93b) ..

Poland

2,930

3,025

5,964

84

85

85

87d)

Portugal

1,389

1,557

2,946

97

96

96

Romania

1,823

2,548

4,371

82

84

84

1,925

78b)

82b)

80b)

Russia (Moscow)

880

1,045

Slovak Republic

1,056

1,220

2,276

86

89

87

Slovenia

1,406

1,379

2,785

88

88

88

Sweden

1,592

1,640

3,232

1,278

1,335

2,613

87 ..

87 ..

87

Switzerland Turkey

2,273

1,904

4,177

91

91

91

Ukraine

1,918

2,255

4,173

1,083

985

2,068

86 ..

83

United Kingdom

81 ..

83

84b)

a) Participating students in participating classes. b) Calculated on all students in participating classes. c) 93% in Flemish and 74% in French speaking schools. d) An estimate not based on classrooms reports. It shows the proportion of participating students out of all 1987 born students b) in the country and not the number of students in participating classes.

40

Methodological considerations

very low in nearly all countries. The rather high response rates in nearly all countries and the reports about the reasons for not participating, do not indicate any major methodological problems connected with the response rates. Absent students are somewhat more prone to be involved in the use of various substances than is the case with students who are consistently in school (Grube and Morgan, 1989, Andersson and Hibell, 1995). A follow up study of students in Sweden shows that absent students had tried alcohol and illegal drugs more often than those present at the regular data collection (Andersson and Hibell ibid). Because of the relatively small number of absent students, the figures for the population as a whole were unchanged or only changed by one percentage point if absent students were included. In the school surveys in USA the corresponding average figure has been calculated to be 1.4% (Johnston et al, 2004). The difference in drug use between present and absent students may of course differ between countries and the effect of such differences is dependent upon the response rate. However, in the ESPAD context the alcohol and drug involvement among absent students is not a major methodological problem when students in different countries are compared.

Summary To summarise the issues related to representativeness one can conclude that the average age of participating students across countries was 15.7–15.9 years, that the samples were representative and that the

number of participating students was in line with the ESPAD protocol. In all countries but two a very large majority of those born in 1987 were enrolled in school (usually 90% or more). In a large majority of participating countries the proportion of students born in 1987 that were found in participating schools categories/grades was high (usually 90% or more). However, it was relatively low (below 80%) in five countries. School co-operation was satisfactory in most countries, even though many countries report problems with schools that were asked to participate in too many school surveys. Seven countries reported that 20% or more of the sampled schools or classes did not participate in the survey for this very reason. The representativeness of the surveys in some countries is somewhat uncertain. Austria, Belgium, Denmark, the Netherlands, Norway and the United Kingdom have a relatively large number of nonparticipating schools or classes. In Austria and Romania the gender distribution was skewed. In Bulgaria and Turkey a substantial proportion of the 1987 birth cohort were not enrolled in school. In Cyprus, Ireland and Romania a substantial proportion of the target population were not in the selected grades and in Greenland the response rate is unknown. The results of the surveys in these countries are nevertheless deemed to be sufficiently representative of students born in 1987. The fact that the Greek students in 2003 were seven months younger than in 1999 must be kept in mind when interpreting changes in the substance use figures from 1999 to 2003.

Reliability Reliability, which is a necessary condition for validity, is the extent to which repeated measurements used under the same conditions produce the same result. Data from different questions within the ESPAD questionnaire have been used to measure reliability. Two measures will be discussed. One is the inconsistency between two sets of questions measuring the lifetime prevalence for different drugs. The other is a quotient between the proportion of students who on the “honesty question” answered that they “already said” that they had used cannabis and the proportion who actually gave this answer. In the ESPAD methodology study in 1998 students in seven countries were asked to complete the

Methodological considerations

questionnaire on their use of alcohol and drugs on two separate occasions with a delay period on 3–5 days (Hibell et al. 2000). Since the studies were completely anonymous it was not possible to do a test-retest study limited only to individuals who participated in both data collections. No significant differences in the consumption patterns were found between the two data collections in any of the countries. This was true for alcohol consumption as well as drug prevalence which suggests that the reliability was very high in all seven ESPAD countries. Similar results with no significant differences were also reported from two repeated studies in Iceland and Hungary (Hibell et al. 1997).

41

Inconsistency in relation to lifetime use For many drugs the questionnaire contained questions about lifetime use. A later set of questions dealt with the age at first use of different drugs. These questions included the alternative “never”, which makes it possible to compare the prevalence of users of each drug according to these two questions. Table D includes information on the proportion of students reporting drug use on one question and not on the other, i.e. giving inconsistent answers. The lowest inconsistency figures were found for anabolic steroids and other illicit drugs than cannabis (explained in Table D). In nearly all countries inconsistency rates are 0 or 1%, demonstrating that 99–100% gave consistent answers in relation to the consumption of these substances. With some very few exceptions the figures were nearly as low for tranquillisers and sedatives without a doctor’s prescription. In about 80% of the countries the proportions with inconsistent answers were 3% or less. The highest figures were 6–7% and were reported from the Netherlands and Poland. The figures are in many cases low also for cannabis. In a majority of the countries inconsistent answers were given by 3% or less of the students. The highest figures were found in Belgium, Bulgaria, Greenland and Ukraine (6–8%). The figures are also rather similar for the use of inhalants as well as tranquillisers or sedatives without a doctor’s prescription. In about half of the countries 3% or less of the students gave inconsistent answers on their use of inhalants. The highest inconsistency figures are found in Greenland and Malta (10– 11%) followed by Cyprus, Greece, Iceland, Isle of Man, Latvia, Poland and Slovenia (6–7%). For cigarette smoking the proportion of inconsistent answers is somewhat higher (4–5%) with a majority of 5% or less. The highest figure is found in Turkey (15%), followed by Bulgaria, the Faroe Islands, Greenland, Latvia and Switzerland in which 7–8% of the students gave inconsistent answers on the lifetime prevalence of smoking cigarettes. Some countries had rather high inconsistency rates for the variable been drunk. The highest are found in Greenland (16%), Bulgaria, Latvia, Ukraine (12–14%) and Portugal (10%). However, rather low figures are found in most countries and in about half of them they are 5% or less. In most countries the inconsistency rates are low for all drugs. However, it is often lowest for anabolic steroids and “other illicit drugs” followed by tranquillisers and sedatives without a doctor’s pre-

42

scription, cannabis and inhalants. Somewhat less consistency is reported for the variables cigarette and drunkenness. Some of the high inconsistency rates can to a certain extent be explained by differences in the questions being matched. For instance the first question on inhalants was “On how many occasions (if any) have you sniffed a substance (glue, aerosols etc.) to get high?” In the second question some examples were omitted and it was written “When (if ever) did you FIRST do each of the following things?” One of the sub-questions was “Try inhalants (glue, etc) to get high”. The different examples might give rise to different perceptions of the variable content. Students may also have been ambivalent when answering the question about the age of the first use of a drug. If a student had only used a drug once or twice and did not define himself or herself as a user and therefore may not have found it appropriate to give an age when he or she started. These students may have answered “never” since they think of their consumption as an experiment rather than use. The question about the age at first use did not include a category like “I do not remember”. If a student did not remember there is probably a risk that he/she answers never instead of “guessing” about an age, especially if the person has used the substance a few times only. An other possibility could be that the student simply do not answer the question. There may also be other factors that complicate the interpretation of inconsistency rates. One is that the inconsistency rate may be affected by the prevalence rate. In other words, there are more people who can report their use inconsistently when there are more users in a country. However, there does not seem to be a strong relationship between high prevalence figures and high inconsistency figures. For none of the drugs the highest inconsistency figures are found in countries with the highest prevalence rates or the lowest found in countries with the lowest prevalence rates. It could also be argued that a given inconsistency figure (e.g. 1%) is more “serious” in country A where 5% admit drug use than in country B where 50% do so. In country A the inconsistency is 20% of the prevalence rate, but in country B it is only 2% of the prevalence rate. The importance of the size of the inconsistency in relation to the prevalence figure can be illustrated by the cannabis figures. In a majority of the countries the inconsistency figures are between 0–3%. The Romanian inconsistency figure of 1% might be seen as high

Methodological considerations

Table D. Some aspects of reliability. Two measures of inconsistency between two questions in a single administration. Percentages and quotients among all students. Country

Students reporting lifetime drug use on one question and not on the other (%)

Cigarettes

Austria

Been drunk

3

6

Inhalants

5

Cannabis

3

Other illicit drugsc)

1

Tranq. or sedat.d)

1

a)

Quotient between two questions b)

Anabolic steroids

Cannabis

1

0.9

Belgium

3

6

3

6

1

4

1

0.7

Bulgaria

8

12

3

7

1

2

2

1.1

Croatia

2

7

4

2

0

2

1

0.8

Cyprus

4

5

6

1

1

3

1

1.5

Czech Republic

2

3

3

3

1

5

0

0.8

Denmark

3

2

3

1

0

2

0

0.9

Estonia

5

4

3

5

1

3

1

0.8

Faroe Islands

7

3

3

2

1

1

0

1.2

Finland France

4 ..

2 ..

3 ..

1 ..

0 ..

2 ..

0 ..

0.9 ..

Germany

2

6

3

2

1

1

0

0.9

Greece

3

5

6

1

0

2

1

1.2

Greenland

7

16

11

6

1

1

0

0.9

Hungary

4

4

2

5

1

4

Iceland

2

2

7

1

0

3

1 ..

1.1

Ireland

1

1

1

1

1

1

1

0.9

Isle of Man

0

0

7

1

0

0

0

0.9

Italy

5

6

5

5

2

4

1

0.8

Latvia

7

13

6

5

1

3

1

1.0

Lithuania

3

6

1

2

0

1

1

0.8

Malta

3

7

1

2

1

1.0

4

5

10 ..

2

Netherlands

2

1

6

0

0.8

Norway

5

3

2

1

0

1

0

1.0

Poland

6

8

6

4

1

7

1

1.6

Portugal

3

10

5

4

1

3

1

0.9

Romania

6

7

1

1

0

2

0

1.7

Russia (Moscow)

5

7

5

3

1

1

1

0.8

Slovak Republic

6

5

3

3

0

2

0

0.8

Slovenia

5

8

6

3

1

2

1

0.9

0.8

Sweden

3

3

3

1

0

0

0

1.2

Switzerland

7

4

3

0

0

1

0

0.8

Turkey

15

8

3

2

3

2

4

0.7

Ukraine

6

14

4

8

1

1

1

0.4

United Kingdom

3

4

5

2

0

1

0

0.9

a) The first question is the self-reported lifetime prevalence question for the drug, while the second is a later one about the age at first use of the drug. b) Quotient a/b between the proportion answering “I already said that I have used it” to the question “If you ever used marijuana or hashish, b) do you think that you would have said so in this questionnaire?” (a) and the proportion who reported that they ever used it (b). c) Other illicit drugs include amphetamines, LSD and other hallucinogenes, crack, cocaine, ecstasy and heroin. The figure is an average for these drugs. d) Tranquillisers or sedatives without a doctor’s prescription.

Methodological considerations

43

considering that only 3% answered that they had used cannabis. Thus for Romania as a country the prevalence figure of 3% could be seen as uncertain. However, in the ESPAD context, when data are compared with results from other countries, it is not of “vital importance” whether the “true figure” is 2 or 4%, if the “true figures” in all other countries are (much) above this level. In the ESPAD context Romania is still a country where very few students have used cannabis. A more problematic inconsistency is found in Ukraine, where 21% admit that they have used cannabis but 8% give inconsistent answers, which means that “the true prevalence figure” may vary quite a lot (13–29%). In 27 of the 34 countries with available information, consistent answers were provided by 92% or more of the respondents, which must be seen as a satisfactory result. In 8 cases the values were 10% or above, which is a cause for concern since “the true prevalence” may very quite substantially compared to the reported figure. However, it seems rather unlikely that (nearly) all students would opt for one of the “extreme positions”, i.e. either denying real use or admitting use that never has occurred. With the exception of cigarette smoking in Turkey and the use of inhalants in Greenland and Malta all 10+ inconsistency rates were found for the variable been drunk. With the exception of Greenland no country has more than one 10+ figure. If one also includes inconsistency figures that are high in comparison to other figures for the same drug, a few countries with relatively high figures might include Bulgaria (been drunk and cannabis use), Greenland (been drunk, use of inhalants and cannabis use), Latvia (been drunk), Malta (use of inhalants), Poland (tranquillisers and sedatives without a doctor’s prescription), Portugal (been drunk), Turkey (cigarette smoking and use of anabolic steroids) and Ukraine (been drunk and cannabis use).

An inconsistency quotient The other measure of reliability is the quotient between the answers to two questions. One is about the willingness to admit the use of marijuana or hashish (the so called “honesty question”). The students were asked: “If you had ever used marijuana or hashish, do you think you would have said so in this questionnaire?”. The question could be used as a measure of validity and it is from this perspective that it is discussed in the next section. However, one of the response alternatives was “I

44

already said I have used it” and this proportion has been compared with the proportion that reported cannabis use on the lifetime prevalence question. Table D includes the quotient between these two proportions, with the “honesty answer” as the numerator and the “lifetime answer” as the denominator. A value of 1.0 means that the proportions are the same on both measures. The quotient is above 1.0 if more students answered that they already had said they have used the drug than actually reporting so on the direct question. Conversely, the quotient is below 1.0 if fewer students indicated that they have already admitted drug use than actually did admit to it on the direct question. The quotient is 1.0±0.2 in 28 out of the 34 countries where it was possible to calculate. It was above 1.2 in Romania (1.7), Poland (1.6) and Cyprus (1.5) and below 0.8 in Ukraine (0.4), Belgium (0.7) and Turkey (0.7). The Ukrainian ESPAD researcher has found that amongst those who reported lifetime cannabis use 7.3% answered “definitely yes” on the honesty question, which in some way also is a correct answer. If these answers are added to the 8.7% that answered “I have already said I have used it” the figure is 16.0%, which is rather close to the lifetime prevalence figure. This seems like a plausible explanation. However, if so, why does this mainly occur in Ukraine? (If one accepts this “recalculation” the quotient is changed to 0.8). For Romania, Cyprus and Turkey the deviant quotient measures are in part due to the low prevalence figures. Only 3–4% reported cannabis use on the lifetime prevalence question, which implies that only a rather few individuals can “cause” a high or a low quotient figure.

Summary In the ESPAD methodology study in 1998 reliability was high in all the seven participating countries. In the 2003 ESPAD study the inconsistency rates are rather satisfactory in most countries and for most measured variables. No country scores high on all variables. However, Greenland shows rather high inconsistencies on three out of the seven measures – having been drunk, inhalants and cannabis. Three countries showed high inconsistency measures for two variables. They are Bulgaria (been drunk and cannabis use), Turkey (cigarette smoking and use of anabolic steroids) and Ukraine (been drunk and cannabis use). Ukraine also reports a low inconsistency quotient for cannabis. Four countries reported a high inconsistency figure for one vari-

Methodological considerations

able, including Latvia (been drunk), Malta (use of inhalants), Poland (tranquillisers and sedatives without a doctor’s prescription) and Portugal (been drunk). Altogether the inconsistency measures demonstrate that reliability is good in most ESPAD

countries. However, in Bulgaria, Greenland, Latvia, Malta, Poland, Portugal, Turkey and Ukraine the reliability is probably somewhat lower for one or a few variables.

Validity The validity of answers is a major concern in survey research, in particular in surveys of sensitive behaviours like substance use. In ESPAD terms, validity could be said to be the degree to which the ESPAD questionnaire (including how data are collected) measures aspects of students’ consumption of different substances that we have decided to measure. Some researchers have used biological tests to study the validity of school surveys. Campanelli, Dielman and Shope (1987) found no significant differences in reported alcohol use between a control group and a group where saliva samples were collected prior to the survey. Kokkevi and Stefanis (1991) used urine samples collected after a school survey on drug use. Their findings validated students’ reports of recent cannabis use. In recent years hair analysis has also been used to validate survey data about drug use. However, Harrison (1997) has argued that most research conducted on validating self-report has focused on criminal justice and treatment populations and is thus limited in its ability to determine how accurately respondents report drug use in general population surveys, such as household and school surveys. Despite of the concerns with the generalizability of the results of most validation studies Harrison (1997) emphasizes some general conclusions. One is that the pattern of reporting is consistent with the social desirability hypothesis, i.e. that more stigmatised drugs are less validly reported than less stigmatised drugs. A second conclusion is that respondents are most willing to report lifetime use and least willing to report use that occurred in the very recent past. Third, self-administrated questionnaires tends to produce more valid data than interviews in which the respondents are required to give a verbal response. In a review of studies about drug use Morgan (1977) concludes that self-report methods for substance use are as reliable and valid as most other forms of behaviour. There are inconsistencies in

Methodological considerations

such reports from time to time as in denial that of earlier admitted use in longitudinal studies, but these also occur with other behaviours. Adding special conditions to enhance validity (like the bo gus pipeline) do not add anything to validity over and above anonymity and confidentiality. Morgan also concludes that when discrepancies occur between self-reports and other indices (physiological, collateral reports), it cannot be assumed that the self-reports are necessarily the less valid measure. Finally, self-reports have the greatest claim to construct validity, that is, the measures related in predicted ways to other outcomes and to antecedent factors. In a discussion on validity in school surveys of USA Johnston and O’Malley (1985) also conclude on the bases of considerable inferential evidence that self report questions produce largely valid data. High reliability is a necessary but not sufficient condition for validity. In the previous section it was concluded that the test-retest reliability was high in seven countries in the ESPAD methodology study as well as in two countries where such studies were conducted separately with the ESPAD questionnaire. It was also concluded that the inconsistency measures using a high level of reliability in most countries and for most drugs. However, this is in itself not enough to secure high validity.

Student co-operation The primary condition for obtaining any data is that students in selected classes actually receive the questionnaire and are willing to respond to it. The first condition is nullified if the school or the teacher refuses to co-operate. If students do receive the questionnaire they must have enough time to complete it, understand the questions and they must be willing to answer the questions honestly. The participation in the study was of course voluntary. However, in nearly all countries none or very few students were reported to have refused to participate. On the contrary, in many countries the

45

classroom reports state that many students were very interested in answering the questionnaire. In a few countries it was necessary to get parental permission before students were allowed to participate in the project. Countries where parental permission was compulsory include France, Norway and the United Kingdom. In France as well as in the United Kingdom 1% of the parents refused their children to take part in the study. The corresponding figure was low also in Norway. Thus, parents refusing their children to participate in the ESPAD study is only a very limited problem. A visual inspection of each questionnaire, sometimes combined with computer screening, was undertaken before data entry into the national databases. With very few exceptions, only a small fraction of all questionnaires were excluded during the scrutinising process. On average 1.0% of the questionnaires were excluded for that reason (Table B). However, there are a few countries which reported higher proportions of eliminated questionnaires, including Cyprus (5.0%), Isle of Man (3.6%), Greece (2.3%) and Portugal (2.3%). Unfortunately, information is not available from two of the ESPAD countries. The survey leaders were asked to fill out classroom reports about disturbances during the data collection, the students interest in the survey as well as whether the students worked seriously. In 21 of 32 countries with available information 60% or more of the survey leaders did not report any disturbances during data collection (Table E). The highest figures were found in Cyprus (100%), Ireland (97%) and Croatia (95%) and the lowest in Russia (Moscow) (24%), the Slovak Republic (36%) and Belgium (41%). The highest proportions which reported disturbances from more than a few students are found in Greece, Russia (Moscow) and Turkey (16–18%) together with Belgium and the Slovak Republic (14% each). In most countries giggles or eye makings were the most commonly reported disturbances. It should be noted that research assistants were responsible for data collection in all countries with widespread reported disturbances. Since they are not used to the “normal level of disturbance” in a classroom they are probably much more sensitive than teachers for different kinds of disturbances and, consequently, report them to a much higher degree. In three of these countries (Belgium, Poland and Russia (Moscow)) the research assistants had received special instructions to report all kinds of disturbances.

46

In nearly all countries a very large majority of the survey leaders (91–100%) reported that “all”, “nearly all” or “a majority” of the students were interested in the study, and 75–100% reported that “all” or “nearly all” students were interested (Table E). The smallest proportions were reported from Slovenia (58%) and Turkey (68%). The figures were very similar on the question of whether the students worked seriously. Nearly all data collection leaders (95–100%) answered that “all”, “nearly all” or “a majority” of the students worked seriously on the questionnaire (Table E). With the exception of three countries the proportions answering “all” or “nearly all” were 75–100%. Again the exceptions were Turkey (65%) and Slovenia (69%), as well as Russia (Moscow) (69%). Unfortunately, data from the survey leaders from Isle of Man and the United Kingdom were not available following an oversight in which the classroom reports were not used. However, from other indices gleaned from the country reports student co-operation was on par with that reported by other countries. In summary, no countries reported problems with many students refusing to participate. The propor tion of eliminated questionnaires was low in nearly all countries with 5.0% as the maximum figure. When disturbances did occur this rarely involved more than a few students. Even if some disturbances were reported in some countries, they seem very seldom to have negatively affected the student co-operation. Most survey leaders reported that the students were interested in the study and worked seriously. Over all, student co-operation seems to have been good or very good in all participating countries.

Student comprehension The number of questions included in the questionnaire varies somewhat between countries. Naturally, the length of the questionnaire has a direct effect on the time taken to complete it. In addition, a difference between students’ experience in participating in these types of studies would also affect the time to complete questionnaires. For these and other reasons, it is not surprising that the time taken to complete the questionnaire varied between countries. The average time to complete the questionnaire varied between 30 and 50 minutes in most countries (Table B). The highest figure (69 minutes) was reported from Greenland. A rather long time was

Methodological considerations

Table E. Opinions of the data collection leaders a). Percentages. Country

Disturbances during the completion of the questionnaire

Kind of disturbances b)

No

Giggles or eye makings

A few students

More

Loud comments

Student co-operation Other comments

Students interested c)

Students worked seriously d)

Austria

76

20

5

5

12

7

95(77)

99(86)

Belgium e)

41

45

14

26

13

34

92(80)

93(78)

Bulgaria

56

34

10

30

14

9

97(85)

97(89)

Croatia

95

4

1

2

3

2

100(100)

100(95)

Cyprus

100





5

3

3

95 (83)

95 (83)

Czech Republic

61

32

6

31

5

3

99(92)

98(88)

Denmark

84

13

2

7

8

9

99(95)

100(99)

Estonia

51

39

10

41

14



89(72)

96(83)

Faroe Islands

81

16

3

10



6

100(100)

100(91)

Finland

76

2 ..

13

13

96(84)

62

22 ..

8

France

30

12

11

96(78)

99(94) ..

Germany

81

15

3

Greece

56

29

Greenland

68

Hungary

75

Iceland

71

Ireland Isle of Man g)

97 ..

Italy

56

37

7

Latvia

67

27

6

5f)

10f)

2f)

96(72)

99(82)

16

..

39

5

92(81)

92(81)

30

2

21

42

37

100(93)

97(93)

20

5

18

5

2

97(87)

98(91)

23

6

16

1



96(88)

100(96)

3 ..

– ..

3 ..

– ..

– ..

100(100) ..

100(100) ..

30

21

98(86)

14

2 ..

94(79)

21

94(79)

95(79)

Lithuania

72

24

11

17

11

1

96(86)

99(88)

Malta

83

17



17





97(88)

5

4

18

98(86) ..

99(96)

18

1

10

7

6

96(89)

99(93)

Netherlands

81

Norway

81

19h)

Poland

54

36

10

32

49

15

90(81)

92(74)

Portugal

69

26

6

25

9

5

98(86)

99(88)

Romania

90

8

2

10

2

0

98(92)

98(92)

Russia (Moscow)

24

60

16

53

7

1

93(72)

92(69)

Slovak Republic

36

50

14

46

16

21

97(86)

97(86)

24

13

98(69)

15

9 ..

92(58)

24

90 (82)

100 (96)

Slovenia

57

Sweden

59

43h) 34

6

Switzerland

70

28

2

25

10

9

94(77)

100(94)

Turkey

54

28

18

36

13

8

89 (68)

92 (65)

Ukraine

48 ..

41 ..

11 ..

40 ..

15 ..

7 ..

99 (88) ..

100 (86) ..

United Kingdom g)

a) In countries where more than one age group participated, the information is usually based on all participating students. b) Percent of participating classes. c) “All”, “Nearly all” or “A majority” of the students were reported to have been uninterested in the survey (within brackets: “All” or “Nearly all” students). d) “All”, “Nearly all” or “A majority” of the students were reported to have worked seriously (within brackets: “All” or “Nearly all” students). e) Information is only available from the Flemish speaking areas. f) Classifications of free text answers. g) The ESPAD classroom report was not used. h) Only two answering categories were used (yes/no).

Methodological considerations

47

also utilised in Isle of Man and Romania with 60 minutes each. No countries reported that students refused to complete the questionnaire as a result of its length. On the other hand, one of the most frequent comments was that the questionnaire was long and repetitive. Nor were there any countries that reported any major problems on the ability of students to understand the questionnaire. Overall, student comprehension seems to have been satisfactory in all participating countries.

Anonymity The validity of answers in surveys related to illegal behaviour, such as drug use, is dependent upon the respondents’ trusting that reporting such behaviour would not result in any negative consequences. Thus, it is important that the students perceive the survey to be anonymous. Several measures were taken to ensure the perceived as well as the actual anonymity of the ESPAD survey. The ESPAD protocol recommends distributing an envelope for each student to seal after having answered the questions. In 25 ESPAD countries individual envelopes were used (Table A). Countries that did not use individual envelopes used other methods to secure that the students felt that their anonymity was secured. These methods included a closed box and a large envelope for the entire class, often sealed in front of the class before being transported to the research institute. It is also important that the students trust that the data collection leaders do not look at their answers. He or she could either be a teacher or a research assistant. In some countries with long traditions of school surveys students are used to teachers taking responsibility for the data collection. In other countries research assistants, or other persons not affiliated to the school, administered the questionnaire. The decision on the most suitable data collection leader was taken by each country independently. The base for that decision should of course be to choose the person most trusted by the students. In a methodological study in Iceland, Bjarnason (1995) found no significant differences in either the reported prevalence or the reported frequency of drug use between randomly selected classes responding to the ESPAD questionnaire administered by their teachers and randomly selected classes that had the questionnaire administered to them by research assistants. These findings suggest that at least in some countries the mode of administration does not significantly affect the results of school

48

surveys on drug use. It can thus be inferred that results obtained by a teacher administrator are fully comparable with results obtained by research assistants in countries where mode of administration may be more sensible. In about half of the ESPAD countries teachers were data collection leaders, while more than one third choose research assistants (Table A). A few schools used health staff. The data collection leader was asked to stress the question of anonymity and to refrain from walking around in the classroom while the questionnaires were completed. The students were instructed verbally and in writing on the first page of the questionnaire that they should not put their names on the questionnaires or the envelopes. No country reported any serious doubts about the anonymity aspect. As a whole, the question of anonymity seems to have been handled satisfactory in all participating countries.

Missing data rates In the instructions to the students it was stressed that it was important to answer each question as thoughtfully and frankly as possible. However, since participation in the study was voluntary they were told that they could skip any questions they found objectionable for any reason. Thus, missing data rates on drug questions can be seen as an indicator of the respondents’ willingness to report drug use. Of special interest are possible differences in missing data rates between different drugs and between drug questions and other questions. Looking at the questionnaire as a whole the proportion of unanswered questions is low in most countries. In about two thirds of the countries with available information only 0–2% of the questions were unanswered (Table F). In only two it exceeded 5%. Because of mistakes in the layout and coding of multiple questions 21% of the data were missing in Estonia. The proportion of unanswered questions in Greenland was 10%. The high rate of missing values in Estonia is limited to a relatively small number of questions and does therefore not signal a threat to validity of the questions about substance use. Some caution should however be exercised in the interpretation of Greenlandic results as the rate of missing values indicates a reluctance by students to provide honest responses. In some few countries the proportion of unanswered questions varies a little between core, module and own questions. The core ESPAD questions are to be situated in the beginning of the national

Methodological considerations

Table F. Proportions of unanswered questions. All students. Country

Cigarettesa)

Alcoholb)

Been drunkb)

Inhalantsb)

Cannabisb)

Other illegal drugsc)

Tranq. or sed.d)

Austria

1

4(4)

5(2)

2(1)

Belgium

1

2(3)

2(2)

1(1)

Bulgaria

2

5(6)

5(4)

Croatia

0

1(1)

1(0)

Anabol- Core ic stero- questidse) ions

Module questions

Own questions

2(1)

1

1

1

1(1)

1

1

2

3(1)

3(1)

2

1

1(0)

1(0)

0

0

All questions

1

2

4

2

2

3

7

3

2

6

3



5

0

2 ..

4 ..

1 ..

4

4

2

Cyprus

0

2(2)

1(1)

0(1)

0(1)

0

0

0

1 ..

Czech Republic

1

2(2)

2(1)

1(0)

1(1)

0

0

0

1

Denmark

0

3(3)

3(2)

2(1)

2(1)

2

2

2

1

2

5

1

Estonia

1

3(3)

3(2)

1(1)

1(1)

2

2

2

25f)

3

0

21f)

Faroe Islands

1

5(2)

4(1)

3(1)

3(1)

2

2

2

6

3

5

5

Finland

1(0)

2(0)

2(0)

2(0)

1

1

1

1

1

5(3)

5(2)

2(1)

3(1)

2

1

3

3

3

2 ..

1

France

0 ..

3

Germany

0

2(2)

1(1)

1(0)

1(0)

0

0

1

1

1

2

1

Greece

1

2(2)

2(1)

1(0)

1(0)

1

0

1

1

1

1

1

Greenland

5

12(11)

13(14)

12(9)

12(10)

8

8

8

10

17

13

10

Hungary

1

4(3)

3(2)

1(0)

1(0)

1

0

1

2

3



2

1g)

4g)

2g)

Iceland

0

2(1)

2(1)

1(1)

1(0)

0

0

0

1g)

Ireland

0

4(4)

5(3)

3(1)

3(1)

2

2

2

2 ..

..

2

..

2

Isle of Man

1

3(3)

3(2)

1(0)

1(1)

1

1

0

2 ..

Italy

0

2(1)

2(1)

3(2)

3(2)

2

2

2

2





2

Latvia

0

3(2)

3(1)

0(0)

0(0)

0

0

0

1

3

4

2

Lithuania

0

0(0)

0(0)

0(0)

0(0)

0

0

0

0

0

..

0

Malta

1

4(3)

2(1)

2(1)

3(1)

1

1

1

3

3

1

4(3)

2(1)

2(0)

2(0)

1

1

2

3

2 ..



Netherlands

13

3

Norway

1

7(3)

6(3)

7(3)

6(3)

4

4

5

Poland

1

2(2)

2(1)

1(0)

1(1)

1

1

1

3 ..

4 ..

10 ..

3 ..

Portugal

1

7(7)

4(3)

3(1)

3(2)

1

1

1

..

..

..

..

Romania

1

4(3)

3(1)

3(1)

3(1)

2

1

1

2

4

Russia (Moscow)

1

3(3)

4(2)

2(0)

2(1)

1

1

1

2

1

– ..

2

2

Slovak Republic

1

2(2)

2(2)

1(0)

1(0)

1

1

1

1

3

11

2

Slovenia

0

3(1)

2(1)

1(0)

1(0)

1

1

1

1

3

1

1

Sweden

1

3(1)

3(1)

2(1)

2(1)

2

2

2

2

2

7

2

Switzerland



0(0)

1(1)

0(0)

0(0)

0

0

0

Turkey

0

5(1)

8(4)

6(2)

5(2)

4

2

5

2 ..

2 ..

8 ..

1 ..

Ukraine

0

5(4)

4(3)

1(1)

2(1)

2

1

2

2

2

..

2

United Kingdom

0

3(3)

2(2)

1(1)

1(1)

1

1

1

1

5

7

1

a) Average for lifetime and 30 days prevalence. b) Average for lifetime, 12 months and 30 days prevalence. Figures within brackets = lifetime prevalence only. c) Other illegal drugs include amphetamines, LSD and other hallucinogenes, crack, cocaine, ecstasy, heroin and drugs by injection. b) The figure is an average of lifetime prevalence for these drugs. d) Tranquillisers or sedatives without a doctor’s prescription. Lifetime prevalence. e) Lifetime prevalence. f) The high proportion of unanswered core quesstions is related to mistakes in how Q37 and some other multiple questions b) were layouted and coded. This also “explain” the large number of unanswered questions in the questionnaire as a whole. g) Based on those students that answered questionnaire A, i.e. the questionnaire that included almost all ESPAD core questions.

Methodological considerations

49

questionnaire and generally the rate of missing values for these questions was equal to or lower than the rate for country-specific question. The proportions of unanswered questions for different substances are low for all drugs in most countries (usually 1–3%). It should be noticed, however, that they are higher in a few countries, including Greenland (high on all questions), Norway (rather high for illigal substances), Turkey (rather high for most substances) and Portugal (rather high for alcohol consumption). Apart from these concerns, the proportions of unanswered questions about the consumption of different substances does not constitute any methodological problems. The proportion of unanswered questions in Greenland in the questionnaire as a whole (10%) was about the same as it was for most drug related variables. Consequently, it is mainly in Greenland that the proportion of unanswered questions, in the questionnaire as a whole as well as for questions on consumption of different substances, is so high that it needs careful consideration when interpreting the results.

Logical consistency Closely related to the inconsistency measures discussed in the reliability section is the logical consistency. In the ESPAD project this is relevant for drug questions measuring the prevalence for the three time periods, namely lifetime, last 12 months and last 30 days. Logically the last 12 months prevalence cannot exceed the lifetime prevalence and the same is true for the last 30 days prevalence when compared with the last 12 months and lifetime prevalence. Table G includes information on the proportion of inconsistent answers related to the three time periods for four variables; alcohol use (any alcoholic beverage), been drunk, cannabis use and use of inhalants. In nearly all countries and for all four variables, the reported proportions of inconsistent answers are very low. In other words, the proportion giving logically consistent answers across the three time periods is very high, usually 98% or more. Rather high proportions of inconsistent answers are only found in a few countries and are concentrated on the two alcohol related variables. Inconsistent answers on these two questions are mainly reported from Greenland (10–12%), Bulgaria (9– 10%), Ukraine (8–10%) and Portugal (7–10%). A high figure for alcohol use is also found in Cyprus (10%).

50

Faking good Social desirability is an important methodological problem in all surveys, i.e. the tendency of respondents to give answers that they believe show them in a desirable light in the eyes of others. This becomes particularly important in surveys on behaviour that is not accepted by some social groups or are even illegal. In addition to the methods discussed above, it is possible to gauge the magnitude of the social desirability effect by asking respondents directly about the honesty of their responses. In the ESPAD methodology study in seven countries data were collected twice with a lag time of 3–5 days (Hibell et al. 2000). The second time the questionnaire included some additional questions about the first study. One of them was whether they answered honestly to the questions on their drug consumption and another whether they thought that their classmates answered honestly. Nearly all students in the seven countries said that they answered honestly to the questions related to their alcohol and drug habits. With some few exceptions, 95% or more of the students said yes. Students were more sceptical about the honesty of their classmates, but the large majority nevertheless thought that “all” or “most” of their classmates answered honestly about their use of alcohol and drugs. About 85% or more of the students said that all or most of their classmates answered honestly to the questions about their consumption of the different substances. At the end of the international ESPAD questionnaire the students were asked two questions on their willingness to admit drug use in a hypothetical fashion. The wording of the first question was “If you had ever used marijuana or hashish, do you think that you would have said so in this questionnaire?” The second question asked in the same fashion about heroin use. The response alternatives were “I already said that I have used it”, “Definitely yes”, “Probably yes”, “Probably not” and “Definitely not”. The proportion of students reporting that they would definitely not report drug use is shown in Table G. In two-thirds of the countries with available information 7% or less answered that they definitely were unwilling to admit cannabis use if they had used it. The highest figure is reported from Greenland (30%) followed by Malta (13%), Croatia (12%), Latvia (12%) and Lithuania (10%). In line with social desirability concerns the willingness to admit heroin use is slightly lower than

Methodological considerations

Table G. Some aspects of validity: Inconsistent answers, unwillingness to admit drug use and reported knowledge and use of the dummy drug “relevin”. Percentages among all students. Country

Inconsistent answers a) Alcohol c)

Been drunk

Unwillingness to admit drug use b)

Dummy drug “relevin”

Cannabis

Inhalants

Cannabis

Heroin

Heard of

11

Reported own use

Austria

3

3

2

2

7

11

Belgium

4

2

1

0

5

9

Bulgaria

10

9

1

1

8

9

10

0.8

Croatia

3

2

1

0

12

15

14

0.2

Cyprus

10

4

1

2

6

6

10

0.3

2

1

0

0

3

7

9

0.2

Czech Republic

8g)

0.5 0.3g)

Denmark

1

1

0

0

3

5

6

0.1

Estonia

3

1

0

0

8

9

9

0.2

Faroe Islands

2

1





3

3

5

0.3

Finland

1

1

0

0

2

2

0

4 ..



5

2 ..

8

France

8d)

0.4d)

Germany

3

2

1

0

4

9

11

0.4

Greece

7

3

1

1

4

4

9

0.2

Greenland

10

12

3

2

30

46

5

0.2

Hungary

4

2

1

0

6

7

7

0.3

Iceland

2

1

1

1

5

8

11

0.7

Ireland

1

1

1

1

5

10

14

0.4

Isle of Man









7

12

16

0.6

Italy

5

3

1

1

4

7

11

1.2

Latvia

2

2

1

0

12

13

6

0.1

Lithuania

0

1

0

0

10

10

0

0.1

Malta

5

3

1

1

13

15

12

0.4

Netherlands

2

2

0

0

6

9

13e)

0.9e)

Norway

1

1

0

0

3

3

11

0.4

Poland

5

5

5

6

8

10

12

1.0

Portugal

10

7

2

1

4

5

9

0.8

Romania

5

4

0

0

8

7

11

0.1

Russia (Moscow)

6

7

4

2

5

8

10

0.1

Slovak Republic

3

3

2

1

3

5

8

0.0

Slovenia

5

3

1

1

4

6

7

0.1

Sweden

1

1

0

0

7

7

12

0.2

Switzerland

3f)

2f)

1f)

0f)

5

9

8

0.4

Turkey

4

3

1

1

3

3

9

1.3

Ukraine

10

8

1

0

8

9

6

0.4

2

2

1

0

7

14

16

0.1

United Kingdom

a) For each drug, inconsistent response pattern is defined as one in which any of the following is found: (a) thirty-day frequency is higher than annual frequency, b) thirty-day frequency is higher than lifetime frequency, or (c) annual frequency is higher than lifetime frequency. b) Students answering “definitely not” on the question “If you had ever used marijuana or hashish, do you think that you would have said so in this questionnaire?” b) and the corresponding question for heroin. c) Any alcoholic beverage. d) MOP was used as a dummy drug instead of relevin. e) NSTC was used as a dummy drug instead of relevin. f) Before the data cleaning process. g) NTSC/BKR was used as a dummy drug instead of relevin.

Methodological considerations

51

for cannabis in many countries. Fifteen countries have proportions of 7% or less. The highest figures are found in Greenland (46%), Croatia (15%), Malta (15%), the United Kingdom (14%), Latvia (13%), Isle of Man (12%), Austria (11%) and Ireland (10%), i.e. to a large extent the same countries that also reported high proportions of students that were unwilling to admit to cannabis use. A high proportion of students answering that they would not be willing to admit drug use may signal problems with validity, but this is not necessarily the case. Students who have never used drugs tend to be rather strongly opposed to their use and this opposition may in part be reflected in their answers to these questions. To the extent that responses to this question reflects the opinions of the non-drug using population these questions give a pessimistic view of the actual willingness of the drug using population to report their use of different substances. It should also be born in mind that the questions are hypothetical. If a student really tries cannabis in the future, he or she might be willing to admit that in a survey even if he or she answered negatively in the ESPAD questionnaire. Combining these two arguments give rise to a third reflection. If a student in the future decides to try an illegal drug for the first time, the same reasons behind that change might also be the reasons for a changed willingness to admit that use. The questions on the hypothetical willingness to report drug use may be most useful in a crosscultural context. In countries where a high proportion would definitely not admit such use many adolescents apparently consider it so shameful that they could not hypothetically imagine reporting it. The figures of unwillingness to admit drug use are rather high in some countries but much smaller in others, indicating that a probable underreporting may differ somewhat between countries. Students in Greenland are extremely reluctant to admit the use of both cannabis (30%) and heroin (46%). Countries with rather high figures (12+%) for both drugs also include Croatia, Latvia and Malta. It can be concluded that self-reported surveys most likely underestimate the prevalence of drug use and that underreporting probably differs somewhat between countries. It also seems reasonable to assume that underreporting to some extent differs between drugs. There is, however, no reason to believe that such differences undermine the overall conclusions of the study. However, the high figures for Greenland should be kept in mind.

52

Faking bad In addition to the risk of underreporting in drug surveys, the tendency of some adolescents to pretend they have used drugs can pose a threat to validity. To test this, the non-existent dummy drug “relevin” was included among real drugs in the questionnaire. The plausibility of this drug name is reflected in the fact that on average 9% of the students believe they have heard about it before. However, as shown in Table G, very few students report having used the dummy drug. In all participating countries but three the figure is 0.9% or less, with an average of 0.4%. However, in neither of these three countries the figure exceeds 1.3%. Very few students have answered that they have used the dummy drug relevin, which could be seen as a clear indicator that students do not routinely exaggerate drug experience. Thus, it seems reasonable to assume that high prevalence rates of drug use in practice nearly are unaffected by a possible general tendency to exaggerate drug use. However, these results also underline the need for caution in interpreting the prevalence of less common drugs such as heroin and LSD. For each country, the proportion reporting use of the non-existant drug relevin could be used as a baseline for plausibility. If 0.9% of students in a given country have used a non-existing drug, the first 0.9% of students reporting using existing drugs should be interpreted with extreme caution.

Construct validity The using of existing theories, results from earlier studies and logical inference, makes it possible to evaluate the extent to which variables are related to one another in a valid fashion. Such construct validity was discussed rather extensively in the Pompidou six-country pilot study which provided the base for the ESPAD questionnaire. The conclusion was that “there is considerable evidence of construct validity in the current data sets” (Johnston et al. 1994). For instance, it is logical to expect that countries with high proportions of students reporting use of different drugs also should have high proportions reporting drug use among friends. This was tested in the 1995 ESPAD report with the outcome of very strong relationships for LSD (rxy = 0.95), cannabis (rxy = 0.92) as well as for drunkenness (rxy = 0.87), which indicate a high validity (Hibell et al. 1997).

Methodological considerations

The validity of the questionnaire The comparability of the questionnaire across countries is of vital importance in any multi-national survey project. The equivalency of the translation of questions into different languages is therefore an important aspect of validity. The standard ESPAD questionnaire is written in English. In nonEnglish speaking countries the questionnaire was translated to the native language and then translated back by another translator and then both the original and the back translated version were compared for anomalies. However, the equivalency of questionnaires is not only a matter of literal translation. It is also a matter of equivalent understanding. Thus, the question per se should be “understood” in the same way in all countries irrespective of the original wording in the model questionnaire. When necessary, the questions have been “culturally adjusted” to the situation in a country. For instance drugs or nicknames should be adjusted to the situation in each single country. If this is not done correctly, it may pose difficulties for comparisons with other countries. In Austria and Germany the fixed answering categories to the questions about alcohol consumption at the last drinking occasion were changed to open alternatives. However, the answers to these open ended questions are judged not to be comparable with the answers given in other countries that have used the fixed answering categories. Hence, these data will be presented separately in the tables. For instance, the concept “drunkenness” is difficult to translate in equivalent terms into different languages. In the 2003 Russian (Moscow) survey a new translation of drunkenness was used. It was a little less harsh than the earlier translation and was tested in a split half test among participating students in Moscow. The new translation resulted in more students providing an affirmative answer on drunkenness (for example 24% compared with 15% for being drunk 20 times or more often). The Russian ESPAD researchers concluded that the new translation is more appropriate and that it should be used in the chapter that describes the situation in 2003. However, the old version will be used for comparisons between the 1995, 1999 and 2003 surveys. With some few exceptions no country reported any major problems with the translation of the questionnaire. Thus, it seems reasonable to assume that the translation of the questionnaire is a non-issue and does not jeopardise the possibility to com-

Methodological considerations

pare results between the ESPAD countries. In the few cases when this was not so it is commented on in the result chapter.

The cultural context The standardisation of the different steps in the data collection procedure was the adopted method by the ESPAD project to provide as much as possible a suitable framework for comparability between countries. This included the target population, the questionnaire and how data were collected and treated, all of which have been described in earlier sections. However, as already stressed in the introduction of this chapter, it has not been possible to standardise every detail. This holds true for the cultural contexts in which the students have provided their replies. The role of cultural context will be discussed from two perspectives. One is whether the questions are understood or perceived in the same way in all countries and the other the willingness to give true/valid answers. To allow comparisons between countries it is necessary that students answer the same questions. All countries but one included (nearly) all core questions while others also used the module and optional questions of the ESPAD questionnaire. In the section “The validity of the questionnaire” it was described how the questionnaires were translated and “culturally adjusted”. No major problems were reported in this process. However, even if no single researcher noticed any “problems” in his or her own country, i.e. that the questions were not technically correct, one cannot automatically assume that students in different countries did not perceive them any differently. Does, for example the word “solvent”, even if exemplified, signify the same thing for a student in Ukraine, Norway or Italy? “Being drunk” may mean many different things for students in Iceland, Hungary and Portugal? Apparently one cannot ascertain in total whether students in different countries have understood the questions in the same way. On the other hand, for most variables the differences between high and low prevalence countries are considerable and it seems very unlikely that possible differences in the understanding or perception of some questions paves the way to “explaining” these differences. Earlier in this section, different indices for cultural context have been elaborated. Student co-operation, missing data rates and reported willingness to answer honestly differ somewhat between coun-

53

tries, which is suggestive that the cultural context in which the questions have been answered may vary between countries. However, for each of these indicators only a rather few countries seem to differ in any major way from any of the others. Other validity indicators, including student comprehension and reported dummy drug use, do not prompt for any important differences between participating countries. The willingness to admit drug use may be influenced by societal attitudes towards a given drug. The results from the ESPAD project show that perceived risk of substance use and disapproval of different types of substance use differ between countries. The same is also true in relation to the availability of different drugs. Taken together, these results indicate that social desirability may vary between countries. Thus, in a country with low availability and negative attitudes towards drugs a student might be less willing to admit drug use than a student in a country with high availability and positive attitudes towards drugs. Similar issues may also be relevant when considering that in some countries drugs and drug use are often mentioned in mass media and discussed at school, while the situation may be the opposite in others. Some ESPAD countries have long traditions in the conduct of school surveys while the ESPAD study was the first in others. These different traditions and, consequently, differences in the students experiences of surveys could in principle affect the willingness to answer honestly and thus this may differ between countries. One of the conclusions of the methodological discussions in the ESPAD 95 report (Hibell et al. 1997) was that the cultural context in which the students answered the questions most probably differed between countries and that one could not exclude that these differences may have differently impacted on the willingness to answer honestly. To obtain a better insight into the effects of cultural context, the ESPAD methodology project was conducted in 1998 (Hibell et al. 2000). The answers from the students about their own honesty and the expected honesty of their classmates, as well as data from the survey leaders, clearly indicated a high reliability and validity in the seven participating countries. It could not be excluded, however, that the validity might have been slightly lower in one or two out of the seven participating countries (Cyprus, Denmark, Lithuania, Malta, Ukraine, the Slovak Republic and Sweden; i.e.

54

countries in different parts of Europe). The cultural context in which the students answered the questions most probably differed between the seven countries. However, it does not seem plausible that validity differed very much. One reason for this outcome, indicated by the methodology study, might be that the students really trusted that anonymity and confidentiality would be observed. Even if some doubts remain on the effect of cultural context for the validity, especially in countries that did not participate in the methodology study, it does not seem likely that the “true” answer in a low prevalence country (e.g. 2% admitting cannabis use) should be more than doubled or tripled (i.e. above 4–6%) and that the “true” figure in a high prevalence country (e.g. 30%) should not be somewhere between ±5% (i.e. between 25–35%). Thus, a low prevalence country is most probably still a low prevalence country “in reality” and a high prevalence country “still” a high prevalence country, even if the exact difference between the two countries is not known for certain. However, it may be difficult to draw any firm conclusions about significant differences between countries with only small differences in prevalence figures.

Summary An analysis of available information strongly suggests that the validity of the ESPAD studies is high in most countries. These indicators include student co-operation, student comprehension, anonymity, reported dummy drug use and construct validity. The main threats to validity are related to missing data rates, logical inconsistencies and reported lack of willingness to answer honestly. Validity problems are encountered in a limited number of countries, mainly Greenland but to some extent also Croatia, Latvia and Malta. However, it should be noted that with the exception of Greenland, none of these countries are indicated on more than one of the validity measures. The importance of the cultural context should not be underestimated, but responses by students and survey leaders in the ESPAD methodology project indicated that the students usually answered rather honestly. These conclusions are also supported in the present study by the very large proportion of the data collection leaders that reported that students were interested in the study and worked seriously. Validity problems seem to be limited in scope and to affect only a few countries.

Methodological considerations

Comparisons with other survey data In some ESPAD countries data are available from other studies measuring alcohol and drug habits among youth. Comparisons between those data and results from the ESPAD study can provide valuable information on whether differences in alcohol and drug habits between students in different ESPAD countries are realistic. In this perspective, results from two studies in a country do not have to be exactly the same. What is important is that they are of the similar magnitude. It could be questioned whether this is a measure of validity or not. Even if the results of two surveys are similar one could argue that this is not sufficient proof for validity. However, the general consensus is that school surveys usually do provide rather valid results, thus comparisons with other data should further provide valuable insights as to the validity of the ESPAD project, at least in countries with comparable data. Comparable data are available in Sweden, Norway and the Netherlands. Comparisons on four variables from the Study of Health Behaviour in School-aged Children (HBSC) (Currie et al. 2004) are discussed below Data accrued in the studies used for comparisons are not always collected in the same way, with the use of same questions or on exactly the same age groups. The most important methodological differences are mentioned in the tables (H–N). Again, these differences stress the importance of focusing on magnitudes rather than on exact figures. In Norway the figures for most variables are similar in both studies (Table H). The proportion that said that they had used any alcohol in their lifetime was slightly higher in the ESPAD study compared to that obtained from a national survey that employed the use of mailed questionnaires. However, the latter survey specified a lower limit of at least a bottle of beer or 10 cl of wine or 2.5 cl of spirits but the ESPAD did not contain any minimum quantities so the difference between the two studies seems reasonable. For all other variables the figures are remarkably similar, including measures related to three different time frames, i.e. lifetime (intoxication, use of cannabis, use of amphetamines and use of inhalants), last 12 months (intoxication, use of cannabis and use of inhalants) and last 30 days (any alcohol and cigarette smoking). In Sweden slightly more boys in the ESPAD study answered that they have ever been drunk and

Methodological considerations

that they were drunk at the age of 13 or younger compared to estimates from the regular national school survey in 2003, while for the remaining five variables there were no differences of note (Table I). Among girls there were no differences at all for any of the seven variables. The questions on drunkenness were not the same in the two surveys, which may be a source for the difference in the answers. However, in the total ESPAD context, figures for lifetime prevalence for boys range from 25 to 87% while figures for being drunk at the age of 13 or earlier range from 8 to 42%, the differences between the two Swedish studies among boys are probably of minor importance. A third country with information from another school survey is the Netherlands. It was conducted in parallel with the ESPAD study and used the same questionnaire with some minor differences. Hence, the Dutch comparison should be seen more Table H. Alcohol and drug use in Norway. Frequency of lifetime, last 12 months and last 30 days use. Data from ESPAD and a national survey in 2003. Percentages among all respondents a). ESPAD

National survey b)

15–16 years

15–16 years

Lifetime Any alcohol

84

072c)

Intoxicated

59

056c)

Cannabis

09

008c)

Amphetamines

02

002c)

Inhalants

06

005c)

Last 12 months Intoxicated

54

52 (last 6 months)

Cannabis

06

06 (last 6 months)

Inhalants

03

02 (last 6 months)

Any alcohol

51

051c)

Smoke cigarettes

28

27 (smoke tobacco)

3,833

563c)

Last 30 days

Number of respondents

a) Percentages are based on respondents answering respective question. b) Data were collected by mailed surveys with a response rate of about 50%. c) Specified to at least a bottle of beer or 10cl of wine or 2,5 cl of spirits. Source: Skretting (2000, 2004).

55

Table I. Alcohol and drug use in Sweden. Frequency of lifetime and last 30 days use. Data from ESPAD and the annual Swedish school survey 2003 in grade 9. Percentages among boys and girls a). Boys ESPAD

Girls Annual school survey 2003

ESPAD

Annual school survey 2003

Lifetime Been drunk

62

56

62

60

Been drunk at the age of 13 or younger

25

19

19

18

Used any illicit drug

10

7

7

7

9

6

6

6

Used cannabis Used inhalants

8

8

8

6

Used anabolic steroids

1

1

0

0

Last 30 days Used cannabis

2

2

1

2

1,592

2,667

1,640

2,559

Number of respondents

a) Percentages are based on students answering respective question. Source: Hvitfeldt et al. (2004).

as a measure of reliability than of validity. Data from the two surveys are very similar for alcohol consumption and cannabis use during lifetime, last 12 months as well as last 30 days (Table J). This is also the case for cigarette smoking during the last 30 days. The slightly higher figures in the ESPAD study can be explained by a slightly larger number of boys in the ESPAD sample. In the 1995 ESPAD report comparisons between ESPAD data and data from national surveys were presented for England, Hungary, Iceland and Scotland. None of them showed any important differences (Hibell et al. 1997). The proportion of Finish ESPAD students that have ever used cannabis increased from 1995 to 1999 and was unchanged in 2003. A similar trend of an increase in the late 90’s and a levelling out in the beginning of this century has also been reported from 15–19 year old Finns in a nation wide survey (Hakkarainen and Metso, 2003). Many countries that participate in the ESPAD project are also involved in the HBSC study. Comparable information was available for alcohol consumption and drunkenness. Many countries in the HBSC study also asked questions on the use of cannabis. The latest round of data collection for the HBSC study was conducted in 2001–2002 with the goal to

56

produce mean ages of 11.5, 13.5 and 15.5 years. Comparisons with the ESPAD study is therefore limited to the oldest age group in the HBSC survey. Table 3 in Annex 1 of the HBSC report (Currie et al. 2004) shows that the mean ages in the oldest age group varied from 14.8 to 16.4 years while the corresponding range in ESPAD was 15.6–15.9. Since a difference of only a few months might indeed have an impact on the experiences with different substances, comparisons between the HBSC and ESPAD studies have been limited to countries in which the differences of the mean ages are not larger than ±0.2 years. There are some small differences between the two surveys in the way in which alcohol consumption and drunkenness have been measured. In ESPAD the figures for alcohol consumption show the proportion of boys and girls that had used alcohol 3 or more times during the last 30 days, while the HBSC survey measured the proportion that drank alcohol at least weekly. ESPAD data for drunkenness show the proportion that have “ever been drunk” while HBSC reports the proportion that has been “drunk” 2 or more times. Possible differences in the measures of lifetime and 12 months prevalence of cannabis use are less obvious between the two surveys. The relationship is rather strong on the alcohol

Methodological considerations

Table J. Alcohol and drug use in the Netherlands. Frequency of lifetime, last 12 months and last 30 days use. Data from ESPAD and a parallell school survey (PEIL). Percentages among all respondents a). ESPAD b)

PEIL c)

Lifetime Any alcohol

92

90

Cannabis

29

27

Last 12 months Any alcohol

88

86

Cannabis

23

22

Last 30 days Any alcohol

76

73

Cannabis

13

13

Smoke cigarettes

31

29

a) Percentages are based on respondents answering respective question. The questions were the same. However, in the PEIL study the answering categories were separate up to 10 (0, 1, 2 etc. till 10 times) while they were combined in ESPAD (1–2, 3–5, 6–9 times). b) Since there are no weight factors for the PEIL study for the selected birth cohort ESPAD figures are also unweighted, wich means that there in a few cases are minor differences compared with data in the result sections. c) The national sample of the PEIL study included students that were 10–18 years. However, for this comparison the selected age group is matced to the ESPAD target population. Source: Dorsselaer and Monshouwer (2004).

use variable, with rxy=0.91 for boys and 0.90 for girls and with Spearmans rank correlation (rrank) on 0.89 and 0.78 respectively (Table K). The rxy figures are about the same for drunkenness with 0.89 for boys and 0.90 for girls, while the rrank values are

Methodological considerations

a little higher with 0.93 and 0.96 (Table L). The cannabis variables also show a high correlation between the ESPAD and HBSC surveys. For lifetime use of cannabis the rxy was 0.96 and rrank 0.93 for boys as well as for girls (Table M). The rxy values are more or less equivalent for both sexes (0.94 for boys and 0.95 for girls) on the 12 months prevalence figures for cannabis, while rrank was a little higher for girls (0.94) than for boys (0.85) (Table N). Overall, the comparisons between ESPAD data in Norway, Sweden and the Netherlands and results from other surveys in these three countries, as well as comparisons between the ESPAD and HBSC surveys, show very similar figures. The same conclusions were also drawn from earlier studies in England, Hungary, Iceland and Scotland. Even if ESPAD data are “validated” by data from other studies, this really only applies to the countries involved and says nothing of the remaining ESPAD countries. On the other hand, it does not seem unrealistic to expect the situation to be rather equivalent in similar countries, i.e. mainly countries from the western part of Europe (since six of the seven countries included in the individual country comparisons were from this part of Europe as well as nine of the thirteen countries in the ESPAD – HBSC comparison). It is more difficult to form an opinion on the countries of central and eastern Europe, even if the comparisons between the two 1995 Hungarian studies indicated very similar results and the com parisons between the ESPAD and HBSC studies included four countries from these parts of Europe.

57

Table K. Alcohol use in the ESPAD and HBSC surveys. Students answering 3 times or more often during the last 30 days (ESPAD) or at least weekly (HBSC). Percentages among boys and girls a), rxy and Spearmans rangcorrelation cofficient (rrank). Country

Boys

Girls

ESPAD

HBSC

ESPAD

HBSC

3+ times last 30 days

1+ times a week

3+ times last 30 days

1+ times a week

Netherlands

62

56

49

47

Malta

60

56

48

40

Denmark

59

50

50

44

Italy

48

48

30

28

Switzerland

47

39

37

28

Poland

43

29

29

10

Slovenia

35

42

24

26

Portugal

34

21

19

11

Ukraine

31

29

24

19

Hungary

30

34

21

19

Norway

22

20

22

19

Finland

21

18

23

16

Sweden

20

23

16

17

rxy=0.91

rxy=0.90

rrank=0.89

rrank=0.78

a) Percentages are based on students answering respective question. Source: Currie et al. (2004).

Table L. Drunkenness in the ESPAD and HBSC surveys. Students who have ever been drunk (ESPAD) and drunk at least twice (HBSC). Percentages among boys and girls a), rxy and Spearmans rangcorrelation cofficient (rrank). Country

Boys

Girls

ESPAD

HBSC

ESPAD

HBSC

Ever been drunk

Drunk 2+ times

Ever been drunk

Drunk 2+ times

Denmark

87

68

84

65

Ukraine

80

61

75

45

Slovenia

74

44

65

38

Finland

68

53

70

56

Poland

67

40

51

23

Hungary

65

47

56

26

Switzerland

64

39

53

27

Sweden

62

40

62

38

Netherlands

60

35

50

22

Norway

55

39

62

41

Italy

53

23

49

17

Malta

52

25

44

18

Portugal

36

26

29

19

rxy=0.89

rxy=0.90

rrank=0.93

rrank=0.96

a) Percentages are based on students answering respective question. Source: Currie et al. (2004).

58

Methodological considerations

Table M. Lifetime use of cannabis in the ESPAD and HBSC surveys. Percentages among boys and girls a), rxy and Spearmans rangcorrelation cofficient (rrank). Country

Boys

Girls

ESPAD

HBSC

ESPAD

HBSC

Switzerland

44

49

36

40

Netherlands

32

29

24

23

Italy

31

27

23

18

Slovenia

31

31

26

25

Ukraine

29

33

12

15

Denmark

27

26

18

21

Poland

23

25

13

12

Hungary

18

17

13

11

Portugal

18

25

12

15

Malta

13

09

08

04

Finland

11

11

11

10

Sweden

09

08

06

07

rxy=0.96

rxy=0.96

rrank=0.93

rrank=0.93

a) Percentages are based on students answering respective question. Source: Currie et al. (2004).

Table N. 12 months prevalence of cannabis use in the ESPAD and HBSC surveys. Percentages among boys and girls a), rxy and Spearmans rangcorrelation cofficient (r rank). Country

Boys

Girls

ESPAD

HBSC

ESPAD

HBSC

Switzerland

35

40

28

35

Netherlands

27

24

18

19

Italy

24

24

19

17

Slovenia

24

27

22

21

Denmark

21

24

13

19

Poland

19

21

09

09

Ukraine

18

21

06

08

Portugal

15

25

11

14

Hungary

13

15

09

10

Malta

10

08

07

04

Finland

07

08

08

07

Sweden

05

05

04

05

rxy=0.94

rxy=0.95

rrank=0.85

rrank=0.94

a) Percentages are based on students answering the respective question. Source: Currie et al. (2004).

Methodological considerations

59

Conclusions The methodological discussion on representativeness, reliability and validity is rather extensive. The most salient conclusions are listed below (they are not ranked in any order).

General conclusions • None of the countries experienced methodological problems that made it impossible to compare their data with the data of other countries. • The drug use figures are probably somewhat underestimated and underreporting appears to differ somewhat between countries. However, the relative ranking of high and low prevalence countries is not likely to be affected by differences in underreporting between countries. • Despite some differences in cultural context the validity of the ESPAD survey is assumed to be high in most ESPAD countries. • The report does not provide confidence intervals for individual figures. It is important to interpret differences in point estimates with caution. • Individual countries suffer from methodological problems that should be taken into account when analysing their figures. These problems are briefly reviewed below. • The magnitude of various kinds of drug use in different ESPAD countries probably reflects country differences quite well, especially between distinguished groups of countries with different experiences of drug use. • It is more important to concentrate on the magnitudes of the estimates than on single figures, both when analysing data in single countries as well as when interpreting trends and differences between countries. • Small discrepancies between countries should be considered carefully. They may not reflect valid differences.

Country-specific conclusions • In Austria there were rather many classes that did not participate, which indicate some uncertainty. Boys were slightly overrepresented, and thus data ought to have been weighted. • A large number of schools and classes in Belgium did not participate in the data collection. There were sufficient reasons to believe that this did not impact on representativity, but the high figure calls for some caution.

60

• The proportion enrolled in school of those born in 1987 was also low in Bulgaria (72%). Inconsistency rates were rather high for alcohol consumption, drunkenness and cannabis use, which call for some caution when interpreting the figures of these variables. • Relatively large proportions in Croatia answered that they were unwilling to report possible use of cannabis (12%) and heroin (15%), which points to some uncertainty. • The sample in the Czech Republic only “covered” about 68% of all students born in 1987, which mainly limits the representativeness to students in grade 1. • The sample in Cyprus only “covered” 74% of all students born in 1987, which mainly limits the representativeness to students in grades 1 and 2. • A large number of schools and classes in Denmark refused to participate. Even though no systematic differences were found between participating and refusing schools, one cannot exclude the risk that the study is not fully representative for Danish students. • The proportion of non-participating schools and classes is unknown in Greenland, which cause some concern since school drop-out rates was rather high in 1999. The proportions of inconsistent answers were rather high as well as the proportions of unanswered questions. Many students reported an unwillingness to admit drug use. Hence, some caution is recommended when comparing data from Greenland with those from other ESPAD countries. • Students in Greece were seven months younger in 2003 than in the 1999 data collection, which must be kept in mind when interpreting changes in the substance use figures from 1999 to 2003. • In Ireland a relatively small proportion of students born in 1987 were found in the only participating grade in the ESPAD study (67%). Consequently Irish data are mainly representative for students born in 1987 that attended grade 5. • Compared with other countries rather large proportions in Latvia reported that they were unwilling to report possible use of cannabis (12%) and heroin (13%). Rather many students gave inconsistent answers to questions on drunkenness. Hence, some caution is recommended when interpreting the figures of these variables.

Methodological considerations

• The participating grade in Malta only included 75% of all students born in 1987. Hence, data are mainly representative for students attending grade 5. The inconsistency figure for inhalants was rather high and relatively large proportions reported that they were unwilling to report possible use of cannabis (13%) and heroin (15%). Hence, some caution is recommended when interpreting the figures of these variables. • Rather many schools in the Netherlands refused to participate, which points to some uncertainty. • Rather many classes in Norway did not participate, which raises some uncertainty. The proportions of unanswered questions on illegal substances were higher in Norway (4–7%) than in nearly all other countries, which might indicate an underreporting to a slightly higher degree than in some other ESPAD countries. • Of all student born in 1987 in Romania only 79% were found in participating school categories and grades. Thus, data were mainly representative for students born in 1987 enrolled in grades 9 and 10 in regular high schools. Boys were underrepresented in the Romanian sample and data should have been weighted to correct for this. • Participating grades in the Slovak Republic only included a rather small proportion of all students born in 1987 (67%), which was smaller than that in 1999 when the coverage was 99%.

Methodological considerations









Thus, data from the Slovak Republic are mainly representative for students born in 1987 that were found in grades 1–4. Some caution is recommended when comparisons are made between data from 1999 and 2003. The proportion of the survey leaders in Slovenia that reported that “all” or “nearly all” students were interested in the study and worked seriously was rather low. However, there are no other indications that the reliability or validity should be lower than in other ESPAD countries. The proportion of the 1987 birth cohort enrolled in school was low in Turkey (60%). The inconsistency figures were high for cigarette smoking and the proportion of unanswered questions on alcohol consumption, drunkenness and the use of inhalants, cannabis and other illegal drugs were rather high, which calls for some caution when interpreting many of the substance use variables. Some reliability and validity measures for drunkenness and cannabis use in Ukraine call for some caution when interpreting the figures for those variables. A large proportion of sampled schools in the United Kingdom did not participate. No differences were found when participating and nonparticipating schools were compared. However, the high proportion calls for some caution.

61

Changes in the use of alcohol and other drugs 1995–2003

This chapter presents changes in the use of alcohol and other drugs between 1995–2003 that are best exemplified by diagrams and scatter plots. Changes between 1999 and 2003 as well as between 1995 and 1999 are also included for selected variables. The variables selected are the same as those used in the 1999 ESPAD report. However, not all countries participated in 1995 or 1999 and in some instances data for one of the years may be missing on a specific variable. In both cases missing data are marked by two dots (..) in the bar graphs. A zero (0) signifies that at least 1 but less than 0.5 % have given this answer, while a short line (–) means that no student has given that answer. A study that is based on a random sample from a specific population will always result in a point estimate within a certain confidence interval. This means that a small difference in proportions can be caused by random sampling fluctuations rather than true differences in the populations under study. The confidence intervals enable the researcher to establish whether a difference should be considered a true difference or not. For various reasons described elsewhere in this report, no confidence intervals have been calculated for the surveys included in this study (see the chapter “Methodical considerations”). Consequently, the comments in this section of the report are based on substantive differences and changes, while differences of only a few percentage points are disregarded. In order to maintain consistency between this and the 1999 report we have only highlighted changes of more than three percentages points. Thus, values for a specific variable for a specific country that are unchanged or only changed within the range of three percentage points are coloured in yellow in the diagrams. Figures that have increases more than three percentage points are marked in red and figures that have decreased by more than three percentages points are marked in green.

It should be pointed out however, that this is only to facilitate interpretation as a difference within the yellow section of the diagrams may very well be statistically significant. The comments on each diagram focus mainly on the pattern of changes and the grouping of countries that fall within this pattern. The actual levels (percentages) of involvement in the various behaviours that are shown are usually disregarded as these findings are discussed in more detail in the next chapter, where the results are presented for each country that participated in the data collection in 2003. The gender pattern is demonstrated in the bar graphs, but is not discussed in the text. However, the next chapter includes some comments about gender differences. When data from 2003 are compared to those from earlier data collections it should be observed that the Romanian figures from the 1999 data collection included in this report are in some cases not those as found in the ESPAD 99 report since it by mistake included answers from students not born in 1983. Hence, to rectify this anomaly in this report, the Romanian figures for 1999 are only based on students belonging to the target population. It should also be observed that the Slovenian figures for cigarette smoking during the last 30 days have been recalculated for 1995 as well as for 1999. Greek students were seven months younger in 2003 than in the 1999 data collection, which must be kept in mind when interpreting changes from 1999 to 2003. A smaller proportion (67%) of the target population in the Slovak Republic participated in 2003 compared to 1999 (99%), which have limited the possibilities to compare data from the two surveys. The same is true for Portugal where the proportion of the target population that was included in the sampling frame increased from 66% in 1995 to 83% in 1999 and to 99% in 2003.

Changes in the use of alcohol and other drugs 2003

63

Changes in cigarette smoking Lifetime use of cigarettes 40 times or more (Figures 1a–c) In many of the countries the proportion of students that smoked at least 40 cigarettes in their lifetime was about the same in 2003 as it was in 1999. However, when changes occurred it was more common that these were downward rather than upward. The highest prevalence of smoking cigarettes at least 40 times is in most cases found in the eastern parts of Europe including the Czech Republic, Estonia, Lithuania and Romania. However, the two countries at the very top are still the same as they were in 1999, Greenland and Faroe Islands, despite the fact that the proportion reporting this behaviour had decreased somewhat in Greenland. The prevalence rates are also almost unchanged for this variable in other countries. This would seem to suggest that in countries where the prevalence rates were quite high in 1999, they have remained so in 2003. Countries where an increase can be observed were mainly found in eastern parts of Europe (the Czech Republic, Estonia, Lithuania and Romania). However, the prevalence rates in Romania are still among the lowest. The proportions reporting lifetime use of cigarettes 40 times or more decreased in some of the ESPAD countries between 1999 and 2003. As mentioned above, this was apparent in the high prevalence country Greenland, but also in Denmark, Finland, Ireland and Norway, all of which were half way up the list in 1999. Decreases, however, were also observed in countries that reported rather low prevalence rates in 1999 (Greece, Iceland, Malta and United Kingdom). When looking at the trend development for this particular behaviour between 1995 and 2003, only Lithuania has a clear upward tendency in this measure of lifetime use, while no country shows a continuous decrease over the years.

ESPAD countries by about 15 percentage points. It was, however, even higher in 1999 and thus they have somewhat decreased in 2003. In many of the top countries the prevalence rates were relatively unchanged between 1999 and 2003. Despite a decrease in Bulgaria between the two surveys, this does not alter the fact that the country is still the second highest on this variable followed by Russia (Moscow) and the Czech Republic. Some countries with relatively high prevalence rates in 1999 have lower figures for 2003, including Denmark, Finland, France, Ireland and Norway. However, this also occurred in countries with somewhat lower prevalence rates such as Greece, Iceland, Malta, Sweden and the United Kingdom. An increase in the prevalence rate for the 30 days smoking was observed in Cyprus, Estonia and Romania, although these countries’ position in the prevalence hierarchy are different – Cyprus and Romania are among the countries with the lowest prevalence rates, while Estonia is somewhere in the middle with respect to all ESPAD countries. Changes in 30 days smoking over the eight years in the countries that have conducted all three ESPAD studies show that very few of them have any continuous trends. However, the Estonian students reported increases from 1995 to 1999 to 2003, while students’ responses in Iceland and Ireland were indicative of a unidirectional decrease between the three surveys.

Daily smoking at the age of 13 or younger

(Figures 2a–c) Having smoked more than 40 times in a lifetime does not in itself refer to most recent habits. The last 30 days prevalence rates on the other hand, give an overall assessment of actual smoking habits. As in 1999 the top countries are still to be found in the eastern parts of Europe together with Greenland and the Faroe Islands. The prevalence rates are extremely high in Greenland and surpass other top

(Figures 3a–c) Many young people who experiment with smoking do so a few times but do not necessarily continue to smoke on a regular basis. Others, however, have already started daily smoking at an early age. Countries where smoking is highly prevalent also generally have a higher proportion of students that started to smoke at the age of 13. From 1999 to 2003 very small changes occurred on this variable in most of the countries. In Estonia, Faroe Islands and Latvia, however, a rather big increase was noted. A change in the opposite direction only occurred in two countries, Ireland and United Kingdom, where a rather big decrease was observed. This results in a change in these countries 1999 position in the prevalence hierarchy; they are replaced at the top of list by the Faroe Islands and Estonia in 2003. In many countries the prevalence rates for daily

64

Changes in the use of alcohol and other drugs 2003

Cigarette smoking during the last 30 days

smoking at the age of 13 have been rather stable over the three ESPAD data collections. No country

shows either a continuous increase or decrease between the three surveys.

Changes in alcohol consumption Alcohol use 40 times or more in lifetime (Figures 4a–c) The diagrams show that the prevalence rates on this variable were relatively unchanged in many ESPAD countries. However, in some of them the proportion of students who report this behaviour have noticeably increased. The twelve countries where this was observed include Bulgaria, Croatia, the Czech Republic, Estonia, the Faroe Islands, Hungary, Italy, Latvia, Lithuania, Russia (Moscow), the Slovak Republic and Ukraine. From the above list it would appear that the increases have predominantly occurred in the eastern parts of Europe, but also in the Faroe Islands and Italy. Changes in the opposite direction were only found in three countries, all of which were among the top countries in 1999: Denmark, Greece and the United Kingdom. Denmark and the United Kingdom still hold onto their top ranking despite the recent decrease, but Greece has fallen down the list. The trend development for this variable over the period 1995 to 2003 shows that in some of the countries there has been a unidirectional increase over the years. An upward trend can be observed in six countries, all of which are found in the eastern parts of Europe and include Croatia, the Czech Republic, Estonia, Lithuania, the Slovak Republic and Ukraine.

Alcohol use 20 times or more during the last 12 months (Figures 5a–c) Changes in the proportion of students who drank alcohol 20 times or more during the last 12 months are very similar to the lifetime prevalence of drinking 40 times or more. Thus, an increase was observed in a large number of countries, mainly in the eastern parts of Europe. The twelve countries where increasing proportions of students report such frequency of drinking include Bulgaria, Croatia, the Czech Republic, Estonia, the Faroe Islands, Hungary, Italy, Latvia, Lithuania, Russia (Moscow), the Slovak Republic and Ukraine. A decrease was found in Denmark, Greece and Ireland.

Changes in the use of alcohol and other drugs 2003

Over the years from 1995 to 2003 a continuous increasing number of students reported drinking 20 times or more in the last 12 months in the Czech Republic, Estonia, Lithuania, and the Slovak Republic. Others were relatively unchanged over the same time period but in no country was there evidence for a decrease in this trend.

Alcohol use 10 times or more during the last 30 days (Figures 6a–c) Among 15–16 year old students in Europe, an alcohol consumption frequency of 10 times or more over the last 30 days is relatively uncommon but the prevalence rates differ substantially. However, between 1999 and 2003 rather small changes were observed and in the main the situation is one of status quo. Nevertheless, changes occurred in a few of the countries, some of which were rather noteworthy. Thus, increased figures were observed in Bulgaria, Croatia, Italy, Latvia and Russia (Moscow). A decrease was reported by only one country, Denmark. From the rather stable situation between 1999 and 2003 it follows that the top countries remain, including Malta, the United Kingdom, Ireland and Denmark. Looking at the trends over the eight years no continuous changes were found, neither in a positive nor negative direction.

Beer consumption 3 times or more during the last 30 days (Figures 7a–c) The pattern of frequent beer consumption has changed in different directions among young people in Europe over the actual four years. Moreover, it is not simply a pattern of an increase in low prevalence countries and a decrease in high prevalence countries, but a mixture of both. However, the increases tend to be mainly found in the eastern parts of Europe. Increases in the proportions reporting that they had consumed beer three times or more during the last 30 days were found in Bulgaria, Croatia, the

65

2003 50 Gr

ia ss co os ) w

30

(M

Estonia Bulgaria Latvia, Slovak Rep. Finland Hungary Croatia Denmark Slovenia Ireland Poland Norway Italy U.K. Sweden Greece Iceland Uk

ne rai

Romania Cyprus

20

Po

Malta

l ga rtu

Figure 1a. Changes between 1999 and 2003 in lifetime use of cigarettes 40 times or more. Countries above the line have increased prevalence rates, and countries below have decreased. All students.

ee n la nd

Ru

Czech Rep. Lithuania

40

Faroe Isl.

10

0 0

10

20

30

40

50 1999

Boys

Girls Greenland (.., 50, 42)

41 34

39

Russia (Moscow) (.., 42, 40)

42 29 49

38 32

29 33 32

32 26

13

Slovak Rep. (20, 30, 32)

35 35

26 30

Latvia (.., 30, 32)

38 39 32 31 33

23 25 24 25

Hungary (28, 28, 31)

30 27

31 32

Croatia (23, 28, 30)

18

Ukraine (29, 29, 28)

18 18 19

25 29

41 39 38 36 25

29 22

24

Denmark (23, 32, 27)

31 26

32 28

16

17

Slovenia (16, 26, 27)

25 26 25

26 28 25

Norway (25, 33, 26)

31 23

34 29

27

13

Poland (20, 26, 26)

32 32 25

20 21 24

Italy (25, 25, 25)

22 25 25 24

28 25 30 28

U.K. (27, 26, 22)

19

24

28 26

28

Sweden (28, 25, 22)

20

25 24

Greece (.., 27, 20)

28 19

27

1995

21

Romania (.., 16, 20)

24 26 27 24

1999

10 15

2003

27 26

Iceland (27, 25, 18)

19

17

14 16 18

12

Portugal (13, 17, 18)

26 26 27

15 17 9 8

Cyprus (18, 16, 18)

11 20 20

20

18

Malta (19, 20, 16)

17

30

38 36

Ireland (37, 34, 27)

31

40

36 38

Finland (35, 39, 32)

41

2003

38 37 17 18

Estonia (25, 27, 35)

41

1999

34

Bulgaria (.., 36, 35)

36 38

1995

20

Czech Rep. (26, 36, 39)

35

66

23 28

30 39 39

50

38 38 12

Lithuania (20, 35, 39)

46

% 60

41 40 42

Faroe Isl. (42, 43, 41)

46

Figure 1b. Changes between 1995 and 2003 in lifetime use of cigarettes 40 times or more. Percentages among boys and girls (values within brackets refer to all students 1995, 1999, 2003). Data sorted by all students 2003.

55 49

42 47

21 16

10

0

0

10

20

30

40

50

60 %

Changes in the use of alcohol and other drugs 2003

Bulgaria

% 60

Finland

% 60

% 60

40

40

40

40

20

20

20

20

0

0 1995

1999

2003

Croatia

% 60

0 1995

40

20

Data not

20

0

2003

France

% 60

40

1999

0 1995

% 60

1999

2003

1999

2003

Latvia

1999

40

40

20

20

2003

1999

2003

Lithuania

% 60

40

40

40

20

20

20

20

0 1995

% 60

1999

2003

Czech rep.

0 1995

% 60

1999

2003

1999

2003

% 60

Malta

40

40

40

20

20

20

20

0 1995

1999

2003

Denmark

% 60

0 1995

1999

2003

Hungary

% 60

1999

2003

Norway

% 60

40

20

20

20

20

0

% 60

2003

0 1995

% 60

Estonia

1999

2003

1999

2003

1995

% 60

Poland

40

40

40

40

20

20

20

20

0

0 1995

1999

2003

Faroe Isl.

% 60

0 1995

% 60

1999

2003

1999

2003

Portugal

% 60

40

40

40

20

20

20

20

0 1995

1999

2003

0 1995

1999

2003

2003

Ukraine

1995

40

0

1999

0 1995

% 60

Ireland

2003

0 1995

% 60

Iceland

1999

Sweden

% 60

40

1999

2003

Slovenia

1995

40

1995

1999

0 1995

40

0

2003

Slovak rep.

1995

40

0

1999

0 1995

% 60

Greenland

2003

Russia (Moscow)

1995

40

0

1999

0 1995

% 60

Greece

1995

% 60

0 1995

% 60

Cyprus

0 1995

% 60

avail able

Romania

% 60

Italy

1999

2003

United Kingdom

0 1995

1999

2003

1995

1999

2003

Figure 1c. Changes between 1995 and 2003 in lifetime use of cigarettes 40 times or more, by country. All students.

Changes in the use of alcohol and other drugs 2003

67

2003 70

60

ly p. Ita Re h a ec ni ia ne Cz ua tv ai th LaUkr ry Li ga un H

50

Slo va k Estonia Rep . Slovenia

40

Poland Portu gal Romania

30 Cyprus

Sweden

20

Bulgaria Russia (Moscow) Faroe Isl. Finland

C Ireroat France lan ia d Norway

Malta U.K. Greece Denmark

Figure 2a. Changes between 1999 and 2003 in cigarette smoking during the last 30 days. Countries above the line have increased prevalence rates, and countries below have decreased. All students.

Greenland

Iceland

10

0 0

10

20

30

40

50

60

70 1999

Boys

Girls Greenland (.., 67, 60)

62 56

Bulgaria (.., 50, 46)

42 44 37 43 40 42 42

18

Lithuania (25, 40, 41)

30 33

Latvia (.., 40, 40)

48 46 51 50 49

34 36 28 29 28

Ukraine (38, 40, 39) 36 37 39 36 35 36 37 35 34

Italy (36, 40, 38)

34 36 22 24 33 28

Croatia (32, 38, 36)

36 26

36 37 27

Slovenia (26, 36, 36)

36 35

38 38

France (.., 44, 33)

41 31

47 36

37

45 42

Ireland (41, 37, 33)

32 28

37

34

23

Poland (28, 33, 31)

39 35 24

28 27 32

Denmark (28, 38, 30)

34 27

41 32

Greece (.., 35, 29)

27

36 30

32 31

40

U.K. (36, 34, 29)

25

37 34

33

39

Norway (36, 40, 28)

36 24 22

44 32 25

Portugal (24, 31, 28)

32 28

1995

30 27

Romania (.., 24, 28)

31 32 33 29 28 32

1999

20 26

2003

30

Malta (31, 32, 27)

34 26 15

Cyprus (23, 16, 25)

25 33

9 18

28 29 30

26 33

Iceland (32, 28, 20)

26 20

30

33 32

Sweden (30, 30, 23)

20

45

20

Estonia (28, 32, 37) 34

40

2003

40 39 43 41 37 43 40

Slovak Rep. (27, 37, 37)

40 39 37 41 40

1999

35

Finland (37, 43, 38)

44

1995

32

Hungary (34, 36, 39)

34

68

43 43 43 41 41

Faroe Isl. (42, 41, 41) 34

49 49

60

42 44 31

Czech Rep. (34, 44, 43)

46

Figure 2b. Changes between 1995 and 2003 in cigarette smoking during the last 30 days. Percentages among boys and girls (values within brackets refer to all students 1995, 1999, 2003). Data sorted by all students 2003.

51 50

Russia (Moscow) (.., 45, 44)

48

% 75

71 65

48

30 20

15

0

0

15

30

45

60

75 %

Changes in the use of alcohol and other drugs 2003

Bulgaria

% 80

Finland

% 80

% 80

60

60

60

60

40

40

40

40

20

20

20

20

0

0

0

1995

1999

2003

Croatia

% 80

1995

1999

2003

France

% 80

0 1995

1999

2003

Latvia

% 80

1995

% 80

60

60

60

60

40

40

40

40

20

20

20

20

0

0

0

1995

% 80

1999

2003

1995

% 80

Cyprus

60

60

40

40

20

20

0

0 1995

% 80

1999

2003

Czech rep.

1999

2003

1995

% 80

1999

2003

1999

2003

Lithuania

% 80

1999

2003

1995

% 80

Malta

60

60

40

40

20

20

20

20

0

0

0

Denmark

% 80

1999

2003

Hungary

% 80

1999

2003

Norway

% 80 60

60

40

40

40

20

20

20

20

0

0

0

% 80

1995

% 80

Estonia

1999

2003

1999

2003

1995

% 80

Poland

60

60

60

60

40

40

40

40

20

20

20

20

0

0

0

1995

1999

2003

Faroe Isl.

% 80

1995

% 80

1999

2003

1999

2003

Portugal

% 80

60

60

60

40

40

40

40

20

20

20

20

0

0

0

1999

2003

1995

1999

2003

2003

Ukraine

1995

60

1995

1999

0 1995

% 80

Ireland

2003

0 1995

% 80

Iceland

1999

Sweden

% 80

60

2003

2003

Slovenia

1995

40

1999

1999

0 1995

60

1995

Slovak rep.

0 1995

40

1995

2003

20

60

2003

1999

40

40

1999

Russia (Moscow)

1995

60

1995

2003

60

% 80

Greenland

1999

0 1995

% 50 40 30 20 10 0

Greece

Romania

% 80

Italy

1999

2003

United Kingdom

0 1995

1999

2003

1995

1999

2003

Figure 2c. Changes between 1995 and 2003 in cigarette smoking during the last 30 days, by country. All students.

Changes in the use of alcohol and other drugs 2003

69

2003 30

25 Cz

Faroe Isl.

h ec

20

d an nl Fi land en re G

ia an hu Lit p., Re

Estonia 15

Russia (Moscow) Ireland U.K.

Latvia

Slovak Rep. Ukraine Bulgaria Poland

10

Slovenia Cyprus 5 Romania

Portugal

Figure 3a. Changes between 1999 and 2003 in daily smoking at the age of 13 or younger. Countries above the line have increased prevalence rates, and countries below have decreased. All students.

Italy

Denmark Croatia, Norway Sweden Iceland Malta Hungary

Greece 0 0

5

10

15

20

25

30 1999

Boys

Girls

21 21

13

Russia (Moscow) (.., 16, 15)

18 18 18

16

Finland (17, 15, 15)

15

14 15

Greenland (.., 14, 15)

9 9

20 21

20

16

Ireland (18, 18, 14)

17 12

19 16

Latvia (.., 9, 14)

13 19

6 10

15

22

U.K. (19, 20, 13)

16 9

24 18

10

6

Czech Rep. (8, 11, 13)

12 14

9 11

13

3

Lithuania (8, 11, 13)

17 19

6 7

11

4

Slovak Rep. (7, 10, 13)

12 15

7 11

9 11

7

Croatia (11, 11, 11)

14 13

8 9

9

11

Norway (10, 11, 11)

11 10

10 12

14

4

Ukraine (9, 10, 11)

15 16

5 5

Bulgaria (.., 9, 10)

11 10

8 11

11

12

Sweden (12, 10, 9)

9 8

10 11

9 9

8 8

Portugal (8, 8, 9)

8

10

9

3 3

Poland (6, 6, 9)

10 13

5 11

1995

9 8 8

Malta (8, 10, 7)

5 5 5

10 8 4

Slovenia (5, 5, 7)

7

5

1995

7

9

5

Hungary (7, 9, 6)

11 7

1999

8 5

5 5

2003

2003

4

Italy (5, 6, 6)

6

7 6

5 8 10

Cyprus (3, 5, 6)

2

Romania (.., 4, 5)

2

3 4

7 8

3

Greece (.., 3, 4)

5 4

10

12

Iceland (12, 9, 8)

9 9 9 9

15

12 13

15

1999

10

Denmark (9, 12, 12)

12

20

15 13

17

70

4 4

Estonia (9, 8, 17)

12 21

% 25

10 20

15

Figure 3b. Changes between 1995 and 2003 in daily smoking at the age of 13 or younger. Percentages among boys and girls (values within brackets refer to all students 1995, 1999, 2003). Data sorted by all students 2003.

18

Faroe Isl. (19, 13, 20)

16

5

3 4

0

0

5

10

15

20

25 %

Changes in the use of alcohol and other drugs 2003

Bulgaria

% 30

Finland

% 30

% 30

20

20

20

20

10

10

10

10

0

0

0

1995

1999

2003

Croatia

1995

1999

2003

0 1995

1999

2003

France

% 30

20

20

20

20

10

10

Data not a vaila ble

10

10

0

0 1995

% 30

1999

2003

% 30

0 1995

% 30

Cyprus

Latvia

1995

% 30

% 30

1999

2003

1999

2003

Lithuania

% 30

20

20

20

10

10

10

10

0 1995

% 30

1999

2003

Czech rep.

0 1995

% 30

1999

2003

1999

2003

% 30

Malta

20

20

20

10

10

10

10

0 1995

1999

2003

Denmark

% 30

0 1995

1999

2003

Hungary

% 30

1999

2003

Norway

% 30

20

10

10

10

10

0

% 30

2003

0 1995

% 30

Estonia

1999

2003

1999

2003

1995

% 30

Poland

20

20

20

20

10

10

10

10

0

0 1995

1999

2003

Faroe Isl.

% 30

0 1995

% 30

1999

2003

1999

2003

Portugal

% 30

20

20

20

10

10

10

10

0 1995

1999

2003

0 1995

1999

2003

2003

Ukraine

1995

20

0

1999

0 1995

% 30

Ireland

2003

0 1995

% 30

Iceland

1999

Sweden

% 30

20

1999

2003

Slovenia

1995

20

1995

1999

0 1995

20

0

2003

Slovak rep.

1995

20

0

1999

0 1995

% 30

Greenland

2003

Russia (Moscow)

1995

20

0

1999

0 1995

% 30

Greece

Romania

% 30

Italy

1999

2003

United Kingdom

0 1995

1999

2003

1995

1999

2003

Figure 3c. Changes between 1995 and 2003 in daily smoking at the age of 13, by country. All students.

Changes in the use of alcohol and other drugs 2003

71

2003 60

50

Denmark Czech Rep. U.K.

Russia (Moscow) 40

Ireland

Lithuania Slovak Rep. Estonia Faroe Isl. Croatia Latvia Bulgaria Poland Italy Slovenia Hungary ine Ukra ia n a m Ro Iceland Greenland

30

Greece Malta

s ce ru d n an yp lan de Fr C Fin we S ay l rw uga No ort P

Figure 4a. Changes between 1999 and 2003 in lifetime use of any alcoholic beverages 40 times or more. Countries above the line have increased prevalence rates, and countries below have decreased. All students.

20

10

0 0

10

20

30

40

50

60 1999

Boys 66

Girls

55

54

45 47 37

40 39 43 39

31

Ireland (34, 40, 39)

41 42

39 36

Russia (Moscow) (.., 30, 39)

34 44

26 34

14

10

Lithuania (12, 23, 38)

29 45

17 31

Greece (.., 42, 35)

54 43 42 39

23 28 29 29 27

Malta (34, 36, 33)

41 28 29

23

Faroe Isl. (26, 23, 32)

34

17 30

17

27

10

Estonia (13, 21, 32)

38

17 26

25

12

Poland (18, 26, 27)

35 36 21

18 18 6

Croatia (14, 18, 27)

24 38

10 16

Bulgaria (.., 16, 27)

21 33

12 21

Latvia (.., 20, 26)

24 30

17 23

19

9

Slovenia (14, 23, 25)

29 32 33

16 18 15 13 16

Italy (26, 17, 24)

23 33

France (.., 20, 22)

28 30 16 18

13 15 13

Ukraine (14, 18, 22)

24 44

18 19 21

Cyprus (32, 21, 21)

32 34 20 16 21 20

Finland (16, 20, 20)

19 23 21

Sweden (16, 19, 17)

2003

10 22 21 20

1995

12 12 13 15 14

1999 2003

7

Norway (8, 16, 15)

18 17

13 14 10 10

Portugal (15, 15, 14) 8 14 15 16

13 14 12

Iceland (14, 14, 14) Greenland (.., 14, 13)

13 17

30

19 20

Romania (.., 18, 18)

27 26

1999

12 12 10 9 14 16

Hungary (15, 13, 21)

17 27

45

13

Slovak Rep. (19, 27, 34)

44

1995

33 28

24 31

72

32

U.K. (42, 47, 43)

51

Figure 4b. Changes between 1995 and 2003 in lifetime use of any alcoholic beverages 40 times or more. Percentages among boys and girls (values within brackets refer to all students 1995, 1999, 2003). Data sorted by all students 2003.

25

Czech Rep. (32, 41, 46)

51

60

53

42 38

% 75

44

Denmark (49, 59, 50)

57

15 9

15

0

0

15

30

45

60

75 %

Changes in the use of alcohol and other drugs 2003

Bulgaria

% 60

Finland

% 60

% 60

40

40

40

40

20

20

20

20

0

0

0

1995

1999

2003

Croatia

% 60

1995

1999

2003

France

% 60

0 1995

1999

2003

Latvia

% 60

1995

% 60

40

40

40

40

20

20

20

20

0

0 1995

% 60

1999

2003

0 1995

% 60

Cyprus

1999

2003

1999

2003

Lithuania

% 60

40

40

40

20

20

20

20

0 1995

% 60

1999

2003

Czech rep.

0 1995

% 60

1999

2003

1999

2003

% 60

Malta

40

40

40

20

20

20

20

0 1995

1999

2003

Denmark

% 60

0 1995

1999

2003

Hungary

% 60

1999

2003

Norway

% 60

40

20

20

20

20

0

% 60

2003

0 1995

% 60

Estonia

1999

2003

1999

2003

1995

% 60

Poland

40

40

40

40

20

20

20

20

0

0 1995

1999

2003

Faroe Isl.

% 60

0 1995

% 60

1999

2003

40

40

40

20

20

20

0

0

0

1999

2003

Ukraine

0 1995

1999

2003

Portugal

% 60

Ireland

2003

0 1995

% 60

Iceland

1999

Sweden

% 60

40

1999

2003

Slovenia

1995

40

1995

1999

0 1995

40

0

2003

Slovak rep.

1995

40

0

1999

0 1995

% 60

Greenland

2003

Russia (Moscow)

1995

40

0

1999

0 1995

% 60

Greece

Romania

% 60

Italy

1995

% 60

1999

2003

United Kingdom

45 30

1995

1999

2003

1995

1999

2003

15 0 1995

1999

2003

1995

1999

2003

Figure 4c. Changes between 1995 and 2003 in lifetime use of any alcoholic beverages 40 times or more, by country. All students.

Changes in the use of alcohol and other drugs 2003

73

2003 60

50

Denmark 40 Ireland

Czech Rep. 30

20

Greece Russia (Moscow) Slovak Rep. Slovenia Poland Latv Cyprus ia Finland

Bulgaria

Hungary Ukraine France

10

MaltaU.K.

Faroe Isl.

a ni ua ia ay th n a w Li sto ati t aly or E ro I N C

Figure 5a. Changes between 1999 and 2003 in use of any alcoholic beverages 20 times or more during the last 12 months. Countries above the line have increased prevalence rates, and countries below have decreased. All students.

Sweden

Iceland

Greenland, Portugal, Romania 0 0

10

20

30

40

50

60 1999

Boys

Girls

45

60

Denmark (42, 51, 42)

48 34 39

31 34

41 38 29 42

39 39

20 22

32

26 23

17

19

Faroe Isl. (18, 18, 27)

23 26

14 26

Russia (Moscow) (.., 21, 26)

24 31

18 23

8

5

Lithuania (7, 16, 23)

20 29

11 17

14 27 12

17

14 16 7

Estonia (9, 14, 21)

24

11 19

13

5

Croatia (10, 13, 21)

18 28 25 17 27

7 13 13

Italy (20, 12, 19)

9

Bulgaria (.., 11, 19)

9

12 14

26

14 16

7

Poland (11, 19, 18)

26 24 13

12 13 6

Slovenia (9, 17, 18)

21 23 35

12 12 13

Cyprus (24, 16, 17)

24 26 19 12 20

Hungary (12, 9, 15)

7 6

Ukraine (9, 10, 14)

7

10 10 10 16 10 13 9

11 12

France (.., 12, 12)

1995

7 8

1999

6

Portugal (10, 12, 11)

8 5 15 11

Romania (.., 12, 11)

21 18 12

6 7 11 11

Sweden (11, 13, 10)

15 11 11 11 10

10

2003

Greenland (.., 12, 11)

10 10

20

14 15 13 5

Norway (7, 13, 12)

15 13 17 17 16 17 16

2003

10 11

Finland (13, 16, 13)

17

1999

11 15

16

1995

9 9

Latvia (.., 13, 17)

15

30

7

Slovak Rep. (11, 17, 22)

21

40

27 26

Greece (.., 32, 27)

41

Figure 5b. Changes between 1995 and 2003 in use of any alcoholic beverages 20 times or more during the last 12 months. Percentages among boys and girls (values within brackets refer to all students 1995, 1999, 2003). Data sorted by all students 2003.

23

Malta (27, 32, 32)

38 39

74

U.K. (32, 36, 34)

30 30 31

39

28 33

50

Ireland (32, 39, 35)

Czech Rep. (24, 30, 34)

39

% 60

34 36 30

9 12 11

Iceland (11, 10, 9) 8

0

0

10

20

30

40

50

60 %

Changes in the use of alcohol and other drugs 2003

Bulgaria

% 60

Finland

% 60

% 60

40

40

40

40

20

20

20

20

0

0

0

1995

1999

2003

Croatia

% 60

1995

1999

2003

France

% 60

0 1995

1999

2003

Latvia

% 60

1995

% 60

40

40

40

40

20

20

20

20

0

0 1995

% 60

1999

2003

0 1995

% 60

Cyprus

1999

2003

1999

2003

Lithuania

% 60

40

40

40

20

20

20

20

0 1995

% 60

1999

2003

Czech rep.

0 1995

% 60

1999

2003

1999

2003

% 60

Malta

40

40

40

20

20

20

20

0 1995

1999

2003

Denmark

% 60

0 1995

1999

2003

Hungary

% 60

1999

2003

Norway

% 60

40

20

20

20

20

0

% 60

2003

0 1995

% 60

Estonia

1999

2003

1999

2003

1995

% 60

Poland

40

40

40

40

20

20

20

20

0

0 1995

1999

2003

Faroe Isl.

% 60

0 1995

% 60

1999

2003

1999

2003

Portugal

% 60

40

40

40

20

20

20

20

0 1995

1999

2003

0 1995

1999

2003

2003

Ukraine

1995

40

0

1999

0 1995

% 60

Ireland

2003

0 1995

% 60

Iceland

1999

Sweden

% 60

40

1999

2003

Slovenia

1995

40

1995

1999

0 1995

40

0

2003

Slovak rep.

1995

40

0

1999

0 1995

% 60

Greenland

2003

Russia (Moscow)

1995

40

0

1999

0 1995

% 60

Greece

Romania

% 60

Italy

1999

2003

United Kingdom

0 1995

1999

2003

1995

1999

2003

Figure 5c. Changes between 1995 and 2003 in use of any alcoholic beverages 20 times or more during the last 12 months, by country. All students.

Changes in the use of alcohol and other drugs 2003

75

2003 30

25

20

Malta

a si us R

U.K. Ireland

) w co os (M

15

Croatia

Denmark

10

Cyprus Poland

Bulgaria

G e ec re

Lithuania France, Slovenia P Latvia Hu ortu 5 Romania Uk ngar gal Faroe Isl. rai y ne Finland Gree nland , Nor way Sweden 0 Iceland 0 5 10 15 Estonia

, p. Re

Italy Slov ak R ep.

h ec Cz

Figure 6a. Changes between 1999 and 2003 in use of any alcoholic beverages 10 times or more during the last 30 days. Countries above the line have increased prevalence rates, and countries below have decreased. All students.

20

25

30 1999

Boys

Girls

20

25 25

18

14 17

16 10

Denmark (15, 18, 13)

18

13 10

12

5

Czech Rep. (9, 14, 13)

21 17

8 10

Greece (.., 14, 13)

19 18

10 9

7 15

3 11

Russia (Moscow) (.., 8, 12)

11 16

5 10

18 12 17 19 14

Figure 6b. Changes between 1995 and 2003 in use of any alcoholic beverages 10 times or more during the last 30 days. Percentages among boys and girls (values within brackets refer to all students 1995, 1999, 2003). Data sorted by all students 2003.

1

Croatia (6, 6, 13)

9

16

5

Italy (13, 7, 12)

4

Cyprus (12, 8, 11)

4

6 6 8

6

12

5 6

Bulgaria (.., 5, 9)

6 13

4 7

6 9 12

Slovak Rep. (4, 7, 9)

1

Lithuania (2, 8, 8)

1

5 6

3 9 13

6 5

6

2

Slovenia (5, 8, 7)

10 10

5 4

France (.., 8, 7)

12 10 8

5 5 2

Portugal (5, 7, 7)

10 11

4 3

6 6 8

Estonia (2, 4, 6)

1

2 3 5 2 4

4

3

Ukraine (3, 5, 5)

5

4 4

Romania (.., 4, 5)

7 9 4

1 3 3

Faroe Isl. (4, 4, 4)

3 6

1995

2 4

1

1999 2003

2

Greenland (.., 3, 3)

4 5

2003 3 4

1 2 3

Finland (1, 1, 2)

1 1

Sweden (1, 2, 1)

1 1 1 1 1 1

2 1 2 2 2

Iceland (1, 1, 1)

1

2

5

1999

1 1

Norway (1, 3, 3)

3 3

10

1

Latvia (.., 2, 6)

4 7

1995

Hungary (4, 5, 6)

4 3 5

8

15

2

Poland (4, 8, 10)

13

6

76

15

9 14

19 23

20

13

Ireland (12, 16, 16)

18

16 16

11

U.K. (13, 16, 17)

17

% 25

12

Malta (16, 20, 20) 16

0

0

5

10

15

20

25 %

Changes in the use of alcohol and other drugs 2003

Bulgaria

% 60

Finland

% 60

% 60

40

40

40

40

20

20

20

20

0

0

0

1995

1999

2003

Croatia

% 60

1995

1999

2003

France

% 60

0 1995

1999

2003

Latvia

% 60

1995

% 60

40

40

40

40

20

20

20

20

0

0 1995

% 60

1999

2003

0 1995

% 60

Cyprus

1999

2003

1999

2003

Lithuania

% 60

40

40

40

20

20

20

20

0 1995

% 60

1999

2003

Czech rep.

0 1995

% 60

1999

2003

1999

2003

% 60

Malta

40

40

40

20

20

20

20

0 1995

1999

2003

Denmark

% 60

0 1995

1999

2003

Hungary

% 60

1999

2003

Norway

% 60

40

20

20

20

20

0

% 60

2003

0 1995

% 60

Estonia

1999

2003

1999

2003

1995

% 60

Poland

40

40

40

40

20

20

20

20

0

0 1995

1999

2003

Faroe Isl.

% 60

0 1995

% 60

1999

2003

1999

2003

Portugal

% 60

40

40

40

20

20

20

20

0 1995

1999

2003

0 1995

1999

2003

2003

Ukraine

1995

40

0

1999

0 1995

% 60

Ireland

2003

0 1995

% 60

Iceland

1999

Sweden

% 60

40

1999

2003

Slovenia

1995

40

1995

1999

0 1995

40

0

2003

Slovak rep.

1995

40

0

1999

0 1995

% 60

Greenland

2003

Russia (Moscow)

1995

40

0

1999

0 1995

% 60

Greece

Romania

% 60

Italy

1999

2003

United Kingdom

0 1995

1999

2003

1995

1999

2003

Figure 6c. Changes between 1995 and 2003 in use of any alcoholic beverages 10 times or more during the last 30 days, by country. All students.

Changes in the use of alcohol and other drugs 2003

77

2003 60

50

Faroe Isl.

40 Ukraine Romania

30

Bulgaria Poland

Latvia

Croatia Estonia Slovak Rep. Iceland

20 Hungary 10

Denmark Czech Rep. Russia (Moscow)

Ireland Italy

ta al a ni a M ni ua th us love Li pr ce S Cy an Fr en ed al Sw tug d r y Po nlan rwa Fi No

Figure 7a. Changes between 1999 and 2003 in beer consumption 3 times or more during the last 30 days. Countries above the line have increased prevalence rates, and countries below have decreased. All students.

U.K.

Greenland

Greece

0 0

10

20

30

40

50

60 1999

Boys

Girls

49

64

Bulgaria (.., 27, 43)

38 36

19 15

Czech Rep. (31, 40, 39)

52

25 28

Russia (Moscow) (.., 40, 38)

48 47 42 42 36

21 22 22

Italy (31, 31, 34)

45 45 16

5

Ukraine (12, 22, 34)

29 45

16 21

Greenland (.., 45, 32)

46 33

43 32

Latvia (.., 30, 32)

41 42

18 22

Romania (.., 20, 32)

30 46

12 22

41

19

U.K. (30, 37, 31)

47 42 23

26 17 19 19 21

Faroe Isl. (20, 25, 31)

33 38 43 45 45

14

Malta (28, 31, 29)

20 16

14

3

Lithuania (9, 28, 28)

36 38 19

19 20 5

Croatia (13, 23, 28)

31 42

11 14

Greece (.., 35, 28)

48 38 48

26 17 19 18 16

Cyprus (34, 27, 27)

41 40 21 38 35 24 31 32 29

26 23

6

14 16 11 11

19 20 19

1999

16 16 16 15 14

17

9

20

9 8

Norway (9, 17, 14)

16

20

5 5

Hungary (13, 12, 17)

23

2003

12 12 10 12 11 13

Finland (15, 17, 17)

19 18

1999

16 17

Portugal (16, 19, 18) 22 22

1995

5

Iceland (17, 17, 19)

25 27 27

30

14 12

Sweden (19, 21, 20)

28 27

40

Slovak Rep. (16, 21, 22)

France (.., 25, 20)

32

2003

6

14

31

1995

Estonia (12, 25, 25)

Slovenia (21, 27, 21)

36

50

31 30 27 29 25

Ireland (34, 35, 36)

47

78

33

14 32

41 59

60

19

Poland (25, 28, 41)

39 50

% 70

42

35

55

Figure 7b. Changes between 1995 and 2003 in beer consumption 3 times or more during the last 30 days. Percentages among boys and girls (values within brackets refer to all students 1995, 1999, 2003). Data sorted by all students 2003.

39

Denmark (44, 53, 44)

54

15 11

10

0

0

10

20

30

40

50

60

70 %

Changes in the use of alcohol and other drugs 2003

Bulgaria

% 60

Finland

% 60

% 60

40

40

40

40

20

20

20

20

0

0

0

1995

1999

2003

Croatia

% 60

1995

1999

2003

France

% 60

0 1995

1999

2003

Latvia

% 60

1995

% 60

40

40

40

40

20

20

20

20

0

0 1995

% 60

1999

2003

0 1995

% 60

Cyprus

1999

2003

1999

2003

Lithuania

% 60

40

40

40

20

20

20

20

0 1995

% 60

1999

2003

Czech rep.

0 1995

% 60

1999

2003

1999

2003

% 60

Malta

40

40

40

20

20

20

20

0 1995

1999

2003

Denmark

% 60

0 1995

1999

2003

Hungary

% 60

1999

2003

Norway

% 60

40

20

20

20

20

0

% 60

2003

0 1995

% 60

Estonia

1999

2003

1999

2003

1995

% 60

Poland

40

40

40

40

20

20

20

20

0

0 1995

1999

2003

Faroe Isl.

% 60

0 1995

% 60

1999

2003

1999

2003

Portugal

% 60

40

40

40

20

20

20

20

0 1995

1999

2003

0 1995

1999

2003

2003

Ukraine

1995

40

0

1999

0 1995

% 60

Ireland

2003

0 1995

% 60

Iceland

1999

Sweden

% 60

40

1999

2003

Slovenia

1995

40

1995

1999

0 1995

40

0

2003

Slovak rep.

1995

40

0

1999

0 1995

% 60

Greenland

2003

Russia (Moscow)

1995

40

0

1999

0 1995

% 60

Greece

Romania

% 60

Italy

1999

2003

United Kingdom

0 1995

1999

2003

1995

1999

2003

Figure 7c. Changes between 1995 and 2003 in beer consumption 3 times or more during the last 30 days, by country. All students.

Changes in the use of alcohol and other drugs 2003

79

Faroe Islands, Hungary, Poland, Romania, the Slovak Republic and Ukraine. Decreases were not only observed in the top two countries in 1999 (Denmark and Greenland) but also in France, Greece, Slovenia and the United Kingdom. Despite the decrease Denmark still ranks highest in this regard in 2003. Other countries that have joined this group after rather pertinent increases in the prevalence figures include Bulgaria, Poland and the Slovak Republic. Over the years 1995 to 2003 an increase in the prevalence rates of having consumed beer 3 times or more in the last 30 days were found in Croatia, the Faroe Islands, the Slovak Republic and Ukraine. No country showed a continuous decreasing trend over the last eight years.

Wine consumption 3 times or more during the last 30 days (Figures 8a–c) The proportions of students who reported wine consumption as frequent as 3 times or more during the last 30 days were unchanged between 1999 and 2003 in most countries, including the highest ranking country in 1999 and 2003 (Malta). In five countries, however, an increase was observed. These were Croatia, Greece, Hungary, Italy and Russia (Moscow). In only Denmark and France was there a notable decrease in the prevalence rate of wine consumption at this frequency. When focusing on the development of this behaviour over the eight years of the ESPAD project, it is clear that the proportions to a large extent remain rather unchanged in many countries. No countries show a unidirectional increasing or decreasing trend between 1995 and 2003.

Consumption of spirits 3 times or more during the last 30 days (Figures 9a–c) There is a wide variety in the 30 days prevalence rates in the consumption of spirits 3 times or more in the past 30 days in the participating countries. In many of them, the figure for 2003 was similar to that in 1999. Hence, the high and low prevalence countries hold their positions. However, an increase in the reported consumption of spirits 3 times or more over the last 30 days was observed in nine countries. They include Cyprus, Estonia, the Faroe Islands, Greece, Greenland, Ireland, Italy, Poland, Portugal, the Slovak Republic and the United Kingdom. The consumption of spirits has declined in Denmark and France.

80

For this variable there was a continuous upward trend between 1995 and 2003 in the Faroe Islands, Ireland, Portugal, the Slovak Republic and the United Kingdom. No country showed a continuous decrease over the period.

Consumption of 101 cl of beer or more on the last drinking occasion (Figures 10a–c) The proportion of students in 2003 that reported that they had consumed at least 101 cl beer the last time they drank any alcohol, were very much the same as they were in 1999. There were some notable decreases, especially among the top prevalence countries like Denmark, Greenland and Ireland. Other countries where decreases were observed include Norway, Sweden and the United Kingdom. An increase was only noted in two countries (Croatia and Latvia). Despite the drop in prevalence rates on this variable, Denmark and Ireland remain ranked higher than other countries in this regard, while Greenland drops to a similar level as several other countries. The overall assessment of the findings from 1995 to 2003 is that the prevalence rates on this variable have remained rather stable over the years in most ESPAD countries. A long term decreasing trend was only found in one country (Sweden).

Consumption of 101 cl of alcopops or more on the last drinking occasion (Figures 11a–b) Alcopops are not available in all ESPAD countries. Thus, only some countries included this beverage when asking about consumption on the last drinking occasion. However, the pattern of consumption of at least 101 cl alcopops on the last drinking occasion is of course of interest to those countries where it is available. The results are very diverse. Generally, the prevalence rates are very low. Moreover, only a few countries showed any change from 1999 to 2003. However, the changes that did occur are relevant and are apparent in only four countries. These are Denmark, Ireland, Norway and the United Kingdom, where big increases in alcopops consumption were in evidence between 1999 and 2003.

Consumption of 15 cl of wine or more on the last drinking occasion (Figures 12a–c) The question related to wine consumption on the last drinking occasion was slightly altered for the

Changes in the use of alcohol and other drugs 2003

2003 survey. The amount indicating one glass was increased from 10 to 15 centilitres. This must be borne in mind when comparing the results on this variable between surveys, although it may not have changed the estimated number of glasses consumed by students. It can be argued, however, that most students would appear to consider 1–2 glasses of wine rather similar irrespective of whether in parenthesis it stated 10–20 cl or 15–30 cl. The proportion of students that indicated 15 centilitres or more on last drinking occasion decreased in ten countries. They include Denmark, the Faroe Islands, Finland, France, Iceland, Latvia, Lithuania, Norway, Romania and the Slovak Republic. The only countries with increasing proportions were Croatia and Russia (Moscow). Since the definition of the volume that relates to a glass of wine was larger in 2003 than it was in 1999, this in itself might have tilted the bias in favour of the number of countries reporting a decrease and thus should be taken into consideration when viewing such figures. A unidirectional increase from 1995 to 2003 was only observed in Croatia.

trend. A continuous decreasing trend was also only found in one country (Iceland).

Drunkenness, 20 times or more in lifetime (Figures 14a–c) The proportion of students who reported been drunk 20 times or more in a lifetime was relatively stable between 1999 and 2003 in many of the ESPAD countries. The increases that were observed were mainly found in the eastern parts of Europe. Increased prevalence rates were reported from Estonia, the Faroe Islands, Hungary, Ireland, Latvia, Lithuania, Russia (Moscow), the Slovak Republic and Ukraine. The only decreases in this respect were reported from Denmark and Iceland. Denmark nevertheless remained the highest ranked country in students reporting having been drunk 20 times or more in their lifetime. Over the years a unidirectional increase in the proportion of students that reported this behaviour was observed in five countries. They include Estonia, Ireland, Lithuania, the Slovak Republic and Ukraine. No one of the countries showed unidirectional decrease from 1995 to 2003.

Consumption of 11 cl of spirits on the last drinking occasion

Drunkenness, 10 times or more in the last 12 months

(Figures 13a–c) In many ESPAD countries the prevalence rates for the consumption of a relatively large quantity of spirits on last drinking occasion did not change between 1999 and 2003. This is true for high prevalence as well as low prevalence countries. However, in a few countries increases were observed and in one of them, the Faroe Islands, which topped the list last time, the increase was 12 percentage points. Other countries where increases were observed include the Czech Republic, Estonia, Italy and the Slovak Republic. Countries where the prevalence decreased include four of the Nordic countries (Denmark, Iceland, Norway, Sweden) together with Russia (Moscow) and the United Kingdom. The highest ranked countries in 1999 were again in the top group in 2003 (the Faroe Islands, Malta and Ireland). However, the top group now also includes some of the countries that showed increased prevalence rates for this variable between the two data collections (the Czech Republic and Estonia). Looking at the development of this variable over the years reveals that in only one country, the Slovak Republic, was there a continuous upward

(Figures 15a–c) In the 15–16 age group, the experience of being drunk is a rather recent event for most of the students. Therefore, the prevalence rates of been drunk 10 times or more over the last year is not very different from been drunk 20 times or more in a lifetime. The response pattern on this variable revealed that the figures were relatively unchanged between 1999 and 2003 in most countries. Increased values were reported from two Baltic States (Estonia and Lithuania) as well as from the Faroe Islands and the Slovak Republic. A decrease was observed in countries, which in 1999 were among the top group , including Denmark, Finland, Iceland and the United Kingdom, i.e. all four from the northern parts of Europe. With the exception of Iceland, these countries along with Ireland rank highest on this measure of adolescent drunkenness in the past 12 months. A long-term increase in the prevalence rates for been drunk 10 times or more in the last 12 months was observed only in Estonia for the period 1995– 2003.

Changes in the use of alcohol and other drugs 2003

81

2003 40

35 Malta 30

p. Re h e ec ec Cz re a G ati o Cr

25

20

Italy Slovenia

Hungary

Slovak Rep.

a ni ua th Li e . in a, i .K a an U Ukr m k ar Ro ia tv nm ce e La D ran F

Figure 8a. Changes between 1999 and 2003 in wine consumption 3 times or more during the last 30 days. Countries above the line have increased prevalence rates, and countries below have decreased. All students.

Russia (Moscow) Estonia Bulgaria

15

Cyprus Ireland

10

Po Fa lan r d, Fin oe Is S lan l., G wed en ree No d nla rw nd ay

Iceland 5

l ga rt u Po

0 0

5

10

15

20

25

30

35

40 1999

Boys

Girls

36

42 42

15

24 11 14

16 20 21 18 17 23 20 16 20 15

19 17 7 8

Croatia (13, 14, 19)

15 9 8

Hungary (13, 13, 19)

16 23 21 22

U.K. (19, 16, 18)

12 13

12

Ukraine (12, 16, 17)

15 17

16 18

Russia (Moscow) (.., 10, 17)

8 16

11 19

5

5

Estonia (5, 13, 15)

14 13

14 15

Bulgaria (.., 14, 14)

16 16

11 9

3

4

Lithuania (4, 15, 13)

15 12

15 16

Romania (.., 15, 13)

20 20

11 9

Latvia (.., 13, 12)

12 12 12 12

15 14 6 5 7 4 8

Cyprus (9, 8, 11)

14 5 7 6

Ireland (4, 8, 10)

12 11

13 13 14

Denmark (13, 13, 9)

6

10

France (.., 12, 8)

15 11

8 4

15 9 5

7

5 5 4 5 5 5 3

2003

2 3 3 5 6 6 5

Finland (5, 5, 5) Iceland (5, 4, 5)

3 5 2

Norway (3, 6, 3)

5

10

1999

5 6

Portugal (5, 4, 6)

8

1995

3 4

Greenland (.., 5, 6)

5 5 6 6

2003

7 6 8 8 5

Faroe Isl. (8, 5, 6)

7 7

1999

4

Sweden (6, 8, 8)

8 9 10

1995

7

Poland (10, 8, 8)

12

20

12

Slovak Rep. (15, 20, 19)

15

82

21

Greece (.., 16, 21)

26

30

21 14

Czech Rep. (13, 18, 21)

23

40

16

10 17

14 18 18

Figure 8b. Changes between 1999 and 2003 in wine consumption 3 times or more during the last 30 days. Percentages among boys and girls (values within brackets refer to all students 1995, 1999, 2003). Data sorted by all students 2003.

13

Slovenia (12, 22, 21)

26

33 32

16

Italy (24, 19, 23)

28

% 50

25

Malta (30, 37, 35) 29 29 29

8 4

0

0

10

20

30

40

50 %

Changes in the use of alcohol and other drugs 2003

Bulgaria

% 45

Finland

% 45

% 45

30

30

30

30

15

15

15

15

0

0

0

1995

1999

2003

Croatia

% 45

1995

1999

2003

France

% 45

0 1995

1999

2003

Latvia

% 45

1995

% 45

30

30

30

30

15

15

15

15

0

0 1995

% 45

1999

2003

0 1995

% 45

Cyprus

1999

2003

1999

2003

Lithuania

% 45

30

30

30

15

15

15

15

0 1995

% 45

1999

2003

Czech rep.

0 1995

% 45

1999

2003

1999

2003

% 45

Malta

30

30

30

15

15

15

15

0 1995

1999

2003

Denmark

% 45

0 1995

1999

2003

Hungary

% 45

1999

2003

Norway

% 45

30

15

15

15

15

0

% 45

2003

0 1995

% 45

Estonia

1999

2003

1999

2003

1995

% 45

Poland

30

30

30

30

15

15

15

15

0

0 1995

1999

2003

Faroe Isl.

% 45

0 1995

% 45

1999

2003

1999

2003

Portugal

% 45

30

30

30

15

15

15

15

0 1995

1999

2003

0 1995

1999

2003

2003

Ukraine

1995

30

0

1999

0 1995

% 45

Ireland

2003

0 1995

% 45

Iceland

1999

Sweden

% 45

30

1999

2003

Slovenia

1995

30

1995

1999

0 1995

30

0

2003

Slovak rep.

1995

30

0

1999

0 1995

% 45

Greenland

2003

Russia (Moscow)

1995

30

0

1999

0 1995

% 45

Greece

Romania

% 45

Italy

1999

2003

United Kingdom

0 1995

1999

2003

1995

1999

2003

Figure 8c. Changes between 1995 and 2003 in wine consumption 3 times or more during the last 30 days, by country. All students.

Changes in the use of alcohol and other drugs 2003

83

2003 50

Malta 40 Faroe Isl.

Italy

Estonia

Croatia

Czech Rep.

en y ed a w ) w Sw Nor e sco in o ra Uk ia (M d s us an R Icel tvia La

Poland 10

Denmark

Cyprus

Bulgaria Slov ak R Slov ep. enia Fran ce

Hungary 20

Ireland

Greece

nd la en l re ga G rt u Po

30

a ni ua th Li

Figure 9a. Changes between 1999 and 2003 in consumption of spirits 3 times or more during the last 30 days. Countries above the line have increased prevalence rates, and countries below have decreased. All students.

U.K.

Finland

Romania 0 0

10

20

30

40

50 1999

Boys

Girls

30

43 44

16

40

43

22

Ireland (20, 34, 38)

26 29

42 46

20

16

Faroe Isl. (19, 23, 37)

27 42

20 34

Greece (.., 29, 36)

34 41

25 32

32 34 20

7

Cyprus (15, 24, 29)

39 25

17 21 22

Czech Rep. (25, 28, 28)

31 30

27 27

Greenland (.., 21, 26)

21 33

22 20

18 23

Portugal (14, 19, 26)

23

Bulgaria (.., 22, 25)

29

11 16 23

27

21 25

13

6

Slovak Rep. (10, 20, 25)

22 27

16 22

22 21

14 14

Italy (18, 15, 25)

30

21 11 11

21 16 18 23

12

Slovenia (11, 22, 20)

12

17 24 21

France (.., 23, 19)

26 21

19 16

15

14

Sweden (14, 20, 18)

22 19 11

18 17 11

Norway (11, 20, 17)

20 17 14 15

18 18 7

Croatia (11, 15, 17)

20 21

12 18 12

Russia (Moscow) (.., 14, 16)

14 17

1999

16

Lithuania (15, 10, 14)

1995

6 12

1999

7

Poland (10, 9, 14)

18 17

4 11 13 13

Latvia (.., 12, 12)

16 13

2003

17

Iceland (18, 13, 13)

14 15

12 10

7

10

Romania (.., 6, 6)

7 10

10

8 8

Finland (8, 9, 9)

10 10

20

12 14

13 14 15 15 13

1995

15 14

Ukraine (18, 17, 16)

18 18

30

Hungary (14, 19, 22)

21 10 20 19

2003

5 6

Estonia (8, 8, 24)

27

40

39 31

29

Figure 9b. Changes between 1999 and 2003 in consumption of spirits 3 times or more during the last 30 days. Percentages among boys and girls (values within brackets refer to all students 1995, 1999, 2003). Data sorted by all students 2003.

28

Denmark (30, 40, 31)

41

84

45 42

27

U.K. (23, 32, 39)

26

33

% 50

40

Malta (35, 44, 43) 17

0

5 5

0

10

20

30

40

50 %

Changes in the use of alcohol and other drugs 2003

Bulgaria

% 45

Finland

% 45

% 45

30

30

30

30

15

15

15

15

0

0

0

1995

1999

2003

Croatia

% 45

1995

1999

2003

France

% 45

0 1995

1999

2003

Latvia

% 45

1995

% 45

30

30

30

30

15

15

15

15

0

0 1995

% 45

1999

2003

0 1995

% 45

Cyprus

1999

2003

1999

2003

Lithuania

% 45

30

30

30

15

15

15

15

0 1995

% 45

1999

2003

Czech rep.

0 1995

% 45

1999

2003

1999

2003

% 45

Malta

30

30

30

15

15

15

15

0 1995

1999

2003

Denmark

% 45

0 1995

1999

2003

Hungary

% 45

1999

2003

Norway

% 45

30

15

15

15

15

0

% 45

2003

0 1995

% 45

Estonia

1999

2003

1999

2003

1995

% 45

Poland

30

30

30

30

15

15

15

15

0

0 1995

1999

2003

Faroe Isl.

% 45

0 1995

% 45

1999

2003

1999

2003

Portugal

% 45

30

30

30

15

15

15

15

0 1995

1999

2003

0 1995

1999

2003

2003

Ukraine

1995

30

0

1999

0 1995

% 45

Ireland

2003

0 1995

% 45

Iceland

1999

Sweden

% 45

30

1999

2003

Slovenia

1995

30

1995

1999

0 1995

30

0

2003

Slovak rep.

1995

30

0

1999

0 1995

% 45

Greenland

2003

Russia (Moscow)

1995

30

0

1999

0 1995

% 45

Greece

Romania

% 45

Italy

1999

2003

United Kingdom

0 1995

1999

2003

1995

1999

2003

Figure 9c. Changes between 1995 and 2003 in consumption of spirits 3 times or more during the last 30 days, by country. All students.

Changes in the use of alcohol and other drugs 2003

85

2003 50

40 Denmark Ireland 30 Finland

Iceland

Croatia Estonia Lithuania

Czech Rep. 20 ,S

v lo

ia tv a La ari . lg p Bu k re

ly Ita

a

Figure 10a. Changes between 1999 and 2003 in consumption of 101 cl beer or more on the last drinking occasion. Countries above the line have increased prevalence rates, and countries below have decreased. All students.

10 Hungary

Faroe Isl.

Greenland

U.K.

Norway Sweden Poland

Malta Russia (Moscow) Slovenia Cyprus Greece Portugal

R Uk oma rai nia ne

0 0

10

20

30

40

50 1999

Boys

Girls

52

27 29

Denmark (39, 42, 37)

58 51 49 50 47

22 28 26

Ireland (38, 37, 33) 17

Finland (.., 23, 25)

34 35

13 15

Greenland (.., 41, 24)

42 23 23

39 25 15

Iceland (19, 26, 24)

33 29 35

20 20 11

United Kingdom (22, 30, 23)

44 34

16 11

20

16

Faroe Isl. (18, 24, 22)

34 31

13 12

26

Czech Rep. (17, 23, 22)

40 37

8 10

18

13

Norway (16, 24, 19)

30 24

19 15

34 33

Figure 10b. Changes between 1995 and 2003 in consumption of 101 cl beer or more on the last drinking occasion. Percentages among boys and girls (values within brackets refer to all students 1995, 1999, 2003). Data sorted by all students 2003.

5

22

Sweden (28, 22, 18)

28

12 9

19

5

Poland (12, 19, 16)

28 27

9 6

11

2 3

Croatia (6, 9, 14)

15 23 27

5 4 6 4

Malta (14, 14, 13)

23 23 7

Lithuania (3, 11, 13)

17 23

0 4 4

Estonia (.., 10, 13)

19 22

3 3

Latvia (.., 8, 12)

14 19

2 6

13

Russia (Moscow) (.., 9, 11)

14 13 14

Slovenia (9, 9, 10)

18

4 5 3 3 4

Bulgaria (.., 7, 10)

12 17 13 16 15

2 4 3 3

Cyprus (8, 10, 8) 1

Greece (.., 9, 7)

15 10

4 4

13 12

Slovak Rep. (4, 5, 7)

13 10

2003

Hungary (6, 4, 7)

8 12 10 12 11

Romania (.., 3, 6)

6 6

Ukraine (3, 2, 4)

3 6

86

50

40

30

20

1995

10

0 1 1 0 1 1

0

1999 2003 4 3 3

Portugal (7, 8, 6)

11

% 60

2 4

8 9

1999

6

Italy (10, 5, 7)

9

1995

0 1 0 1 1

0

10

20

30

40

50

60 %

Changes in the use of alcohol and other drugs 2003

Bulgaria

% 45

Finland

% 45

% 45

30

30

30

30

15

15

15

15

0

0

0

1995

1999

2003

Croatia

1995

1999

2003

0 1995

1999

2003

France

% 45

30

30

30

30

15

15

Data not a vaila ble

15

15

0

0 1995

% 45

1999

2003

% 45

0 1995

% 45

Cyprus

Latvia

1995

% 45

% 45

1999

2003

1999

2003

Lithuania

% 45

30

30

30

15

15

15

15

0 1995

% 45

1999

2003

Czech rep.

0 1995

% 45

1999

2003

1999

2003

% 45

Malta

30

30

30

15

15

15

15

0 1995

1999

2003

Denmark

% 60

0 1995

2003

Hungary

% 45

45

1999

2003

Slovak rep.

1995

30

0

1999

0 1995

% 45

Greenland

2003

Russia (Moscow)

1995

30

0

1999

0 1995

% 45

Greece

Romania

% 45

Italy

1999

2003

Slovenia

0 1995

1999

2003

Norway

% 45

1995

30

30

15

15

15

2003

Sweden

% 45

30

1999

30 15 0

0 1995

% 45

1999

2003

0 1995

% 45

Estonia

1999

2003

0 1995

% 45

Iceland

1999

2003

1995

% 45

Poland

30

30

30

30

15

15

15

15

0

0 1995

1999

2003

Faroe Isl.

% 45

0 1995

% 45

1999

2003

1999

2003

Portugal

% 45

30

30

30

15

15

15

15

0 1995

1999

2003

0 1995

1999

2003

Ukraine

1995

30

0

2003

0 1995

% 45

Ireland

1999

1999

2003

United Kingdom

0 1995

1999

2003

1995

1999

2003

Figure 10c. Changes between 1995 and 2003 in consumption of 101 cl beer or more on the last drinking occasion, by country. All students.

Changes in the use of alcohol and other drugs 2003

87

2003 25 U.K.

Ireland

20

Norway 15

Denmark

10

Portugal, Slovenia

Figure 11a. Changes between 1999 and 2003 in consumption of 101 cl alcopops or more on the last drinking occasion. Countries above the line have increased prevalence rates, and countries below have decreased. All students.

5

0

Iceland

Croatia, Estonia

Malta, Russia (Moscow) Sweden

Ro ma Ukraine nia 0 5

10

15

20

25 1999

Boys

Girls 9

U.K. (12, 24)

17 4 3 3 5 5 4 3

12 2 3 2

Estonia (3, 4)

5

3 3 3

Malta (3, 3)

6

1999

3 3

2003

3

2 2

Russia (Moscow) (3, 3) 2

4

2

Portugal (2, 2)

1

3 3

Slovenia (2, 2)

1 1

0

0

2003

2

3

2 2

1999 3

Sweden (3, 2)

3

88

14 6

Croatia (3, 4)

6

10

1

Iceland (5, 8)

5

20

19

Denmark (2, 15)

16

30

31 3

Norway (3, 17)

14

% 40

31 7

Ireland (5, 20)

10

Figure 11b. Changes between 1995 and 2003 in consumption of 101 cl alcopops or more on the last drinking occasion. Percentages among boys and girls (values within brackets refer to all students 1995, 1999, 2003). Data sorted by all students 2003.

16

Romania (0, 2)

0 0

Ukraine (2, 1)

0

1

0

10

20

30

40 %

Changes in the use of alcohol and other drugs 2003

Drunkenness, 3 times or more during the last 30 days (Figures 16a–c) The prevalence rates for been drunk 3 times or more in the last 30 days did not change very much in the participating countries between 1999 and 2003. Countries where an increase was found include Estonia, the Faroe Islands, Italy and Ukraine, i.e. countries that are rather disparate geographically. A decrease was only reported in Denmark and Sweden. The former remained, despite the decrease, in the top position for this behaviour followed by Ireland and the United Kingdom. During the eight years of the ESPAD project a continued increasing in prevalence rates were found in Estonia and Ukraine.

Binge drinking 3 times or more in the last 30 days (Figures 17a–c) The proportion of students, who reported “binge drinking”, i.e. drinking five or more drinks in a row at one drinking occasion, have increased in many ESPAD countries between 1999 and 2003. These countries include Bulgaria, Estonia, the Faroe Islands, Latvia, Lithuania, Portugal, the Slovak Republic, Sweden and Ukraine. Thus increases predominantly occurred in low prevalence countries across disparate parts of the European map. Decreasing figures were reported from Denmark, Greece, Greenland, Hungary, Iceland and Poland. Despite these changes the top countries more or less retained their positions, although two of them, Denmark and Poland, dropped down somewhat from 1999 to 2003. In both surveys the highest

figures were reported from Ireland. A continuous increase in the prevalence rates for binge drinking between 1995, 1999 and 2003 was only found in Estonia.

Drunk at the age of 13 or younger (Figures 18a–c) Many young people start drinking alcohol at a rather early age and some of them drink to the point of intoxication, as showed in the previous parts of this chapter. The proportion of students who reported been drunk at the age of 13 or younger differed to quite a degree among ESPAD countries. From 1999 to 2003 the proportions that report this behaviour remain rather unchanged in many of them, while in others rather large changes occurred. The proportion of students that have been drunk at the age of 13 or younger mainly increased in the eastern parts of Europe, including Bulgaria, Croatia, Estonia, Latvia, Lithuania, Russia (Moscow), the Slovak Republic, Slovenia and Ukraine. However, an increase was also reported in the Faroe Islands. Decreased percentages were only found in Denmark, Greenland and Romania. The top group still includes Denmark, Finland, Russia (Moscow) and the United Kingdom and they have been joined by Estonia. Greenland, which was in the top group in 1999 reported a decrease in 2003. Between 1995 and 2003 Ukraine was the only country in which a continuous increasing proportion of students reported been drunk at the age of 13. No country showed a continuous trend in the opposite direction.

Changes in illicit drug use prevalence Lifetime use of any illicit drug (Figures 19a–c) The proportion of students that have tried illicit drugs varies to a significant extent amongst countries, from less than 5% to almost half (44%) of the student population. Between 1999 and 2003 the prevalence rates for this variable increased in nine of the ESPAD countries. They include Bulgaria, Croatia, the Czech Republic, Estonia, Greenland, Hungary, Ireland, Portugal and the Slovak Republic. Decreasing prevalence rates were found in Greece, Latvia, Norway and Romania.

Among the four top countries from 1999 a further increase occurred in the Czech Republic and Ireland, while France and the United Kingdom remained relatively unchanged. Increases of 7–8 percentage points in the lifetime experiences of any illicit drug use were found in Bulgaria, the Czech Republic, Ireland and the Slovak Republic. The trend in prevalence rates over time between 1995 and 2003 show that a continuous increase has occurred in six ESPAD countries. The sizes of these increases vary but in many countries the rates have doubled or tripled. The countries in which increases

Changes in the use of alcohol and other drugs 2003

89

2003 40

Slovenia Malta Czech Rep.

Hungary Greece Italy

20

Lithuania Slovak Rep.

U.K.

Poland

Russia (Moscow)

Denmark

nd la Ire

Ukraine

Cyprus

Sw

10

Latvia Romania Bulgaria Finland Norway

en ed

G

ce an d l. Fr lan e Is en ro re Fa

d an el Ic

Figure 12a. Changes between 1999 and 2003 in consumption of 15 cl wine or more on the last drinking occasion. Countries above the line have increased prevalence rates, and countries below have decreased. All students.

Estonia

Croatia

30

Portugal

0 0

10

20

30

40 1999

Boys

Girls

24

39

Slovenia (19, 39, 39)

43 37 38

41

27 28

33 36 12

Croatia (20, 25, 31)

30 38 17

20 26 18

Lithuania (17, 40, 29)

41 24 37 37

40 35 33

Slovak Rep. (35, 39, 29)

32 24

40 27 34

United Kingdom (30, 25, 24)

20 17 39

31 30 20

Hungary (29, 22, 24)

29 29

27

19 16 17 15

Poland (21, 20, 20)

25 24

30

17 16 15 13 14

Italy (23, 18, 19)

27 26

Russia (Moscow) (.., 15, 19)

11 14 25

18 23 31 30

Denmark (28, 27, 18)

22 18

19

Latvia (.., 23, 17)

19 12

28 20

Romania (.., 19, 15)

33 23 25 15

10 10

13 12 13 15

Ireland (12, 12, 13)

9 16 17 16

17 8

Cyprus (12, 10, 12)

7 15

1999

15

19 12 19 17

Sweden (16, 14, 11)

12

1995

10 9

Finland (.., 16, 11)

9 12 11

12 12

Norway (11, 17, 10)

11

9 19 6 7 9

17 8 5 6 5 3 4

Faroe Isl. (7, 10, 4)

5

0

2003 8

6

Portugal (6, 8, 5)

10 11 12

10

5

Iceland (17, 9, 5)

8

1999

8

Greenland (.., 10, 8)

15

1995

19 10

France (.., 12, 8)

12

20

20 18 16

Bulgaria (.., 16, 14)

19 18

2003

9 9

Ukraine (21, 17, 15)

17

30

15

Greece (.., 20, 23)

25 29

14

90

36 37 38

Estonia (.., 31, 31)

29

40

30 30

Czech Rep. (35, 34, 35)

31 32

% 50

38 35 22

Malta (29, 34, 36)

35

Figure 12b. Changes between 1995 and 2003 in consumption of 15 cl wine or more on the last drinking occasion. Percentages among boys and girls (values within brackets refer to all students 1995, 1999, 2003). Data sorted by all students 2003.

14

7

2

0

10

20

30

40

50 %

Changes in the use of alcohol and other drugs 2003

Bulgaria

% 45

Finland

% 45

% 45

30

30

30

30

15

15

15

15

0

0

0

1995

1999

2003

Croatia

% 45

1995

1999

2003

France

% 45

0 1995

1999

2003

Latvia

% 45

1995

% 45

30

30

30

30

15

15

15

15

0

0 1995

% 45

1999

2003

0 1995

% 45

Cyprus

1999

2003

1999

2003

Lithuania

% 45

30

30

30

15

15

15

15

0 1995

% 45

1999

2003

Czech rep.

0 1995

% 45

1999

2003

1999

2003

% 45

Malta

30

30

30

15

15

15

15

0 1995

1999

2003

Denmark

% 60

0 1995

2003

Hungary

% 45

45

1999

2003

Slovak rep.

1995

30

0

1999

0 1995

% 45

Greenland

2003

Russia (Moscow)

1995

30

0

1999

0 1995

% 45

Greece

Romania

% 45

Italy

1999

2003

Slovenia

0 1995

1999

2003

Norway

% 45

1995

30

30

15

15

15

2003

Sweden

% 45

30

1999

30 15 0

0 1995

% 45

1999

2003

0 1995

% 45

Estonia

1999

2003

0 1995

% 45

Iceland

1999

2003

1995

% 45

Poland

30

30

30

30

15

15

15

15

0

0 1995

1999

2003

Faroe Isl.

% 45

0 1995

% 45

1999

2003

1999

2003

Portugal

% 45

30

30

30

15

15

15

15

0 1995

1999

2003

0 1995

1999

2003

Ukraine

1995

30

0

2003

0 1995

% 45

Ireland

1999

1999

2003

United Kingdom

0 1995

1999

2003

1995

1999

2003

Figure 12c. Changes between 1995 and 2003 in consumption of 15 cl wine or more on the last drinking occasion, by country. All students.

Changes in the use of alcohol and other drugs 2003

91

2003 40 Faroe Isl.

30 Malta Czech Rep.

Estonia

Ireland Greenland Denmark Norway U.K.

Poland

20 Slovak Rep. e ec re G ry ia ga en un v a H Slo ati ro C

Lithuania Sweden

France Russia (Moscow)

Italy 10

Iceland

Finland

ne ai via kr U Lat

Figure 13a. Changes between 1999 and 2003 in consumption of 11 cl spirits or more on the last drinking occasion. Countries above the line have increased prevalence rates, and countries below have decreased. All students.

Bulgaria Portugal C s ru yp

Romania 0 0

10

20

30

40 1999

Boys 40

Girls

33

42 28

31

12 16 18

Malta (22, 26, 27)

21

27

26

33 10 12 18

Estonia (.., 12, 23)

18

8 17

Greenland (.., 23, 22)

26 29

19 15

23

8

Poland (15, 19, 21)

27 30 30 30

11 13 19

26 27

22

United Kingdom (17, 22, 18)

17 14

25 20

9

14

4

Slovak Rep. (7, 11, 18)

22

8 14

33

26

Lithuania (30, 15, 17)

20 22

9 12

34 16

21 13

20

19

Sweden (20, 19, 15)

22 17

16 13

Finland (.., 16, 13)

21 15

12 12

Greece (.., 14, 12)

18 15

12 9

France (.., 13, 12)

16 15 11 11

10 9 8 8 9

Hungary (10, 10, 12)

15 24

13

Ukraine (17, 12, 11)

17 16

9 6

4

4

Slovenia (4, 10, 11)

10 12 8 9

10 12 3

Croatia (6, 8, 11)

12

6 10

10 10

5 4

Italy (8, 6, 11)

14

7

Russia (Moscow) (.., 16, 10)

19 13

1995

14 8

Latvia (.., 12, 9)

17 12

2003

1999

7 7

Bulgaria (.., 8, 9)

10 10

2003

5 7

8 9 11 12 10

6 6 3

Cyprus (7, 8, 6)

5 3

Romania (.., 1, 2)

3 3

10

4

Portugal (6, 7, 7)

12

20

30

Iceland (32, 23, 15)

23

30

24 19

12

1999

21

Norway (23, 26, 19)

20

1995

26 26

Denmark (28, 28, 20)

21

40

17 30

29

Figure 13b. Changes between 1995 and 2003 in consumption of 11 cl spirits or more on the last drinking occasion. Percentages among boys and girls (values within brackets refer to all students 1995, 1999, 2003). Data sorted by all students 2003.

25 25

Czech Rep. (15, 16, 23)

20

92

21

Ireland (14, 26, 23)

19

% 50

Faroe Isl. (34, 27, 39)

36 23

0

0 1

0

10

20

30

40

50 %

Changes in the use of alcohol and other drugs 2003

Bulgaria

% 45

Finland

% 45

% 45

30

30

30

30

15

15

15

15

0

0

0

1995

1999

2003

Croatia

% 45

1995

1999

2003

France

% 45

0 1995

1999

2003

Latvia

% 45

1995

% 45

30

30

30

30

15

15

15

15

0

0 1995

% 45

1999

2003

0 1995

% 45

Cyprus

1999

2003

1999

2003

Lithuania

% 45

30

30

30

15

15

15

15

0 1995

% 45

1999

2003

Czech rep.

0 1995

% 45

1999

2003

1999

2003

% 45

Malta

30

30

30

15

15

15

15

0 1995

1999

2003

Denmark

% 60

0 1995

2003

Hungary

% 45

45

1999

2003

Slovak rep.

1995

30

0

1999

0 1995

% 45

Greenland

2003

Russia (Moscow)

1995

30

0

1999

0 1995

% 45

Greece

Romania

% 45

Italy

1999

2003

Slovenia

0 1995

1999

2003

Norway

% 45

1995

30

30

15

15

15

2003

Sweden

% 45

30

1999

30 15 0

0 1995

% 45

1999

2003

0 1995

% 45

Estonia

1999

2003

0 1995

% 45

Iceland

1999

2003

1995

% 45

Poland

30

30

30

30

15

15

15

15

0

0 1995

1999

2003

Faroe Isl.

% 45

0 1995

% 45

1999

2003

1999

2003

Portugal

% 45

30

30

30

15

15

15

15

0 1995

1999

2003

0 1995

1999

2003

Ukraine

1995

30

0

2003

0 1995

% 45

Ireland

1999

1999

2003

United Kingdom

0 1995

1999

2003

1995

1999

2003

Figure 13c. Changes between 1995 and 2003 in consumption of 11 cl spirits or more on the last drinking occasion, by country. All students.

Changes in the use of alcohol and other drugs 2003

93

2003 50

40 Denmark Ireland

30

U.K.

Estonia

20

Faroe Isl.

Finland

Greenland Czech Rep. Sweden Iceland Norway

n ma Ro

10

ia en ov d Sl an ol P ria a ni ine a lg tia ua ra ) th k Bu oa Li U r ow . c C ep os R y (M vak ar a lo si ng , S Hu Portugal us R tvia Italy La ia

Figure 14a. Changes between 1999 and 2003 in the proportion who have been drunk 20 times or more in lifetime. Countries above the line have increased prevalence rates, and countries below have decreased. All students.

0

Malta France, Greece Cyprus

0

10

20

30

40

50 1999

Boys

Girls

36

47

32

28

18

29 26 27 27 30 27 28

United Kingdom (28, 29, 27)

27 26

Finland (28, 28, 26)

29 25 12

3

Estonia (7, 12, 26)

19 33

6 19

22

23 22 18 12

5 6

Lithuania (9, 13, 20)

20 29

12 16

5

Czech Rep. (11, 16, 18)

23 25

8 13

3 24 20

23

2

Ukraine (2, 13, 18)

16

10 11 17 16 15

Sweden (19, 19, 17)

18 20 21

20 19

Iceland (19, 20, 16)

16

15 9

4

Slovenia (7, 13, 15)

15 20

8 10

Russia (Moscow) (.., 10, 15)

12 18

9

Norway (9, 16, 14)

14

16 13

10

2

Slovak Rep. (6, 10, 14)

12 20

8 10

Latvia (.., 10, 14)

15 19

6 10

11 11

4

Hungary (8, 7, 11)

16 14 16 15

2 5 4 6 5

Poland (9, 11, 10) Bulgaria (.., 8, 10)

11 15

4 7

7

Croatia (4, 7, 9)

11 14

0 3 5

6 5 5

1995

Malta (4, 4, 4)

7

1999

Romania (.., 2, 3)

4 7

2003

4 4 4 4 3 3 2

Cyprus (2, 2, 2) 0

2003

2 2

Greece (.., 4, 3)

6

1999

2 2 2 1

France (.., 4, 3)

7

1995

0

Portugal (2, 3, 3)

6

10

2 2 3 3 2 3

Italy (4, 2, 5)

4 8

20

9 13

11 18

30

12

Greenland (.., 22, 21)

21 24

Figure 14b. Changes between 1995 and 2003 in the proportion who have been drunk 20 times or more in lifetime. Percentages among boys and girls (values within brackets refer to all students 1995, 1999, 2003). Data sorted by all students 2003.

17

Faroe Isl. (20, 19, 24)

26 26

40

15

Ireland (19, 25, 30)

28

33

94

34 31

22

% 50

28

Denmark (32, 41, 36)

41

2 2 0 0 0

0

10

20

30

40

50 %

Changes in the use of alcohol and other drugs 2003

Bulgaria

% 45

Finland

% 45

% 45

30

30

30

30

15

15

15

15

0

0

0

1995

1999

2003

Croatia

% 45

1995

1999

2003

France

% 45

0 1995

1999

2003

Latvia

% 45

1995

% 45

30

30

30

30

15

15

15

15

0

0 1995

% 45

1999

2003

0 1995

% 45

Cyprus

1999

2003

1999

2003

Lithuania

% 45

30

30

30

15

15

15

15

0 1995

% 45

1999

2003

Czech rep.

0 1995

% 45

1999

2003

1999

2003

% 45

Malta

30

30

30

15

15

15

15

0 1995

1999

2003

Denmark

% 45

0 1995

1999

2003

Hungary

% 45

1999

2003

Norway

% 45

30

15

15

15

15

0

% 45

2003

0 1995

% 45

Estonia

1999

2003

1999

2003

1995

% 45

Poland

30

30

30

30

15

15

15

15

0

0 1995

1999

2003

Faroe Isl.

% 45

0 1995

% 45

1999

2003

1999

2003

Portugal

% 45

30

30

30

15

15

15

15

0 1995

1999

2003

0 1995

1999

2003

2003

Ukraine

1995

30

0

1999

0 1995

% 45

Ireland

2003

0 1995

% 45

Iceland

1999

Sweden

% 45

30

1999

2003

Slovenia

1995

30

1995

1999

0 1995

30

0

2003

Slovak rep.

1995

30

0

1999

0 1995

% 45

Greenland

2003

Russia (Moscow)

1995

30

0

1999

0 1995

% 45

Greece

Romania

% 45

Italy

1999

2003

United Kingdom

0 1995

1999

2003

1995

1999

2003

Figure 14c. Changes between 1995 and 2003 in the proportion who have been drunk 20 times or more in lifetime, by country. All students

Changes in the use of alcohol and other drugs 2003

95

2003 40

Denmark 30

Ireland U.K. Finland

Faroe Isl.

Estonia

ay rw No ia en ov e Sl ain r d Uk via an t l La Po ) tia a oa ni ow Cr ua osc p. th e Li (M R aria ry a k si va ulg a us Slo B ung R H

20

Czec h Re

Greenland

p.

Sweden Iceland

10

Malta Italy s u Cypr

Figure 15a. Changes between 1999 and 2003 in the proportion who have been drunk 10 times or more during last 12 months. Countries above the line have increased prevalence rates, and countries below have decreased. All students.

Romania France, Greece, Portugal

0 0

10

20

30

40 1999

Boys

Girls

34

43

30

26 24 25 21

26 25 31 28 25

24

20

Faroe Isl. (22, 18, 22)

28 23

10 22

9

3

Estonia (6, 10, 21)

15 26

7 17

Greenland (.., 17, 19)

20 19

15 19

14

7

Czech Rep. (11, 14, 16)

19 21

9 12

18

18

Sweden (18, 18, 15)

21 15

16 13

9

4 5

Lithuania (7, 9, 15)

15 20 21 19

9 18 15 12

10

Norway (11, 16, 14)

17 14

16 14

9

3

Slovenia (7, 12, 13)

14 15 4

9 10 1

Ukraine (1, 11, 13)

12 17

9 9

Russia (Moscow) (.., 9, 12)

10 14

7 9

7

1

Slovak Rep. (5, 7, 11)

9 16

5 7

Latvia (.., 9, 10)

11 12

6 7

Bulgaria (.., 7, 10)

10 15 9 9 12 10 13 12

5 6

Hungary (6, 6, 9)

3 2

Poland (7, 9, 8)

3

5

7

Croatia (4, 6, 8)

9 6 5 6 6 4 7

1995 1999 2003

0 3 5 2 3 4 3 1 2

Italy (4, 2, 4) Romania (.., 3, 4) 2

Portugal (1, 3, 2)

6 3

France (.., 3, 2)

4 3 4 2

Cyprus (2, 1, 1) 0

1995 1999

0 1 1 1 1

Greece (.., 3, 2)

4 3 6

10

5 4

Malta (4, 4, 4)

4 4

20

22

Iceland (21, 19, 14)

14

12

96

28

24

Finland (28, 29, 23)

29

30

25

U.K. (25, 28, 24)

30

Figure 15b. Changes between 1995 and 2003 in the proportion who have been drunk 10 times or more during last 12 months. Percentages among boys and girls (values within brackets refer to all students 1995, 1999, 2003). Data sorted by all students 2003.

16

Ireland (20, 27, 29)

28

40

35

29 22

% 50

30

Denmark (32, 39, 34)

40

2003

2 2 1 1 0 0 0

0

10

20

30

40

50 %

Changes in the use of alcohol and other drugs 2003

Bulgaria

% 45

Finland

% 45

% 45

30

30

30

30

15

15

15

15

0

0

0

1995

1999

2003

Croatia

% 45

1995

1999

2003

France

% 45

0 1995

1999

2003

Latvia

% 45

1995

% 45

30

30

30

30

15

15

15

15

0

0 1995

% 45

1999

2003

0 1995

% 45

Cyprus

1999

2003

1999

2003

Lithuania

% 45

30

30

30

15

15

15

15

0 1995

% 45

1999

2003

Czech rep.

0 1995

% 45

1999

2003

1999

2003

% 45

Malta

30

30

30

15

15

15

15

0 1995

1999

2003

Denmark

% 60

0 1995

2003

Hungary

% 45

45

1999

2003

Slovak rep.

1995

30

0

1999

0 1995

% 45

Greenland

2003

Russia (Moscow)

1995

30

0

1999

0 1995

% 45

Greece

Romania

% 45

Italy

1999

2003

Slovenia

0 1995

1999

2003

Norway

% 45

1995

30

30

15

15

15

2003

Sweden

% 45

30

1999

30 15 0

0 1995

% 45

1999

2003

0 1995

% 45

Estonia

1999

2003

0 1995

% 45

Iceland

1999

2003

1995

% 45

Poland

30

30

30

30

15

15

15

15

0

0 1995

1999

2003

Faroe Isl.

% 45

0 1995

% 45

1999

2003

1999

2003

Portugal

% 45

30

30

30

15

15

15

15

0 1995

1999

2003

0 1995

1999

2003

Ukraine

1995

30

0

2003

0 1995

% 45

Ireland

1999

1999

2003

United Kingdom

0 1995

1999

2003

1995

1999

2003

Figure 15c. Changes between 1995 and 2003 in the proportion who have been drunk 10 times or more during last 12 months, by country. All students.

Changes in the use of alcohol and other drugs 2003

97

2003 30

Denmark 25

Ireland U.K.

20 Greenland Faroe Isl.

Estonia

Finland

Ukraine

) w co os d an (M el ia Ic ss nd u la , R a Po aria ni lg ve lo Bu S a ni ep. ua R th k Li ova y ar tvia Sl ng a

15 C

5

Czech Rep. Norway

u L H tia,

a ro

10 ni a ma Ro

Figure 16a. Changes between 1999 and 2003 in the proportion who have been drunk 3 times or more during last 30 days. Countries above the line have increased prevalence rates, and countries below have decreased. All students.

Sweden

Italy

Malta

Po rtu France ga l

Greece Cyprus 0 0

5

10

15

20

25

30 1999

Boys

Girls

24

36

20

U.K. (22, 24, 23)

25 25

Greenland (.., 19, 19)

25 20

17 19

11

10

Faroe Isl. (11, 9, 18)

13 20

6 16

7

3

Estonia (4, 8, 17)

12 23

6 13

19 19 4

1

Ukraine (2, 11, 16)

14 18 14

10 10 5

Czech Rep. (10, 13, 13)

18 17

9 10

9

8

Norway (8, 14, 12)

14 12

13 10

8

13

5

Slovenia (7, 11, 12)

16

9 8

11 12

6 6

Lithuania (9, 9, 12)

17

8 9

2

Slovak Rep. (4, 9, 11)

10 14 13 15

6 8 12 11 11 13 12

Sweden (13, 14, 10)

11 14

Iceland (14, 12, 10)

12 9

9

10

4

Poland (7, 10, 10)

14 13

6 5

Bulgaria (.., 8, 10)

12 17

4 8

Russia (Moscow) (.., 8, 10)

7 12

7 9

9 9

4

Hungary (5, 7, 9)

11

3 5

8

1

Croatia (4, 7, 8)

9 12

3 5

Latvia (.., 7, 8)

12 12

4 7

8

7 3

1999

3 3

2003

2003

3 3

Romania (.., 2, 4)

5 6

1999

3 3

Greece (.., 3, 3)

4 4

1995

4 4 1

Portugal (3, 4, 4)

6 6

1 3

France (.., 6, 3)

7 5 5

4 2

4

1 1

Cyprus (2, 3, 2)

3

10

2

Malta (4, 5, 5)

6

1995

4

Italy (8, 3, 7)

5 9 9

20

18 17 17

Finland (18, 18, 16)

15

30

23

15 24 23 22

Figure 16b. Changes between 1995 and 2003 in the proportion who have been drunk 3 times or more during last 30 days. Percentages among boys and girls (values within brackets refer to all students 1995, 1999, 2003). Data sorted by all students 2003.

14

Ireland (26, 24, 25)

27 27

98

26 21

17

% 40

18

Denmark (21, 30, 26)

30

0

0

0

10

20

30

40 %

Changes in the use of alcohol and other drugs 2003

Bulgaria

% 30

Finland

% 30

% 30

20

20

20

20

10

10

10

10

0

0

0

1995

1999

2003

Croatia

% 30

1995

1999

2003

France

% 30

0 1995

1999

2003

Latvia

% 30

1995

% 30

20

20

20

20

10

10

10

10

0

0 1995

% 30

1999

2003

0 1995

% 30

Cyprus

1999

2003

1999

2003

Lithuania

% 30

20

20

20

10

10

10

10

0 1995

% 30

1999

2003

Czech rep.

0 1995

% 30

1999

2003

1999

2003

% 30

Malta

20

20

20

10

10

10

10

0 1995

1999

2003

Denmark

% 30

0 1995

1999

2003

Hungary

% 30

1999

2003

Norway

% 30

20

10

10

10

10

0

% 30

2003

0 1995

% 30

Estonia

1999

2003

1999

2003

1995

% 30

Poland

20

20

20

20

10

10

10

10

0

0 1995

1999

2003

Faroe Isl.

% 30

0 1995

% 30

1999

2003

1999

2003

Portugal

% 30

20

20

20

10

10

10

10

0 1995

1999

2003

0 1995

1999

2003

2003

Ukraine

1995

20

0

1999

0 1995

% 30

Ireland

2003

0 1995

% 30

Iceland

1999

Sweden

% 30

20

1999

2003

Slovenia

1995

20

1995

1999

0 1995

20

0

2003

Slovak rep.

1995

20

0

1999

0 1995

% 30

Greenland

2003

Russia (Moscow)

1995

20

0

1999

0 1995

% 30

Greece

Romania

% 30

Italy

1999

2003

United Kingdom

0 1995

1999

2003

1995

1999

2003

Figure 16c. Changes between 1995 and 2003 in the proportion who have been drunk 3 times or more during last 30 days, by country. All students.

Changes in the use of alcohol and other drugs 2003

99

2003 40

Ireland 30

20

Sw ed en

Bulgaria

Ukraine ak ov Sl

U.K. Malta Norway

Poland Denmark

Latvia

Slovenia Faroe Isl. Greenland Czech Rep. Russia (Moscow) Finland

Estonia

tia nia oa ma Cr Ro

ep R

.

Figure 17a. Changes between 1999 and 2003 in the proportion who have reported “binge drinking” 3 times or more during last 30 days. Countries above the line have increased prevalence rates, and countries below have decreased. All students.

Portugal Lithuania

Cyprus France Hungary

10

Iceland Greece

0 0

10

20

30

40 1999

Boys

Girls

25

32 31

24 33

26 18 23

32 26

7 23 11 18 19 19

Denmark (22, 30, 24)

37 31

22 18

19 19 27

22 10 23 19 24

19 18

Latvia (.., 14, 22)

9

Ukraine (11, 10, 22)

9 8

18 14 12

28

15

Bulgaria (.., 11, 21)

15 26

6 16

14

5

Estonia (10, 14, 20)

18 26

12 15

Greenland (.., 25, 19)

25 23

22 16

18

6

Faroe Isl. (12, 15, 19)

21 21

8 17

19

7

Czech Rep. (14, 17, 18)

25 24

11 13

Russia (Moscow) (.., 16, 17)

20 22 22 21 19 15 19

18 15 14

Croatia (8, 12, 15)

3

Slovak Rep. (7, 8, 15)

3

7 10

10 12 20

7 12

5

2

Portugal (4, 7, 15)

11 20

4 10

13 12 25

9

Italy (20, .., 13)

19 12

8 9

Iceland (11, 17, 11)

18 13

1995

15 9

Greece (.., 18, 11)

24 14

1999

18 8

18 19 18 15

Romania (.., 11, 11)

6 5

Cyprus (.., 12, 10)

6 5

France (.., 12, 9)

16 13 18 18

2003

7 7 7 8

Hungary (13, 12, 8)

12

20

6 5 7

Lithuania (10, 9, 13)

19

1995

12 12

Finland (19, 18, 16) 13

30

5

Slovenia (7, 25, 22)

29

2003

23 24 12 13

Sweden (16, 17, 24)

22

1999

15

Norway (17, 24, 24)

26 25

100

32 33 27 29

Malta (16, 22, 25)

25

40

20

13 20

% 50

U.K. (22, 30, 27) Poland (11, 31, 25)

41

Figure 17b. Changes between 1995 and 2003 in the proportion who have reported “binge drinking” 3 times or more during last 30 days. Percentages among boys and girls (values within brackets refer to all students 1995, 1999, 2003). Data sorted by all students 2003.

Ireland (23, 31, 32)

20

5

10

0

0

10

20

30

40

50 %

Changes in the use of alcohol and other drugs 2003

Bulgaria

% 45

Finland

% 45

% 45

30

30

30

30

15

15

15

15

0

0

0

1995

1999

2003

Croatia

% 45

1995

1999

2003

France

% 45

0 1995

1999

2003

Latvia

% 45

1995

% 45

30

30

30

30

15

15

15

15

0

0 1995

% 45

1999

2003

0 1995

% 45

Cyprus

1999

2003

1999

2003

Lithuania

% 45

30

30

30

15

15

15

15

0 1995

% 45

1999

2003

Czech rep.

0 1995

% 45

1999

2003

1999

2003

% 45

Malta

30

30

30

15

15

15

15

0 1995

1999

2003

Denmark

% 45

0 1995

1999

2003

Hungary

% 45

1999

2003

Norway

% 45

30

15

15

15

15

0

% 45

2003

0 1995

% 45

Estonia

1999

2003

1999

2003

1995

% 45

Poland

30

30

30

30

15

15

15

15

0

0 1995

1999

2003

Faroe Isl.

% 45

0 1995

% 45

1999

2003

1999

2003

Portugal

% 45

30

30

30

15

15

15

15

0 1995

1999

2003

0 1995

1999

2003

2003

Ukraine

1995

30

0

1999

0 1995

% 45

Ireland

2003

0 1995

% 45

Iceland

1999

Sweden

% 45

30

1999

2003

Slovenia

1995

30

1995

1999

0 1995

30

0

2003

Slovak rep.

1995

30

0

1999

0 1995

% 45

Greenland

2003

Russia (Moscow)

1995

30

0

1999

0 1995

% 45

Greece

Romania

% 45

Italy

1999

2003

United Kingdom

0 1995

1999

2003

1995

1999

2003

Figure 17c. Changes between 1995 and 2003 in the proportion who have reported “binge drinking” 3 times or more during last 30 days, by country. All students.

Changes in the use of alcohol and other drugs 2003

101

2003 50

40 Russia (Moscow)

U.K.

Estonia

Denmark

Finland Slovenia Lithuania Bulgaria

30

Faroe Isl.

Czech Rep. Malta Portugal

Poland Hungary Italy

10

Cyprus

Greenland

tia oa Cr ia tv La

Slovak Rep. 20

Greece

Figure 18a. Changes between 1999 and 2003 in the proportion of all students who have been drunk at the age of 13 or younger. Countries above the line have increased prevalence rates, and countries below have decreased. All students.

Ukraine Ireland Sweden

Iceland, Norway Romania

0 0

10

20

30

40

50 1999

Boys

Girls Russia (Moscow) (.., 33, 37)

35 40

36 24

36 35 10

Estonia (17, 19, 35)

26 42

14 27

46 37

37 32

34 34

33 34

Greenland (.., 35, 29)

37 29

32 28

7

3

Ukraine (5, 22, 26)

26 33 25

18 19 16

Ireland (20, 25, 24)

28 25 24 25

21 22

Croatia (18, 19, 23)

11 12

22

Slovenia (16, 17, 23)

12 12

23 22

Lithuania (18, 16, 23)

30

15 20

Figure 18b. Changes between 1995 and 2003 in the proportion of all students who have been drunk at the age of 13 or younger. Percentages among boys and girls (values within brackets refer to all students 1995, 1999, 2003). Data sorted by all students 2003.

37

Finland (35, 33, 33)

32

27

19

30

13 9 17

26 26 25

19 23

25

Bulgaria (.., 15, 22)

11 7

Slovak Rep. (12, 14, 22)

11 17

15

9

Czech Rep. (12, 16, 19)

20 21 21

12 16 8

Faroe Isl. (15, 15, 19)

18 20

11 18

23

21

Iceland (22, 17, 17)

19 18 13

16 15 11

Norway (12, 17, 17)

19 17

15 18

Romania (.., 22, 16)

32 22

15 11

16 15

6 6

Poland (11, 11, 14)

20

9 14 15 10 14 14

Greece (.., 9, 10)

11 14 12 13 10 10 10

20

10

Cyprus (12, 7, 7) 0

2003

6 7 8 7 8 9 5 5

Italy (11, 7, 10)

8

1999 9 10 9 9

Portugal (12, 12, 10)

12

1995

6 7

Hungary (8, 10, 13)

13 17

2003

11 11 12

Malta (12, 14, 13)

17

30

11

20 16 18

27

1999

Latvia (.., 16, 22)

21 19

1995

22 22

Sweden (24, 24, 22)

27

40

34

Denmark (39, 42, 34)

48

102

38

U.K. (40, 38, 36)

40

% 50

30 34

42

0

10

20

30

40

50 %

Changes in the use of alcohol and other drugs 2003

Bulgaria

% 45

Finland

% 45

% 45

30

30

30

30

15

15

15

15

0

0

0

1995

1999

2003

Croatia

% 45

1995

30

15

Data not

15

0

2003

France

% 45

30

1999

0 1995

% 45

1999

2003

1999

2003

Latvia

1999

30

30

15

15

2003

1999

2003

Lithuania

% 45

30

30

30

15

15

15

15

0 1995

% 45

1999

2003

Czech rep.

0 1995

% 45

1999

2003

1999

2003

% 45

Malta

30

30

30

15

15

15

15

0 1995

1999

2003

Denmark

% 45

0 1995

1999

2003

Hungary

% 45

1999

2003

Norway

% 45

30

15

15

15

15

0

% 45

2003

0 1995

% 45

Estonia

1999

2003

1999

2003

1995

% 45

Poland

30

30

30

30

15

15

15

15

0

0 1995

1999

2003

Faroe Isl.

% 45

0 1995

% 45

1999

2003

1999

2003

Portugal

% 45

30

30

30

15

15

15

15

0 1995

1999

2003

0 1995

1999

2003

2003

Ukraine

1995

30

0

1999

0 1995

% 45

Ireland

2003

0 1995

% 45

Iceland

1999

Sweden

% 45

30

1999

2003

Slovenia

1995

30

1995

1999

0 1995

30

0

2003

Slovak rep.

1995

30

0

1999

0 1995

% 45

Greenland

2003

Russia (Moscow)

1995

30

0

1999

0 1995

% 45

Greece

1995

% 45

0 1995

% 45

Cyprus

0 1995

% 45

avail able

Romania

% 45

Italy

1999

2003

United Kingdom

0 1995

1999

2003

1995

1999

2003

Figure 18c. Changes between 1995 and 2003 in the proportion of all students who have been drunk at the age of 13 or younger, by country. All students.

Changes in the use of alcohol and other drugs 2003

103

2003 50

Czech Rep. Ireland France

40

Sl

nd la p. en e re R G vak o

30

Iceland Bulgaria

Croatia

Denmark Russia (Moscow) Ukraine Latvia

20 Portugal Hungary Finland Malta Faroe Isl. Sweden

10

Norway Greece

us pr Cy

0

Slovenia Italy

nd la a a ni ni Po ua to th Es Li

Figure 19a. Changes between 1999 and 2003 in lifetime experience of any illicit drug. Countries above the line have increased prevalence rates, and countries below have decreased. All students.

U.K.

Romania

0

10

20

30

40

50 1999

Boys

Girls

26

40

Czech Rep. (23, 35, 44)

48

32

Ireland (37, 32, 40)

35

41 44

29

United Kingdom (42, 36, 38)

39 42

40 40 33 35

France (.., 35, 38)

38 43

32 34

15

17

Italy (21, 26, 28)

29 33

24 24

Greenland (.., 21, 27)

21 29

21 26

13

6

Slovak Rep. (10, 20, 27)

24 32

17 22

11

12 19

20

15

Denmark (18, 25, 23)

31 27 10

19

20 19 5

Croatia (8, 17, 23)

24

14 22

Russia (Moscow) (.., 24, 22)

26 26

22 19

Bulgaria (.., 14, 22)

15 24

12 19

20

9

Ukraine (14, 21, 21)

27 29

14 12

13 23 25 11 16 21

Poland (9, 18, 19)

6

Portugal (8, 12, 18)

6

13 14 9 15

Latvia (.., 22, 17)

26 21

18 13

4 5

4 8 14

12

8

Iceland (10, 16, 13)

18 15

13 11

5

9

Malta (2, 8, 11)

9 9 8

Faroe Isl. (12, 8, 10)

13 12

9 7 11 10

8 9 11

1995

6

Norway (6, 13, 9)

5

Sweden (6, 9, 8)

5 6 7

8

1999

11 10

Greece (.., 10, 7)

13

2003

7 5

11

3 2 3

Cyprus (6, 3, 5)

6 8

Romania (.., 10, 3)

10 5

10

2

10

15

2003

10 12

3

1995

6

Finland (5, 10, 11)

11 11

20

10 10

Hungary (5, 12, 16)

17 18

1999

3

Lithuania (3, 15, 16)

21 21

30

5

Estonia (8, 16, 24)

21 28

40

23 27

24

Figure 19b. Changes between 1995 and 2003 in lifetime experience of any illicit drug (values within brackets refer to all students 1995, 1999, 2003). Data sorted by all students 2003.

12

Slovenia (13, 26, 29)

28 31

104

30 40

42

% 50

19

9 2

0

0

10

20

30

40

50 %

Changes in the use of alcohol and other drugs 2003

Bulgaria

% 45

Finland

% 45

% 45

30

30

30

30

15

15

15

15

0

0

0

1995

1999

2003

Croatia

% 45

1995

1999

2003

France

% 45

0 1995

1999

2003

Latvia

% 45

1995

% 45

30

30

30

30

15

15

15

15

0

0 1995

% 45

1999

2003

0 1995

% 45

Cyprus

1999

2003

1999

2003

Lithuania

% 45

30

30

30

15

15

15

15

0 1995

% 45

1999

2003

Czech rep.

0 1995

% 45

1999

2003

1999

2003

% 45

Malta

30

30

30

15

15

15

15

0 1995

1999

2003

Denmark

% 45

0 1995

1999

2003

Hungary

% 45

1999

2003

Norway

% 45

30

15

15

15

15

0

% 45

2003

0 1995

% 45

Estonia

1999

2003

1999

2003

1995

% 45

Poland

30

30

30

30

15

15

15

15

0

0 1995

1999

2003

Faroe Isl.

% 45

0 1995

% 45

1999

2003

1999

2003

Portugal

% 45

30

30

30

15

15

15

15

0 1995

1999

2003

0 1995

1999

2003

2003

Ukraine

1995

30

0

1999

0 1995

% 45

Ireland

2003

0 1995

% 45

Iceland

1999

Sweden

% 45

30

1999

2003

Slovenia

1995

30

1995

1999

0 1995

30

0

2003

Slovak rep.

1995

30

0

1999

0 1995

% 45

Greenland

2003

Russia (Moscow)

1995

30

0

1999

0 1995

% 45

Greece

Romania

% 45

Italy

1999

2003

United Kingdom

0 1995

1999

2003

1995

1999

2003

Figure 19c. Changes between 1995 and 2003 in lifetime experience of any illicit drug, by country. All students.

Changes in the use of alcohol and other drugs 2003

105

occurred between 1995 and 2003 include Croatia, the Czech Republic, Estonia, Hungary, Portugal and the Slovak Republic. No country showed a continuous decrease between the three data collections.

Lifetime use of cannabis (Figures 20a–c) The majority of those who tried any illicit drug have used marijuana or hashish. The lifetime prevalence rates for cannabis use are thus rather similar to the figures presented above and the changes that are found are almost all in the same countries. Increased prevalence rates were reported from Bulgaria, Croatia, the Czech Republic, Estonia, Greenland, Hungary, Ireland, Poland, Portugal and the Slovak Republic. The top four countries in 2003 are the same as those in 1999. The Czech Republic and Ireland report further increases while France and the United Kingdom remain relatively unchanged. No country had decreasing figures on this variable. Thus, in a majority of the countries the figures for 2003 are as they were in 1999. Looking at the trend development since 1995 reveals that a continuous increase over time seems in order for Croatia, the Czech Republic, Estonia, Hungary, Poland, and the Slovak Republic, i.e. countries in the eastern part of Europe. In no ESPAD country was there a continuous development in the opposite direction.

Cannabis use during the last 30 days (Figures 21a–c) The proportion of students in various ESPAD countries that used marijuana or hashish during the last 30 days as expected were much lower than the lifetime prevalence rates. In most of the countries there were no changes from 1999 to 2003. Only a few countries show increasing figures. They include Bulgaria, the Slovak Republic and the United Kingdom. No country has decreasing values for this variable. The seven top countries in 1999 were again at the top in 2003, with France, the United Kingdom and the Czech Republic in the top 3 positions. The impression that the situation was rather constant is reinforced on viewing the trends from 1995 to 2003. No continuous long term increases or decreases were evident.

106

Lifetime use of any illicit drug other than cannabis (Figures 22a–c) The drugs included in this definition are amphetamines, LSD or other hallucinogens, cocaine/crack, ecstasy and heroin. The proportion of students that used any illicit drug other than cannabis is much lower than the cannabis prevalence rates in all ESPASD countries. In most countries the relatively low figures are unchanged. No country reported an increase, while lower figures for 2003 as compared to 1999 were found in four countries including Latvia, Poland, Russia (Moscow) and Romania. There were no continuous upward or downward trends in the lifetime use of any illicit drug other than cannabis between 1995 and 2003.

Lifetime use of tranquillisers or sedatives without a doctor’s prescription (Figures 23a–c) The prevalence rates for the use of tranquillisers or sedatives without a doctor’s prescription are relatively low in most ESPAD countries. Moreover, there were very few changes from 1999 to 2003. In only one country, Estonia, a substantial increase occurred whereas there was a decrease the Czech Republic, one of the two top countries in 1999 together with Poland. The other two top countries in 1999 (Lithuania and France) show no change for 2003, and alas they still form part of the top group. From 1995 to 2003 only small changes have been noted in this behaviour. In no country, however, has a continuous upward or downward trend been observed.

Lifetime use of alcohol together with pills (Figures 24a–c) In many ESPAD countries students have tried the combination of alcohol and pills of various types. The assumption for such use is based on the expectation that mixing products induces a higher degree of intoxication. Whatever the assumption or for that matter reason for the use of this cocktail, it would appear to be a rather common, yet dangerous, phenomenon in many ESPAD countries, especially amongst girls. In a large majority of the countries the proportion of students reporting this behaviour remained relatively stable between 1999 and 2003. However, there was an increase in the Slovak Republic, one of the high prevalence countries in 1999, which became the highest ranked country in this regard in 2003.

Changes in the use of alcohol and other drugs 2003

Decreasing prevalence rates for this variable were only reported in Denmark, Sweden and the United Kingdom. As a result Denmark drops from being ranked 1st on this measure to being ranked 12th, and Sweden drops from 2nd to 10th place in this ranking. The tendency for the increase in the prevalence rates for the Slovak Republic is visible through out the period 1995 to 2003. In Sweden and the United Kingdom the proportion of students that reported this behaviour have decreased continuously during the same period.

Lifetime use of inhalants (Figures 25a–c) The lifetime prevalence rates for the use of inha-

lants have not changed very much between the two last surveys in a large majority of the ESPAD countries. The differences between countries are rather great; from a few percentages to about one fourth of the student population. Increases were mainly found in the Faroe Islands and Portugal, and decreases in Ireland and Lithuania. The decrease in Ireland, however, did not effect its top position together with Greenland and Malta. The trends over the three surveys revealed the same pattern of relatively unchanged prevalence rates. Only two countries showed continuous changes. A substantial increase was indicated in Cyprus (that did not have any data in 1999) whereas a decrease was observed in Lithuania.

Changes in perceived availability of drugs The perceived availability of different substances varies substantially between the ESPAD countries. The students were asked to indicate their opinion about how easy or difficult it would be for them to acquire any of the substances listed in the questionnaire.

Proportion of students who perceive inhalants “very easy” or “fairly easy” to obtain (Figures 26a–c) The proportion of students who indicated that inhalants would be “very easy” or “fairly easy” for them to obtain, increased in eleven countries between 1999 and 2003. These changes occurred in both low and high prevalence countries. They include Bulgaria, Croatia, the Czech Republic, Estonia, the Faroe Islands, Finland, Greenland, Ireland, Malta, Romania and the United Kingdom. Interestingly, a decrease was observed in more or less an equal number of countries (12). They include Cyprus, Denmark, Greece, Hungary, Iceland, Italy, Latvia, Lithuania, Norway, Portugal, Russia (Moscow) and Slovenia. Unchanged figures were thus only found in a small number of countries (6). Two of the top three countries in 1999 were also in this group in 2003 ( Ireland and Slovenia). However, Cyprus has moved down the order to be replaced by Finland. The trends from 1995 to 2003 are rather divergent. There are, however, indications of continuous increasing rise in this figure over time in Estonia,

Changes in the use of alcohol and other drugs 2003

the Faroe Islands, Finland and Malta. During the same period this figure has decreased in Norway.

Proportion of students who perceive cannabis “very easy” or “fairly easy” to obtain. (Figures 27a–c) An increasing proportion of the ESPAD students perceive cannabis to be easy to obtain. Increased proportions indicating “very easy” and “fairly easy” in relation to cannabis were found in fourteen countries, including Bulgaria, Croatian, the Czech Republic, Estonia, the Faroe Islands, Greenland, Latvia, Lithuania, Malta, Poland, Romania, the Slovak Republic, Slovenia, Romania and the United Kingdom. Decreased proportions were mainly found in Denmark, Greece and Norway, the outcome of which resulted in Denmark falling within the top group and both Greece and Norway from their position in the upper half of the table. From 1995 to 2003 there was a continuous increase in perceived availability of cannabis in seven ESPAD countries. They include Croatia, the Czech Republic, Estonia, Lithuania, Poland, the Slovak Republic and Slovenia, all of which are in the eastern parts of Europe. In five of these countries, the proportion doubled over the eight years. In Estonia the increase was threefold while in Lithuania it was six fold (starting from a low level). No country reported a continuous decrease between 1995, 1999 and 2003.

107

2003 50

Czech Rep. 40

nd la en re p. Re

30

tia oa Cr

Estonia

Slovenia Italy

Denmark Russia (Moscow) Ukraine

Bulgaria

20

France, U.K.

G

ak ov Sl

Figure 20a. Changes between 1999 and 2003 in lifetime experience of marijuana or hashish. Countries above the line have increased prevalence rates, and countries below have decreased. All students.

Ireland

Poland Hungary Latvia Portugal Lithuania Iceland Finland Malta Norway Faroe Isl. Sweden Greece Cyprus Romania

10

0 0

10

20

30

40

50 1999

Boys

Girls

25

40

Czech Rep. (22, 35, 44)

48 38

31

Ireland (37, 32, 39)

35

44 39 41

29

39 38

United Kingdom (41, 35, 38)

32

France (.., 35, 38)

32

35 38

42

35 14

16

Italy (19, 25, 27)

23 23

Greenland (.., 23, 27)

23 29

23 26

12

6

Slovak Rep. (9, 19, 27)

24 32

15 22

20 27

20 18

10

5

Estonia (7, 13, 23)

18 28

8 18

Russia (Moscow) (.., 22, 22)

25 26

20 18

13

5

Croatia (9, 16, 22)

18 24

13 20

20

9

Ukraine (14, 20, 21)

26 29

13 12

Bulgaria (.., 12, 21)

14 23

11 19

12

5

Poland (8, 14, 18)

19 23

10 13

Latvia (.., 17, 16)

22 20 5

18 12

8

Iceland (10, 15, 13)

14

13 11

2

1

Lithuania (1, 12, 13)

17 18

6 9

5

9 11 7 7 8

Malta (8, 7, 10)

7 13 7

5

Norway (6, 12, 9)

14 9 11 8 9 7

1995

10 9 11

Faroe Isl. (11, 7, 9)

6

Sweden (6, 8, 7)

5 6 6

1999

10

11 9

Greece (.., 9, 6)

11 7 7

2003

7 5

5 7 2 4

10

5

Finland (5, 10, 11)

10 11 10

20

7 12

18

2003

7 5

Portugal (7, 9, 15)

12

1999

4 13

9

1995

12 12

Hungary (4, 11, 16)

16 18

30

15

Denmark (17, 24, 23)

30

40

23 26

21 28 31

Figure 20b. Changes between 1995 and 2003 in lifetime experience of marijuana or hashish (values within brackets refer to all students 1995, 1999, 2003). Data sorted by all students 2003.

12

Slovenia (13, 25, 28)

27 31

108

30 40

42

% 50

18

0

Cyprus (5, 2, 4)

2 1 2

Romania (.., 1, 3)

1 2

0

10

20

30

40

50 %

Changes in the use of alcohol and other drugs 2003

Bulgaria

% 45

Finland

% 45

% 45

30

30

30

30

15

15

15

15

0

0

0

1995

1999

2003

Croatia

% 45

1995

1999

2003

France

% 45

0 1995

1999

2003

Latvia

% 45

1995

% 45

30

30

30

30

15

15

15

15

0

0 1995

% 45

1999

2003

0 1995

% 45

Cyprus

1999

2003

1999

2003

Lithuania

% 45

30

30

30

15

15

15

15

0 1995

% 45

1999

2003

Czech rep.

0 1995

% 45

1999

2003

1999

2003

% 45

Malta

30

30

30

15

15

15

15

0 1995

1999

2003

Denmark

% 45

0 1995

1999

2003

Hungary

% 45

1999

2003

Norway

% 45

30

15

15

15

15

0

% 45

2003

0 1995

% 45

Estonia

1999

2003

1999

2003

1995

% 45

Poland

30

30

30

30

15

15

15

15

0

0 1995

1999

2003

Faroe Isl.

% 45

0 1995

% 45

1999

2003

1999

2003

Portugal

% 45

30

30

30

15

15

15

15

0 1995

1999

2003

0 1995

1999

2003

2003

Ukraine

1995

30

0

1999

0 1995

% 45

Ireland

2003

0 1995

% 45

Iceland

1999

Sweden

% 45

30

1999

2003

Slovenia

1995

30

1995

1999

0 1995

30

0

2003

Slovak rep.

1995

30

0

1999

0 1995

% 45

Greenland

2003

Russia (Moscow)

1995

30

0

1999

0 1995

% 45

Greece

Romania

% 45

Italy

1999

2003

United Kingdom

0 1995

1999

2003

1995

1999

2003

Figure 20c. Changes between 1995 and 2003 in lifetime experience of marijuana or hashish, by country. All students.

Changes in the use of alcohol and other drugs 2003

109

2003 25

France 20

U.K. Czech Rep. Ireland Italy Slovenia

15

H

0 0

Slovak Rep. Denmark

ia ss nd la ia Ru Pooat al, a g i e Cr tu n in a r o a t vi Po s kr at way e E U L r ec No re G

Romania

5

Greenland

d ia a ni ar an lg el a ua c t I l Bu Lith a d M an s nl ru FiCyp

10

, ry ga un

Figure 21a. Changes between 1999 and 2003 in the proportion of all students who have used marijuana or hashish during the last 30 days. Countries above the line have increased prevalence rates, and countries below have decreased. All students.

Sw Faro eden e I sl.

5

10

15

20

25 1999

Boys

Girls

25 26

France (.., 22, 22)

29 23

8

13

16

17 13 7

5 11 14

Greenland (.., 10, 11)

12 12

8 11

5

1

Slovak Rep. (3, 6, 10)

8 10

5 9

8

4

Denmark (6, 8, 8)

11 10

6 5

4 10 10

Poland (3, 7, 8)

1

Croatia (3, 6, 8)

1

4 5

4

7 9

7

10 4

7 2 3 5

Russia (Moscow) (.., 5, 7)

5 7

5 6

Estonia (.., 5, 6)

7 8 1

3 4 1

Hungary (1, 4, 6)

5 7 1

Lithuania (0, 4, 6)

6 8 6

2 5 0 2 3 2

Ukraine (5, 5, 5)

7 8

3 2

Latvia (.., 5, 4)

8 5 5 5

3 2 3 3

Iceland (4, 4, 4)

4

4 3 3

1

Malta (2, 3, 4)

5

2 3

4 3

Greece (.., 4, 3)

2 1 3 3

2003

2 2 3 2 2 2 2 2

1

Cyprus (2, 1, 2)

1

Faroe Isl. (2, 1, 1)

1995

2 2

Finland (1, 2, 3)

1999

2 2

2003

0 1 1 1 1

Sweden (1, 2, 1)

3

3 0 1

Romania (.., 0, 1)

1 1

5

3 2

7

1999

2

Norway (3, 4, 3)

5

10

3

Portugal (4, 5, 7)

7 11

1995

5

Bulgaria (.., 4, 8)

5

15

12 12

Slovenia (5, 13, 14)

14 14

Figure 21b. Changes between 1995 and 2003 in the proportion of all students who have used marijuana or hashish during the last 30 days (values within brackets refer to all students 1995, 1999, 2003). Data sorted by all students 2003.

10

Italy (13, 14, 15)

17 19

20

12 11

Ireland (19, 15, 17)

18

110

6 17

25

25

15 16

Czech Rep. (7, 16, 19)

20 21

% 30

19 18 20

U.K. (24, 16, 20)

18

0

0 0

0

5

10

15

20

25

30 %

Changes in the use of alcohol and other drugs 2003

Bulgaria

% 30

Finland

% 30

% 30

20

20

20

20

10

10

10

10

0

0

0

1995

1999

2003

Croatia

% 30

1995

1999

2003

France

% 30

0 1995

1999

2003

Latvia

% 30

1995

% 30

20

20

20

20

10

10

10

10

0

0 1995

% 30

1999

2003

0 1995

% 30

Cyprus

1999

2003

1999

2003

Lithuania

% 30

20

20

20

10

10

10

10

0 1995

% 30

1999

2003

Czech rep.

0 1995

% 30

1999

2003

1999

2003

% 30

Malta

20

20

20

10

10

10

10

0 1995

1999

2003

Denmark

% 30

0 1995

1999

2003

Hungary

% 30

1999

2003

Norway

% 30

20

10

10

10

10

0

% 30

2003

0 1995

% 30

Estonia

1999

2003

1999

2003

1995

% 30

Poland

20

20

20

20

10

10

10

10

0

0 1995

1999

2003

Faroe Isl.

% 30

0 1995

% 30

1999

2003

1999

2003

Portugal

% 30

20

20

20

10

10

10

10

0 1995

1999

2003

0 1995

1999

2003

2003

Ukraine

1995

20

0

1999

0 1995

% 30

Ireland

2003

0 1995

% 30

Iceland

1999

Sweden

% 30

20

1999

2003

Slovenia

1995

20

1995

1999

0 1995

20

0

2003

Slovak rep.

1995

20

0

1999

0 1995

% 30

Greenland

2003

Russia (Moscow)

1995

20

0

1999

0 1995

% 30

Greece

Romania

% 30

Italy

1999

2003

United Kingdom

0 1995

1999

2003

1995

1999

2003

Figure 21c. Changes between 1995 and 2003 in the proportion of all students who have used marijuana or hashish during the last 30 days, by country. All students.

Changes in the use of alcohol and other drugs 2003

111

2003 15

12 Czech Rep. Estonia Portugal

9

Italy Lithuania

France 6

Poland

Hungary

Slovenia

Croatia ia ar lg d Bu lan e e c en e a in re Gre kr l. G U Is e d ro an Fa Finl s, ru yp

Iceland, Slovak Rep.

Denmark

Malta Swe den

3

U.K.

Ireland

Latvia Russia (Moscow)

Norway

C

Figure 22a. Changes between 1999 and 2003 in lifetime experience of any illicit drug other than marijuana or hashish. Countries above the line have increased prevalence rates, and countries below have decreased. All students.

0 0

3

Romania

6

9

12

15 1999

Boys

Girls 5

10 11

Czech Rep. (4, 9, 11) 3

23 9

8 10 6 7 6

5 9

8 6

2

1

Lithuania (2, 9, 7)

11 8 4

6 6 2

Portugal (3, 6, 7)

8 9

4 6

France (.., 5, 7)

6 8

5 7

3 9 7 6 6

5

Denmark (3, 7, 6)

3

Croatia (4, 6, 6)

3

5 5 5 6

5 5

3

Iceland (4, 5, 6)

7 3

4 5 1

Slovak Rep. (2, 5, 6)

6 6

5 5

Latvia (.., 11, 5)

12 5 3

10 4 2

Slovenia (3, 7, 5)

7 4 1

7 5 1

Hungary (1, 5, 5)

6 5

4 5

Russia (Moscow) (.., 9, 4)

7 5

10 4

Bulgaria (.., 5, 4)

5 5 3 2 4 4 2 3 2

1999

3

Norway (3, 6, 3)

2

Greece (.., 4, 3)

2 2

5

Sweden (2, 3, 3)

1

Cyprus (2, 2, 3)

1 1 1 1

1995

2 3

2003

1 1

Finland (1, 2, 3)

2

1999 2003 3 3

Romania (.., 9, 2)

9 2 2

9 1 1

Ukraine (1, 4, 2)

5 3 3 3

3 1 1

Faroe Isl. (2, 3, 2)

1

5

3

3

6

4 4 4

4 4 1 4

7

4

4

Malta (1, 3, 4)

3

1995

5

Greenland (.., 4, 4)

5

10

3

Poland (4, 11, 7)

15

112

12

Italy (8, 8, 8)

11

15

20 11

Ireland (16, 9, 9)

8 9 9

20

11 9

11

25

12 7

U.K. (22, 12, 9)

13 19

% 30

8

1

Estonia (2, 9, 10)

11 10

Figure 22b. Changes between 1995 and 2003 in lifetime experience of any illicit drug other than marijuana or hashish (values within brackets refer to all students 1995, 1999, 2003). Data sorted by all students 2003.

4

0

0

3 3

5

10

15

20

25

30 %

Changes in the use of alcohol and other drugs 2003

Bulgaria

% 30

Finland

% 30

% 30

20

20

20

20

10

10

10

10

0

0

0

1995

1999

2003

Croatia

% 30

1995

1999

2003

France

% 30

0 1995

1999

2003

Latvia

% 30

1995

% 30

20

20

20

20

10

10

10

10

0

0 1995

% 30

1999

2003

0 1995

% 30

Cyprus

1999

2003

1999

2003

Lithuania

% 30

20

20

20

10

10

10

10

0 1995

% 30

1999

2003

Czech rep.

0 1995

% 30

1999

2003

1999

2003

% 30

Malta

20

20

20

10

10

10

10

0 1995

1999

2003

Denmark

% 30

0 1995

1999

2003

Hungary

% 30

1999

2003

Norway

% 30

20

10

10

10

10

0

% 30

2003

0 1995

% 30

Estonia

1999

2003

1999

2003

1995

% 30

Poland

20

20

20

20

10

10

10

10

0

0 1995

1999

2003

Faroe Isl.

% 30

0 1995

% 30

1999

2003

1999

2003

Portugal

% 30

20

20

20

10

10

10

10

0 1995

1999

2003

0 1995

1999

2003

2003

Ukraine

1995

20

0

1999

0 1995

% 30

Ireland

2003

0 1995

% 30

Iceland

1999

Sweden

% 30

20

1999

2003

Slovenia

1995

20

1995

1999

0 1995

20

0

2003

Slovak rep.

1995

20

0

1999

0 1995

% 30

Greenland

2003

Russia (Moscow)

1995

20

0

1999

0 1995

% 30

Greece

Romania

% 30

Italy

1999

2003

United Kingdom

0 1995

1999

2003

1995

1999

2003

Figure 22c. Changes between 1995 and 2003 in lifetime experience of any illicit drug other than marijuana or hashish, by country. All students.

Changes in the use of alcohol and other drugs 2003

113

2003 20

Poland

16 Lithuania France 12 Czech Rep.

a ni to Es

4

Fi nl an d

Iceland

Ita ly

en ed a e ta l c Sw ni a ee Ma om , Gr ay R k rw ar m Isl. No via en at D roe ,L Fa nd la en re

8

Hungary

G

Figure 23a. Changes between 1999 and 2003 in lifetime experience of tranquillisers or sedatives without a doctor’s prescription. Countries above the line have increased prevalence rates, and countries below have decreased. All students.

Croatia Portugal, Slovenia

Slovak Rep. Russia (Moscow) Ireland

Ukraine

Bu lga ria Cyprus ,U .K. 4 8

0 0

12

16

20 1999

Boys

Girls

11

13

22 8 8

10

18

8

2 2

6

Finland (5, 6, 7)

3 4 6 6

9 9 11

Croatia (8, 8, 6)

4 8

9 9 15

Italy (11, 7, 6)

5 5 5 5 5

8 7 7

Sweden (6, 6, 6)

6 7

Romania (.., 5, 6)

4 3 8

7 7 8

Portugal (8, 8, 5)

6 4

9 7

5

10

Slovenia (8, 8, 5)

7 3

9 8

5 5 5

2 2

Faroe Isl. (4, 3, 5)

5 3 3

12

Denmark (11, 5, 4)

5 4

5 5

Greece (.., 5, 4)

5 3

5 5

Russia (Moscow) (.., 6, 3)

4 2

9 3

8

10

Malta (9, 5, 3)

5 2

5 3

Greenland (.., 3, 3)

3 3

2 4

2

3 3 3

Norway (3, 4, 3)

4 3

Latvia (.., 3, 3)

3 2 6 2

1995

4 4

1999

9

Ireland (7, 5, 2)

5

4 2

7 2

2003

10

U.K. (8, 4, 2)

6

3 1

Bulgaria (.., 4, 2)

3 2

4 2

3 3 3

3

Ukraine (3, 3, 2)

2 1

Cyprus (8, 6, 1)

6 1

5

9 5

9

1995

6

Slovak Rep. (4, 7, 4)

5

7

1 13

4

1999

2

Estonia (2, 2, 9)

5

114

13 13 10 10 10

Iceland (9, 10, 9)

8

10

11

Hungary (8, 10, 10)

9 10

15

21 14

5 7 7

2003

14 15 15

Czech Rep. (11, 18, 11)

8

20

17

France (.., 12, 13)

14

% 25

20

Lithuania (15, 12, 14)

10 10

Figure 23b. Changes between 1995 and 2003 in lifetime experience of tranquillisers or sedatives without a doctor’s prescription (values within brackets refer to all students 1995, 1999, 2003). Data sorted by all students 2003.

25 24

Poland (18, 18, 17)

12

5

9

1

0

0

5

10

15

20

25 %

Changes in the use of alcohol and other drugs 2003

Bulgaria

% 30

Finland

% 30

% 30

20

20

20

20

10

10

10

10

0

0

0

1995

1999

2003

Croatia

% 30

1995

1999

2003

France

% 30

0 1995

1999

2003

Latvia

% 30

1995

% 30

20

20

20

20

10

10

10

10

0

0 1995

% 30

1999

2003

0 1995

% 30

Cyprus

1999

2003

1999

2003

Lithuania

% 30

20

20

20

10

10

10

10

0 1995

% 30

1999

2003

Czech rep.

0 1995

% 30

1999

2003

1999

2003

% 30

Malta

20

20

20

10

10

10

10

0 1995

1999

2003

Denmark

% 30

0 1995

1999

2003

Hungary

% 30

1999

2003

Norway

% 30

20

10

10

10

10

0

% 30

2003

0 1995

% 30

Estonia

1999

2003

1999

2003

1995

% 30

Poland

20

20

20

20

10

10

10

10

0

0 1995

1999

2003

Faroe Isl.

% 30

0 1995

% 30

1999

2003

1999

2003

Portugal

% 30

20

20

20

10

10

10

10

0 1995

1999

2003

0 1995

1999

2003

2003

Ukraine

1995

20

0

1999

0 1995

% 30

Ireland

2003

0 1995

% 30

Iceland

1999

Sweden

% 30

20

1999

2003

Slovenia

1995

20

1995

1999

0 1995

20

0

2003

Slovak rep.

1995

20

0

1999

0 1995

% 30

Greenland

2003

Russia (Moscow)

1995

20

0

1999

0 1995

% 30

Greece

Romania

% 30

Italy

1999

2003

United Kingdom

0 1995

1999

2003

1995

1999

2003

Figure 23c. Changes between 1995 and 2003 in lifetime experience of tranquillisers or sedatives without a doctor’s prescription, by country. All students.

Changes in the use of alcohol and other drugs 2003

115

2003 20

15

Fi nl an d

Slovak Rep.

Czech Rep.

Hungary Faroe Isl.

10 Croatia, Poland Iceland

5

0 0

Malta

nd la Ire

Lithuania Estonia Latvia Po rt u ga l Ukraine Bu l Italy Roga ria ma nia Gr Cy Gree e ec pru nla e s nd

France

Sweden Denmark

U.K. Slovenia Norway

Russia (Moscow)

Figure 24a. Changes between 1999 and 2003 in lifetime experience of alcohol together with pills. Countries above the line have increased prevalence rates, and countries below have decreased. All students.

5

10

15

20 1999

Boys

Girls 5

7

Slovak Rep. (5, 11, 15)

11 7

7

8 13 13

Faroe Isl. (10, 11, 10)

4

12 16

10

15 14

Malta (13, 12, 9)

9 7

11

Ireland (.., 11, 9)

9 6 6

14 13 8

Poland (7, 10, 9)

8 6 6

12 11 7

Croatia (6, 10, 9)

9 7 12

11 12 24

Sweden (18, 14, 8)

9 5

18 12

Iceland (.., 10, 8)

8 6

13 11

9

16

Denmark (13, 15, 7)

11 6

19 8

14

25

U.K. (20, 11, 7)

9 6

13 8

France (.., 8, 7)

6 5

9 10

2

2

Lithuania (2, 7, 7)

7 6 6

6 8 8

Slovenia (7, 9, 6)

7 4

12 9

Latvia (.., 7, 6)

7 5

7 7

Russia (Moscow) (.., 7, 6)

5 6

9 6

Estonia (.., 4, 6)

3 4 3

3 3

Bulgaria (.., 4, 4)

3

Ukraine (4, 3, 4)

3 3

1995

4 2

4

1999

4 4

2003

7

Italy (6, 3, 3)

3

3 2

Greece (.., 4, 2)

4 2

4 3

Greenland (.., 2, 2)

3 2

1 2

Cyprus (5, 3, 0)

4 1

5

5 5

Romania (.., 4, 3)

5

5

6 4

5

4

5

Portugal (5, 5, 4)

3

1999

10 6

4 4

4

12

Norway (9, 8, 5)

6

1995

5 8

7

116

11

Hungary (10, 8, 11)

7 8

10

19 18

9

15

25

Finland (17, 13, 12)

5 9

20

19

15

7

2003

10

Czech Rep. (9, 14, 12)

11

% 25

13 18

8 9

Figure 24b. Changes between 1995 and 2003 in lifetime experience of alcohol together with pills (values within brackets refer to all students 1995, 1999, 2003). Data sorted by all students 2003.

5

2

4

0

0

0

5

10

15

20

25 %

Changes in the use of alcohol and other drugs 2003

Bulgaria

% 30

Finland

% 30

% 30

20

20

20

20

10

10

10

10

0

0

0

1995

1999

2003

Croatia

% 30

1995

1999

2003

France

% 30

0 1995

1999

2003

Latvia

% 30

1995

% 30

20

20

20

20

10

10

10

10

0

0 1995

% 30

1999

2003

0 1995

% 30

Cyprus

1999

2003

1999

2003

Lithuania

% 30

20

20

20

10

10

10

10

0 1995

% 30

1999

2003

Czech rep.

0 1995

% 30

1999

2003

1999

2003

% 30

Malta

20

20

20

10

10

10

10

0 1995

1999

2003

Denmark

% 30

0 1995

1999

2003

Hungary

% 30

1999

2003

Norway

% 30

20

10

10

10

10

0

% 30

2003

0 1995

% 30

Estonia

1999

2003

1999

2003

1995

% 30

Poland

20

20

20

20

10

10

10

10

0

0 1995

1999

2003

Faroe Isl.

% 30

0 1995

% 30

1999

2003

1999

2003

Portugal

% 30

20

20

20

10

10

10

10

0 1995

1999

2003

0 1995

1999

2003

2003

Ukraine

1995

20

0

1999

0 1995

% 30

Ireland

2003

0 1995

% 30

Iceland

1999

Sweden

% 30

20

1999

2003

Slovenia

1995

20

1995

1999

0 1995

20

0

2003

Slovak rep.

1995

20

0

1999

0 1995

% 30

Greenland

2003

Russia (Moscow)

1995

20

0

1999

0 1995

% 30

Greece

Romania

% 30

Italy

1999

2003

United Kingdom

0 1995

1999

2003

1995

1999

2003

Figure 24c. Changes between 1995 and 2003 in lifetime experience of alcohol together with pills, by country. All students.

Changes in the use of alcohol and other drugs 2003

117

2003 25

Greenland 20 Ireland ark , Denm

15

Fa roe Isl. Re p., Slo v 10 Por tug Finla ak R ep nd al . Cz

ec h

Iceland

ia

Hungary Bulgaria Romania

U.K. France

Latvia Italy 5

Malta

Greece, Slovenia Croatia

Eston

Figure 25a. Changes between 1999 and 2003 in lifetime experience of inhalants. Countries above the line have increased prevalence rates, and countries below have decreased. All students.

No rw ay

Poland Sweden Russia (Moscow) Ukraine Lithuania

0 0

5

10

15

20

25 1999

Boys

Girls Greenland (.., 19, 22)

21 23

Ireland (.., 22, 18)

14 3

15 14

10

Slovenia (12, 14, 15)

15 15

13 15

Greece (.., 14, 15)

18 17

12 13

13 14

12 14

20 12

17 13

6

10

Iceland (8, 11, 12)

13 12

8 11

France (.., 11, 11)

12 12 12

9 10 4

Faroe Isl. (8, 5, 11)

7 10

3 13

11

8

Poland (9, 9, 9)

10 10

7 8

8 8

7

Czech Rep. (8, 7, 9)

9

6 9

8 8

5

Slovak Rep. (6, 7, 9)

10

6 7

15

9

Sweden (12, 8, 8)

9 8 8 8

8 8 7

Estonia (8, 7, 8)

9

6 7

6

6

Denmark (6, 7, 8)

7 9

8 7

5 5

4

Finland (4, 5, 8)

8

6 8

4 4

2

Portugal (3, 3, 8)

10

3 6

Russia (Moscow) (.., 9, 7)

11

1995

7

1999

7

Latvia (.., 6, 7)

1995 1999

4 7

7

4

Ukraine (5, 8, 6)

9 9 9

2003

7 4 6

Italy (8, 6, 6)

7 8 18

5 5 14

Lithuania (16, 10, 5)

13 6

6 4

7

7

Norway (7, 6, 5)

6 6

5 4

7

5

Hungary (6, 4, 5)

6 6

3 4

Bulgaria (.., 3, 3)

4 4

5

2 3

Romania (.., 1, 2)

2 2

10

8 6

8

2003

21

United Kingdom (20, 15, 12)

14

15

14

Croatia (13, 13, 14)

15

Figure 25b. Changes between 1995 and 2003 in lifetime experience of inhalants (values within brackets refer to all students 1995, 1999, 2003). Data sorted by all students 2003.

17 17

Malta (17, 16, 16)

15 16

118

1 15

17

20

21 21

Cyprus (3, .., 18)

19

% 25

17 22

22

0

1 1

0

5

10

15

20

25 %

Changes in the use of alcohol and other drugs 2003

Bulgaria

% 30

Finland

% 30

% 30

20

20

20

20

10

10

10

10

0

0

0

1995

1999

2003

Croatia

% 30

1995

1999

2003

France

% 30

0 1995

1999

2003

Latvia

% 30

1995

% 30

20

20

20

20

10

10

10

10

0

0 1995

% 30

1999

2003

0 1995

% 30

Cyprus

1999

2003

1999

2003

Lithuania

% 30

20

20

20

10

10

10

10

0 1995

% 30

1999

2003

Czech rep.

0 1995

% 30

1999

2003

1999

2003

% 30

Malta

20

20

20

10

10

10

10

0 1995

1999

2003

Denmark

% 30

0 1995

1999

2003

Hungary

% 30

1999

2003

Norway

% 30

20

10

10

10

10

0

% 30

2003

0 1995

% 30

Estonia

1999

2003

1999

2003

1995

% 30

Poland

20

20

20

20

10

10

10

10

0

0 1995

1999

2003

Faroe Isl.

% 30

0 1995

% 30

1999

2003

1999

2003

Portugal

% 30

20

20

20

10

10

10

10

0 1995

1999

2003

0 1995

1999

2003

2003

Ukraine

1995

20

0

1999

0 1995

% 30

Ireland

2003

0 1995

% 30

Iceland

1999

Sweden

% 30

20

1999

2003

Slovenia

1995

20

1995

1999

0 1995

20

0

2003

Slovak rep.

1995

20

0

1999

0 1995

% 30

Greenland

2003

Russia (Moscow)

1995

20

0

1999

0 1995

% 30

Greece

Romania

% 30

Italy

1999

2003

United Kingdom

0 1995

1999

2003

1995

1999

2003

Figure 25c. Changes between 1995 and 2003 in lifetime experience of inhalants, by country. All students.

Changes in the use of alcohol and other drugs 2003

119

2003 80 Ireland

Slovenia

. .K U l. Is

e ro Fa

60

20

Lithuania

Greenland Ukraine Romania

Cyprus Iceland

ry ga un ay H rw No ) w co p. os Re (M a ak a si tvi ov Sl us La R

Estonia Bulgaria

Finland Denmark

e ec re G den Swe

Malta

Croatia Czech Rep. Poland

ce an Fr

40

l ga rtu Po

ly Ita

Figure 26a. Changes between 1999 and 2003 in the proportion of all students who perceive inhalants “very easy” or “fairly easy” to obtain. Countries above the line have increased prevalence rates, and countries below have decreased. All students.

0 0

20

40

60

80 1999

Boys 78

Girls Ireland (80, 63, 77)

59

83 65

59

53

65 51

57 60

58 55 52 61 56 51

Denmark (57, 57, 52)

52 60

Sweden (61, 54, 51)

53 51 26

28

Faroe Isl. (27, 44, 51)

44 47 42 42 47 38

44 54 43 44

Croatia (42, 43, 50)

53 25

Czech Rep. (32, 47, 47)

55 54

39 42

Greece (.., 58, 46)

64 47 55

53 45 47

Iceland (51, 61, 45)

65 45 45 48 44

57 44 39

Poland (42, 46, 44) 12

45 44 12

Malta (12, 28, 39)

28 36

28 42

49 51

53 50

Hungary (51, 50, 37)

37

38

65

54

Norway (59, 49, 37)

52 38

46 35

34

21 24

Estonia (19, 24, 35)

22

Bulgaria (.., 21, 32)

36

23 21 34

Russia (Moscow) (.., 35, 30)

30 31

36 30 26 32 29

Lithuania (28, 34, 28)

35 28

Greenland (.., 22, 27)

25 31

1995

19 22

Latvia (.., 37, 26)

38 24

1999

36 28

10

20 16 15 27 24

Portugal (28, 26, 15)

16

13

Romania (.., 5, 13)

8 15

20

14 16

Italy (21, 20, 16)

30 29

2003

5

Ukraine (8, 17, 19)

20 22 21 25 17

40

33 36 16 35

30

2003

31 31

France (.., 33, 36)

34 36

1995

22

Slovak Rep. (29, 37, 37)

44 44

34

120

12

Cyprus (14, 65, 52)

65

60

68 49 55

16

80

54

U.K. (68, 48, 53)

48 51

1999

68 65 46

Finland (48, 53, 58)

56 68

Figure 26b. Changes between 1995 and 2003 in the proportion of all students who perceive inhalants “very easy” or “fairly easy” to obtain (values within brackets refer to all students 1995, 1999, 2003). Data sorted by all students 2003.

58

Slovenia (59, 66, 61)

57 50 51

% 100

82

66 71

60

4 12

0

0

20

40

60

80

100 %

Changes in the use of alcohol and other drugs 2003

Bulgaria

% 80

Finland

% 80

% 80

60

60

60

60

40

40

40

40

20

20

20

20

0

0

0

1995

1999

2003

Croatia

% 80

1995

1999

2003

France

% 80

0 1995

1999

2003

Latvia

% 80

1995

% 80

60

60

60

60

40

40

40

40

20

20

20

20

0

0

0

1995

% 80

1999

2003

1995

% 80

Cyprus

1999

2003

1999

2003

Lithuania

% 80

60

60

60

40

40

40

40

20

20

20

20

0

0

0

% 80

1999

2003

Czech rep.

1995

% 80

1999

2003

1999

2003

% 80

Malta

60

60

60

40

40

40

40

20

20

20

20

0

0

0

1999

2003

Denmark

% 80

1995

1999

2003

Hungary

% 80

1999

2003

Norway

% 80 60

60

40

40

40

20

20

20

20

0

0

0

% 80

1995

% 80

Estonia

1999

2003

1999

2003

1995

% 80

Poland

60

60

60

60

40

40

40

40

20

20

20

20

0

0

0

1995

1999

2003

Faroe Isl.

% 80

1995

% 80

1999

2003

1999

2003

Portugal

% 80

60

60

60

40

40

40

40

20

20

20

20

0

0

0

1999

2003

1995

1999

2003

2003

Ukraine

1995

60

1995

1999

0 1995

% 80

Ireland

2003

0 1995

% 80

Iceland

1999

Sweden

% 80

60

2003

2003

Slovenia

1995

40

1999

1999

0 1995

60

1995

2003

Slovak rep.

1995

60

1995

1999

0 1995

% 80

Greenland

2003

Russia (Moscow)

1995

60

1995

1999

0 1995

% 80

Greece

Romania

% 80

Italy

1999

2003

United Kingdom

0 1995

1999

2003

1995

1999

2003

Figure 26c. Changes between 1995 and 2003 in the proportion of all students who perceive inhalants “very easy” or “fairly easy” to obtain, by country.

Changes in the use of alcohol and other drugs 2003

121

2003 60

40

Ireland

Croatia

Italy Poland

Bulgaria

Iceland Portugal

) w co os (M a si d us R an nl Fi Hungary a ni Faroe Isl. to ia v Es Lat a ni ua d th n Li nla a t ne l e ai re a kr G M U

Figure 27a. Changes between 1999 and 2003 in the proportion of all students who perceive marijuana or hashish “very easy” or “fairly easy” to obtain. Countries above the line have increased prevalence rates, and countries below have decreased. All students.

U.K. Czech Rep. Slovenia Denmark Slovak Rep. France

Norway

Sweden

20

Greece

Cyprus

Romania

0 0

20

40

60 1999

Boys 65

Girls

31

Czech Rep. (35, 50, 58)

60

48 56

29 57 48

44

Denmark (46, 57, 52)

53 27

53 50 21

Slovak Rep. (24, 40, 49)

46 56

35 43

France (.., 44, 47)

47 53 20

42 42 17

Croatia (19, 29, 45)

30 44

27 46

31

32

Italy (32, 43, 44)

50 48 22

38 40 15

Poland (18, 30, 37)

34 39

26 35

30

38

35

25

Norway (25, 38, 26)

25

39 27

Russia (Moscow) (.., 22, 24)

23 25 25 25 22

21 23 25 27 25

Sweden (25, 26, 23) 11

5

Estonia (8, 19, 23)

25 26

15 20

Latvia (.., 18, 22)

21 22

16 22

Greece (.., 33, 20)

38 21

28 19

11

8

Hungary (9, 19, 20)

23 21 5

16 19 2

Lithuania (3, 15, 20)

18 21

12 19

Greenland (.., 13, 20)

13 25 9 10

Finland (14, 20, 19)

20 17 19 16 20

Faroe Isl. (18, 17, 17) 7

10 11 14

Cyprus (9, 9, 12)

12

1999 2003

8 9 8 7 10

Romania (.., 5, 11)

7

1995

3

Ukraine (5, 11, 13)

14 18

20

13 16 11 11 19 18 21 20 18 18 14

Malta (10, 11, 20)

12 21

40

23 22 25 26

Portugal (25, 26, 29)

37

2003

21 37

27 30 34

1999

37 39

Bulgaria (.., 22, 36)

24

1995

25

Iceland (27, 38, 36)

34

60

45 53

60

Figure 27b. Changes between 1995 and 2003 in the proportion of all students who perceive marijuana or hashish “very easy” or “fairly easy” to obtain (values within brackets refer to all students 1995, 1999, 2003). Data sorted by all students 2003.

25

Slovenia (27, 47, 55)

49

122

54 51 54

U.K. (56, 52, 58) 37

53

% 80

60 60 60

Ireland (62, 59, 60)

57 60 58 54 61

5 10

0

0

20

40

60

80 %

Changes in the use of alcohol and other drugs 2003

Bulgaria

% 80

Finland

% 80

% 80

60

60

60

60

40

40

40

40

20

20

20

20

0

0

0

1995

1999

2003

Croatia

% 80

1995

1999

2003

France

% 80

0 1995

1999

2003

Latvia

% 80

1995

% 80

60

60

60

60

40

40

40

40

20

20

20

20

0

0

0

1995

% 80

1999

2003

1995

% 80

Cyprus

1999

2003

1999

2003

Lithuania

% 80

60

60

60

40

40

40

40

20

20

20

20

0

0

0

% 80

1999

2003

Czech rep.

1995

% 80

1999

2003

1999

2003

% 80

Malta

60

60

60

40

40

40

40

20

20

20

20

0

0

0

1999

2003

Denmark

% 80

1995

1999

2003

Hungary

% 80

1999

2003

Norway

% 80 60

60

40

40

40

20

20

20

20

0

0

0

% 80

1995

% 80

Estonia

1999

2003

1999

2003

1995

% 80

Poland

60

60

60

60

40

40

40

40

20

20

20

20

0

0

0

1995

1999

2003

Faroe Isl.

% 90 75 60 45 30 15 0 1995

1999

1995

% 80

2003

1999

2003

1999

2003

Portugal

% 80

60

60

40

40

40

20

20

20

0

0 1999

2003

2003

Ukraine

1995

60

1995

1999

0 1995

% 80

Ireland

2003

0 1995

% 80

Iceland

1999

Sweden

% 80

60

2003

2003

Slovenia

1995

40

1999

1999

0 1995

60

1995

2003

Slovak rep.

1995

60

1995

1999

0 1995

% 80

Greenland

2003

Russia (Moscow)

1995

60

1995

1999

0 1995

% 80

Greece

Romania

% 80

Italy

1999

2003

United Kingdom

0 1995

1999

2003

1995

1999

2003

Figure 27c. Changes between 1995 and 2003 in the proportion of all students who perceive marijuana or hashish “very easy” or "fairly easy” to obtain, by country.

Changes in the use of alcohol and other drugs 2003

123

Proportion of students who perceive LSD or other hallucinogens “very easy” or “fairly easy” to obtain (Figure 28a–c) The proportion of students who perceive LSD or other hallucinogens easy to obtain differ substantially between countries. In some countries only very few students think so, while about one fifth of the students in the top prevalence countries think that it would be “very easy” or “fairly easy” to obtain LSD or other hallucinogens. In most countries the figures for 2003 were more or less equivalent to those reported in 1999. Increases were mainly found in Bulgaria and Croatia, while seven countries reported a decrease. They include Denmark, Greece, Hungary, Ireland, Nor-

way, Russia (Moscow), Slovenia and the United Kingdom. The most pronounced decrease was found in Ireland that resulted in the accompanying drop down the table from its previous top position in 1999. In 2003 Poland was in the top group together with one of the countries with a clear increase, Croatia. From 1995 to 2003 a continuous increasing in the proportion of students thinking that LSD or other hallucinogens are easy to obtain were mainly found in Bulgaria and Croatia. A continuous downward trend was observed in Ireland and the United Kingdom. In both of these countries the figures have approximately halved, from 43% in 1995 to 17% in 2003.

Summary The prevalence figures for smoking provide a relative stable trend pattern in a majority of the ESPAD countries. Increasing figures were mainly found in countries in the eastern parts of Europe and decreasing figures in the western parts. The more recent smokers, those who had smoked during the last 30 days, only increased in a few countries. All but one of the twelve countries with decreasing figures is found in the northern, western and southern parts of Europe. Still, however, proportions varying between 20 and 50% had smoked during the last 30 days in the ESPAD countries. The proportions reporting daily smoking at the age of 13 were also relatively unchanged between 1999 and 2003 in a large majority of the countries. Only a few countries reported increased figures (Estonia, the Faroe Islands and Latvia) on this variable. Only a minority of the ESPAD students have drank alcohol as many times as 40 or more. In almost half of the countries only a quarter of the students report such behaviour and in no country has the proportion exceeded 50%. There was, however, a clear increase in the prevalence rates, especially in countries from the eastern part of Europe. It is clear that alcohol consumption among students in most cases is a relatively new experience. The proportions answering that they had consumed alcohol 20 times or more over the last 12 months were not very unlike the rates reported for lifetime prevalence. Accordingly, nearly all countries with

increasing figures were found in the eastern part of Europe. Much lower proportions have drank alcohol 10 times or more during the last 30 days. A very large majority show rather unchanged figures. However, four of the five countries with increases are found in the eastern part of Europe. For all these variables concerning prevalence rates of alcohol consumption, the increases in the eastern parts of Europe do result in them moving up the ladder to occupy the top positions for these behaviours. However, for most variables Malta, the United Kingdom, Ireland, Denmark, the Czech Republic and Greece are still the top ranked countries. Some students drink alcohol rather frequently. A comparison between students that had drank 3 times or more during the last 30 days shows that spirits was the beverage of choice in 2003. The figure for spirits (12 countries) was about twice as high as it was for beer or wine (6–7). A decreased proportion reporting such a high drinking frequency was mainly reported for beer (5 countries). In Denmark and France the figures dropped between 1999 and 2003 for all three beverages. For beer most of the increases occurred in the eastern parts of Europe, while for wine it was most apparent in countries that traditionally are viewed as wine countries like Croatia, Cyprus, Greece, Hungary and Italy. The increase in the frequent consumption of spirits is on the other hand more diverse from a geographical point of view.

124

Changes in the use of alcohol and other drugs 2003

There are more countries that report a reduced rather than an increased number of students that drank large quantities on the last drinking occasion for beer (>100 cl) and wine (>15 cl). For spirits (>10 cl) the number of countries showing an increase and a decrease was more or less the same. Four countries (Denmark, Ireland, Norway and the United Kingdom) reported rather remarkable increases in the proportion of students that drank more than 100 cl of alcopops at the last drinking occasion, while the remaining countries in which this beverage is sold remained at rather low levels. The proportion of students who had been drunk 20 times or more in a lifetime increased mainly in some countries in the eastern part of Europe. However, this also occurred in some countries in the western part that in turn were already high prevalence countries in 1999, like Ireland and the Faroe Islands. Moreover, the same pattern was found in relation to the 12 months prevalence rates. The pattern of the prevalence rates for being drunk 3 times or more during the last 30 days were rather unchanged in a large majority of the ESPAD countries. This is suggestive of the fact that the prevalence rates have remained low in the Mediterranean countries, e.g. Greece, France, Italy, Malta and Portugal, but also in Romania. Much higher figures were recorded for Denmark, Finland, Greenland, Ireland and the United Kingdom, which in the main reinforces the apparent accepted pattern of frequent intoxication in the north but much less so in the southern part of Europe. The figures for binge drinking 3 times or more often during the last 30 days changed more than the drunkenness figures. Six of the nine countries with an increased proportion were found in the eastern parts of Europe. The lifetime prevalence for any illicit drug use was mainly stable in about half of the countries. The increases that occurred in nine countries were geographically rather spread, without any clear pattern. The top countries in 2003 are still those that were in the same position in 1999 (the Czech Republic, France, Ireland and the United Kingdom), but further increases were noted in two of them; the Czech Republic and Ireland. Changes in the prevalence rates of cannabis use were very similar to the changed rates for any illicit drug use. The use of other kinds of drugs is not that prevalent and the number of countries that have changed are rather few. Four countries reported lower figures in 2003 (Latvia, Poland, Russia (Moscow) and

Romania). The former top countries in 1999 are still on top in 2003, i.e. the Czech Republic, Estonia, Ireland, Italy and the United Kingdom. Tranquillisers or sedatives are mostly used in the Czech Republic (although a decrease was noted between 1999 and 2003), France, Lithuania and Poland, and the prevalence rates have not changed very much. Only one country, Estonia, showed an increased prevalence rate between 1999 and 2003. Taking pills together with alcohol is about as common in terms of prevalence rates as those for tranquillisers or sedatives, and the proportions were about the same in the two last surveys. However, the top country in this respect in 2003, the Slovak Republic, showed an increase. Other countries at the top were Finland, Hungary and the Czech Republic. The use of inhalants is rather spread geographically and the proportions indulging in such behaviour vary between 2 and 22%. Only two countries reported increased prevalence rates from 1999 to 2003 (the Faroe Islands and Portugal) and two decreased rates (Ireland and Lithuania). The top country for this variable is Greenland, which remained so together with Ireland and Malta. Figures were missing for Cyprus in 1999 but following the latest survey the country joined the top group of countries. Even though the proportion of students that had ever tried inhalants were rather unchanged the perceived availability changed in a large number of countries. In about 10, the students reported an increased availability and in about the same number of countries the figures dropped, without any clear geographical pattern. An increased availability of cannabis between 1999 and 2003 was reported from more than half of the countries while only three changed in the opposite direction (Denmark, Norway and Greece). The large number of countries reporting an increase was spread all over Europe. However, if looking only at countries with more extensive increases (8 percentage points or more) six of the seven are found in the eastern parts of Europe. The perceived availability of LSD or other hallucinogens is unchanged in a majority of the countries. However, there are more countries (8) that report a decreased availability than an increased (Croatia and Bulgaria). The decreases were remarkable in Ireland and the United Kingdom, especially if one also includes the 1995 exercise. Between 1995 and 2003 the proportion answering “very easy” or “fairly easy” dropped in both countries from

Changes in the use of alcohol and other drugs 2003

125

2003 30

25

Croatia

Poland

20 U.K.

Slovenia

Czech Rep.

Italy

Iceland Hungary Norway Greece

ce an Fr

Ru ss ia (

Mo sc ow )

G

d l. an Is nl Fi ta oe al ar M F e n ai kr U nd la en re

5

Latvia

10

Ireland

Denmark

S we de n

Slovak Rep. Bulgaria Estonia Portugal Lithuania

15

s ru yp C ia an om R

Figure 28a. Changes between 1999 and 2003 in the proportion of all students who perceive LSD or other hallucinogens “very easy” or “fairly easy” to obtain. Countries above the line have increased prevalence rates, and countries below have decreased. All students.

0 0

5

10

15

20

25

30 1999

Boys

Girls 13

22 22

14

21

14 18 18 17

20 18

16

43 33 16

12

12

Denmark (12, 20, 16)

21 18

19 15

7

6

Slovak Rep. (7, 13, 15)

15 17

12 13

Bulgaria (.., 9, 14)

10 13 14 15 12 13 14 14

9 15 11

Iceland (12, 16, 13)

18 14 13

Sweden (13, 15, 13)

16 13

3

2

Estonia (3, 14, 13)

17 14 12

12 12 10

Norway (11, 16, 12)

16 13 11

16 11 12 12 11

Portugal (12, 13, 12)

15 14 7

5

Hungary (6, 16, 11)

17 11

15 11

16

17

Italy (17, 13, 11)

21 13

8 10

Latvia (.., 11, 11)

14 12 3

9 10 1

Lithuania (2, 8, 11)

9 12

6 10

Greece (.., 16, 10)

19 12

14 9

Russia (Moscow) (.., 14, 10)

16 11

12 9

France (.., 9, 8)

10 9 4 4 6 6 5 6 7 8

2003

Malta (6, 6, 6) Cyprus (6, 5, 6) Romania (.., 3, 5)

5 6 6 4 4 4 4 5

2003

8

Ukraine (3, 4, 4)

0

1999

4 2 3 3

Greenland (.., 1, 4)

5

1995

3 4 2

Faroe Isl. (4, 6, 4)

1

10

7 7 7 8 7 6 5 7 6 4 5

Finland (5, 7, 6)

6

126

12

Ireland (43, 30, 16)

27

20

42

Czech Rep. (12, 20, 17)

21 44

1999

24 23 18

12

1995

22

11

United Kingdom (42, 23, 18)

24

30

18 15

19

42

40

9

Slovenia (12, 24, 18)

25

% 50

Croatia (9, 14, 21)

21 10

Figure 28b. Changes between 1995 and 2003 in the proportion of all students who perceive LSD or other hallucinogens “very easy” or “fairly easy” to obtain (values within brackets refer to all students 1995, 1999, 2003). Data sorted by all students 2003.

Poland (11, 20, 21)

9

2 3

0

10

20

30

40

50 %

Changes in the use of alcohol and other drugs 2003

Bulgaria

% 45

Finland

% 45

% 45

30

30

30

30

15

15

15

15

0

0

0

1995

1999

2003

Croatia

% 45

1995

1999

2003

France

% 45

0 1995

1999

2003

Latvia

% 45

1995

% 45

30

30

30

30

15

15

15

15

0

0 1995

% 45

1999

2003

0 1995

% 45

Cyprus

1999

2003

1999

2003

Lithuania

% 45

30

30

30

15

15

15

15

0 1995

% 45

1999

2003

Czech rep.

0 1995

% 45

1999

2003

1999

2003

% 45

Malta

30

30

30

15

15

15

15

0 1995

1999

2003

Denmark

% 45

0 1995

1999

2003

Hungary

% 45

1999

2003

Norway

% 45

30

15

15

15

15

0

% 45

2003

0 1995

% 45

Estonia

1999

2003

1999

2003

1995

% 45

Poland

30

30

30

30

15

15

15

15

0

0 1995

1999

2003

Faroe Isl.

% 45

0 1995

% 45

1999

2003

1999

2003

Portugal

% 45

30

30

30

15

15

15

15

0 1995

1999

2003

0 1995

1999

2003

2003

Ukraine

1995

30

0

1999

0 1995

% 45

Ireland

2003

0 1995

% 45

Iceland

1999

Sweden

% 45

30

1999

2003

Slovenia

1995

30

1995

1999

0 1995

30

0

2003

Slovak rep.

1995

30

0

1999

0 1995

% 45

Greenland

2003

Russia (Moscow)

1995

30

0

1999

0 1995

% 45

Greece

Romania

% 45

Italy

1999

2003

United Kingdom

0 1995

1999

2003

1995

1999

2003

Figure 28c. Changes between 1995 and 2003 in the proportion of all students who perceive LSD or other hallucinogens “very easy” or “fairly easy” to obtain, by country.

Changes in the use of alcohol and other drugs 2003

127

about 43% to about 17%. To sum up, the trend development over the 8 years of the ESPAD history is indicative of the fact that smoking remains at about the same level or decreased in a majority of the countries. With regard to alcohol an unchanged or a somewhat decreasing consumption was observed in the western parts of Europe while increases mainly were found

in the eastern parts. The use of drugs is still dominated by the use of cannabis. The high prevalence countries in 1999 are still at the top in 2003, but a clear increasing tendency can be observed in the eastern parts of Europe. It is also clear that an increasing number of students in many European countries find cannabis easily available.

128

Changes in the use of alcohol and other drugs 2003

The alcohol and drug situation 2003

This chapter presents the results of the 2003 ESPAD survey, mainly following the same structure as in the two earlier reports from 1995 and 1999. Each variable is presented with reference to the relevant table in the table section (Appendix II). In addition, the results of many of the variables are illustrated by a map, a bar graph by sex and a graph describing the changes from 1995 to 2003. In the maps the prevalence rates of each variable have been divided into five groups. The cut-off points for the intervals have been chosen to fit the emerging pattern, with the aim of giving a picture as comprehensive as possible. Thus, the maps show the differences in prevalence rates over the countries for all students, while in the bar graphs the variables are presented by sex. The order of appearance in the bar graphs is determined by the results for all students (the figure within brackets). However, the differences between countries are sometimes very small. When available, corresponding figures from USA and Spain are presented in tables, maps and bar graphs. The American figures origins from the “Monitoring the Future” study in Michigan, from which many of the ESPAD questions originally are taken. It ought to be observed that data from USA relates to students in grade 10, in which the large majority, but not all students, were born in 1987. The Spanish data are from a national survey in 2002 and calculated for the same agegroup as the ESPAD target group, i.e. students born in 1987. Since Spain and USA are not parts of the ESPAD project and data not collected with the same proto-

col, their results are not fully comparable as data are between the ESPAD countries. To show this, data from these two countries are presented below the bottom line in the tables and marked differently in the maps and graphs. In some few cases also an ESPAD country is found below the line in a table. It happens if the formulation of a question or the answering categories differ so much from the standardised ESPAD questionnaire that the results are judged not to be fully comparable. The first part of the chapter deals with tobacco use, followed by a section on alcohol consumption, including prevalence rates of consumption as well as drunkenness and binge drinking. The alcohol section also includes findings from some related variables like expected consequences of alcohol consumption, risk perception etc. The third part presents prevalence rates of illicit drug use, use of inhalants and tranquillisers or sedatives, with and without a doctor’s prescription, onset of drug use and the students’ perception of the availability of drugs. The students’ views of possible drug use among friends and siblings are also included. In the maps, a few of the smallest countries (islands) have been enlarged. This has been done to entrance the visibility of the 5-colour division of the countries into different prevalence groups. In the tables a zero represents a value ranging from 0.1 to 0.4. Values ranging from 0.5 to 0.9 are rounded to 1. The mark “–” means that no student has given that answer, while “..” means that data are not comparable or available.

Tobacco use In this section the lifetime prevalence rates of smoking cigarettes, the rates of smoking 40 times or more during lifetime, the last 30 days prevalence rates and the prevalence of daily smoking at the age of 13 are presented.

The alcohol and drug situation 2003

Lifetime use of cigarettes (Tables 1a–c, figures 29a–b) In nearly all the ESPAD countries 50–80% of the students had smoked cigarettes, at least once in their lifetime. The highest lifetime prevalence rates of smoking cigarettes were found in the Faroe

129

; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ;

19 % 20 24 % 25 30 % 31 35 % 36

; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ;

; ; ; ; ; ; ; ; ; ;

% ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ;

Data uncertain or not available Non-participating country

; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ;

; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ;

; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ;

; ; ; ; ; ; ; ; ; ; ; ; ; ;

; ; ; ; ; ; ;

Figure 29a. Lifetime use of cigarettes 40 times or more. Percentages among all students. 2003. Germany and Turkey: Limited geographical coverage.

Boys

Girls

41 34 39 40 42 39 49 32 41 32 39 35 33 32 38 28 26 25 28 26 23 32 25 24 15 20 19 19 26 27 19 19 17 17 % 50

40

30

20

44

Austria (42) Greenland (42) Faroe Isl. (41) Germany (6 Bundesl.) (40) Russia (Moscow) (40) Czech Rep. (39) Lithuania (39) Bulgaria (35) Estonia (35) Finland (32) Latvia (32) Slovak Rep. (32) Hungary (31) Croatia (30) Ukraine (28) Belgium (27) Denmark (27) Ireland (27) Netherlands (27) Slovenia (27) Norway (26) Poland (26) Italy (25) Switzerland (24) Isle of Man (22) Sweden (22) United Kingdom (22) Greece (20) Romania (20) Cyprus (18) Iceland (18) Portugal (18) Malta (16) Turkey (6 cities) (13) 10

0

49 42 39 38 38 28 37 29 32 25 30 30 29 19 26 28 29 26 28 29 21 25 24 28 24 24 21 15 11 17 17 16 7 0

20

40

60

80 %

Figure 29b. Lifetime use of cigarettes 40 times or more. Percentages among boys and girls. 2003. Values within brackets refer to all students. Data sorted by all students. Germany and Turkey: Limited geographical coverage.

130

The alcohol and drug situation 2003

Islands (83%) followed by Austria, the Czech Republic and Lithuania in which 80% had ever smoked. Next to those are Estonia, Germany, Greenland and Latvia, with lifetime prevalence rates just below 80%. The lowest figures are found in Cyprus, Greece, Iceland, Malta and Turkey, but also in these countries about half of the study population had ever tried to smoke (46 – 52%). Thus, it seems as if smoking is somewhat less prevalent in the eastern part of the Mediterranean area, while Iceland makes out a contrast as a Nordic country at the bottom of the list, especially in comparison with other Nordic islands like the Faroe Islands and Greenland where the prevalence rates are among the highest. Looking at figures 29a–b, where the prevalence rates for smoking 40 times or more in lifetime are presented, it is obvious that there are more students reporting this frequency of smoking in countries where the lifetime prevalence figures are the highest. In Austria, the Czech Republic, the Faroe Islands, Greenland, Germany, Lithuania and Russia (Moscow) about 40% had smoked 40 times or more in their lifetime. The lowest prevalence rates are found in Turkey (13%), Malta (16%), Iceland and Portugal (18% each). It is obvious that smoking is especially prevalent in the central and eastern parts of Europe, but also in the North Atlantic islands, the Faroe Islands and Greenland. On the other hand, the other island in the same area, Iceland, is one of the low prevalence countries. In eight of the 35 ESPAD countries more boys than girls had smoked 40 times or more in their lifetime. They are mainly found in the eastern parts of Europe such as Estonia, Latvia, Lithuania, Poland, Romania and Ukraine, but also in Cyprus and Turkey. Large differences in the other direction with more girls reporting this behaviour are mainly found in two islands countries, Greenland and Isle of Man.

Cigarette smoking during the last 30 days (Tables 2a–c, figures 30a–b) The highest percentage of students having been smoking during the last 30 days is found in Greenland, which is outstanding on this variable (60%). High rates are also found in Austria (49%), Bulgaria (46%), Germany (45%), Russia (Moscow) (44%) and the Czech Republic (43%). Particularly low proportions were found in Cyprus, Iceland, Sweden and Turkey with figures varying between 18 and 25%. In Spain 27% of the students had been smoking during the last 30 days.

The alcohol and drug situation 2003

The gender pattern reveals that countries with substantially higher rates of last month smoking among boys include Cyprus, Latvia, Lithuania, Turkey and Ukraine. The other way around, i.e. considerably higher figures among girls, are mainly found in Greenland, Ireland, Isle of Man and the United Kingdom. The distributions do not follow any strict geographical pattern although the male smokers are predominantly found in the eastern parts of Europe (and eastern Mediterranean) and females in the west, predominantly in the British Isles. Some students are more or less occasional smokers and do not smoke every day. However, on average 2% of the students have smoked 21 cigarettes or more during the last 30 days. The variations between countries are not important and only in two countries as much as about 5% of the students reported this (Croatia and Ireland). A closer look at those who have smoked 6 or more cigarettes during the last 30 days gives a better picture of high and low prevalence countries. The country where the highest number of students had done this is the Faroe Islands, where 30% reported this frequency of smoking. Other countries where this to a higher extent was reported are Austria and Bulgaria (24–25%), Germany, Greenland and Ireland (21–22%). Very few had been smoking that often in Malta, Sweden and Turkey (6%).

Age at first use (Table 3) Young people may have tried occasionally to smoke early in life, and some of them continue to a habitual smoking, while others do not. The number of students, who have smoked their first cigarette at the age of 13 or younger, vary considerably over the countries, from 20 to 60%. In eleven countries more than half of the students have tried to smoke at the age of 13 or younger. The highest percentages are found in the Faroe Islands and Germany (59% each), Latvia (57%), Austria, Estonia and Greenland (56% each). The lowest proportions of students who have tried to smoke at this early age are found in Greece (20%), Turkey (23%), Cyprus and Iceland (26% each ) and in Malta (29%). In many ESPAD countries the prevalence rates of early initiation to cigarette smoking is rather equal between boys and girls. In some countries, however, there are more boys that report early use of cigarettes. The largest gender differences are found in Cyprus, Estonia, Latvia, Lithuania, Poland, Romania, the Slovak Republic, Switzerland

131

                             

24 % 25 30 % 31 39 % 40 48 % 49

                             

         

% ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ;

Data uncertain or not available Non-participating country

; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ;

; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ;

; ;

              ; ; ; ; ; ;

                    

                    

                    

             

             

Figure 30a. Cigarette smoking during the last 30 days. Percentages among all students. 2003. Germany and Turkey: Limited geographical coverage. Spain: Limited comparability.

Boys

Girls

56 42 43 44 43 42 49 46 39 35 35 40 39 36 35 33 49 31 28 32 32 35 27 23 32 25 27 24 28 28 22 33 20 20 22 % 80

60

40

20

65

Greenland (60) Austria (49) Bulgaria (46) Germany (6 Bundesl.) (45) Russia (Moscow) (44) Czech Republic (43) Faroe Islands (41) Lithuania (41) Latvia (40) Hungary (39) Finland (38) Italy (38) Estonia (37) Slovak Republic (37) Croatia (36) Slovenia (36) Switzerland (34) Ukraine (34) France (33) Ireland (33) Belgium (32) Netherlands (31) Poland (31) Denmark (30) Isle of Man (30) Romania (29) United Kingdom (29) Greece (28) Norway (28) Portugal (28) Malta (27) Spain (27) Cyprus (25) Sweden (23) Iceland (20) Turkey (6 cities) (18)

48

0

56 50 46 44 43 41 33 36 40 41 40 33 36 37 38 34 28 36 37 33 31 27 32 36 26 34 30 32 27 26 31 18 26 20 12 0

20

40

60

80 %

Figure 30b. Cigarette smoking during the last 30 days. Percentages among boys and girls. 2003. Values within brackets refer to all students. Data sorted by all students. Germany and Turkey: Limited geographical coverage. Spain: Limited comparability.

132

The alcohol and drug situation 2003

and Ukraine. Not so many countries are reporting more girls than boys that have tried their first cigarette by the age of 13 years. The most important include Greenland, Ireland, Isle of Man and the United Kingdom. Overall, there seem to be more early smokers among boys in the eastern parts of Europe than in the western. Countries with more female early smokers are mainly found in the British Isles.

Daily smoking at the age of 13 (Table 3, figures 31a–b) The number of students who have been daily smokers at the age of 13 or younger is rather high in some countries in the northern parts of Europe, but lower in the south. The highest proportions are found in the Faroe Islands (20%), Germany (18%) and Estonia (17%) followed by Finland, Greenland

and Russia (Moscow) (15% each). The lowest percentages are found in Turkey (3%), Greece (4%), Romania (5%), Hungary and Italy (6% each). However, in most countries the prevalence rates range from 7 to 14%. In a majority of the ESPAD countries, both in the north and the south, the number of students that report daily smoking at this early age are rather equally distributed between the sexes. However, in some countries the gender differences are important and they go in different directions. In a number of countries the proportions among boys double (or more than double) those of the girls. This is true in Cyprus, Latvia, Lithuania, Poland, Romania and Ukraine. Another country with a male majority of early smokers is Estonia. Gender differences in the opposite direction are mainly found in Greenland, Isle of Man and the United Kingdom.

Alcohol consumption Lifetime use of any alcoholic beverage (Tables 4a–c, figures 32a–b) In two thirds of the ESPAD countries the vast majority (90% or more) of the students have been drinking alcohol at least once in their lifetime. The highest percentages are found in the Czech Republic, Lithuania (98% each), the Slovak Republic (97%), Austria, Denmark, Estonia, Germany, Greece, Isle of Man and Latvia (96% each). In some few countries, however, smaller proportions report this experience. The country that deviates the most from this pattern is Turkey, where only slightly less than half of the students (45%) report having been drinking any alcohol at all. Other countries with low prevalence rates include Iceland (75%), Greenland and Portugal (78–80%). Not all of those who have tried alcohol at least once in their lifetime drink on a regular basis. Thus the number of students that have been drinking at least 40 times can be viewed as more of a regular customer. The prevalence rates of this frequency of drinking are much lower than the total lifetime prevalence. The highest percentages reporting use of alcohol 40 times or more in their lifetime are found in more or less the same countries that also had the highest lifetime prevalence rates. They include Denmark (50%), Austria (48%), the Czech Republic (46%), Isle of Man, the Netherlands (45% each) and the

The alcohol and drug situation 2003

United Kingdom (43%). The lowest proportion is reported from Turkey (7%) followed by Greenland, Iceland, Norway and Portugal (13–15%). The gender pattern reveals that in almost all countries there are more boys than girls who report this behaviour. In a few countries, Isle of Man, Finland and Norway, the gender distribution is about equal. However, no country reports prevalence rates among girls that exceed those of the boys.

Last 12 months (Tables 5a–c, figures 33a–b) Not all students who have reported lifetime experience of alcohol have used it as recently as during the last 12 months. Only in 10 of the 35 countries 90% or more had indicated alcohol use during the last 12 months. They include the Czech Republic, Denmark (95% each), Isle of Man, Lithuania (94% each), Austria, Germany (93% each), Greece, the United Kingdom (91% each), Malta and the Slovak Republic (90% each). Of those reporting the lowest 12 months prevalence rates Turkey is again the country with the lowest frequency. Only 35% of the Turkish students had been drinking alcohol during the last 12 months. Other countries with low numbers include Iceland (64%) together with the Faroe Islands, Greenland, Norway, Portugal and Sweden (73–77%). The percentage that report drinking 20 times or

133

; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ;

5% 6 9% 10 13 % 14 15 % 16

; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ;

; ; ; ; ; ; ; ; ; ;

% ; ; ; ; ;; ; ; ; ; ; ; ; ; ;

Data uncertain or not available Non-participating country

; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ;

; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ;

; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ;

; ; ; ; ; ; ; ; ; ; ; ; ; ;

; ; ; ; ; ; ; ; ; ; ; ; ; ;

Figure 31a. Daily smoking at the age of 13 or younger. Percentages among all students. 2003. Germany and Turkey: Limited geographical coverage. USA: Limited comparability.

Boys

Girls Faroe Islands (20) Germany (6 Bundesl.) (18) Estonia (17) Finland (15) Greenland (15) Russia (Moscow) (15) Ireland (14) Latvia (14) Austria (13) Czech Republic (13) Isle of Man (13) Lithuania (13) Slovak Republic (13) United Kingdom (13) Denmark (12) Netherlands (12) Croatia (11) Norway (11) Ukraine (11) Belgium (10) Bulgaria (10) Poland (9) Portugal (9) Sweden (9) Switzerland (9) Iceland (8) Malta (7) Slovenia (7) Cyprus (6) Hungary (6) Italy (6) Romania (6) Greece (4) USA (4) Turkey (6 cities) (3)

21 18 21 15 9 18 12 19 13 14 7 19 15 9 11 10 13 10 16 10 10 13 8 8 9 9 5 7 10 7 6 9 4 5 % 25

20

15

10

5

0

20 19 13 15 21 13 16 10 14 11 18 7 11 18 13 14 9 12 5 11 11 5 10 11 9 8 8 7 4 5 6 3 4 2 0

5

10

15

20

25 %

Figure 31b. Daily smoking at the age of 13 or younger. Percentages among boys and girls. 2003. Values within brackets refer to all students. Data sorted by all students. Germany and Turkey: Limited geographical coverage. USA: Limited comparability.

134

The alcohol and drug situation 2003

; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ;

15 % 16 23 % 24 29 % 30 39 % 40

; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ;

; ; ; ; ; ; ; ; ; ;

% ; ; ; ; ;; ; ; ; ; ; ; ; ; ;

Data uncertain or not available Non-participating country

; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ;

; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ;

; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ;

; ; ; ; ; ; ; ; ; ; ; ; ; ;

; ; ; ; ; ; ; ; ; ; ; ; ; ;

Figure 32a. Lifetime use of any alcoholic beverage 40 times or more. Percentages among all students. 2003. Germany and Turkey: Limited geographical coverage. USA: Limited comparability.

Boys

Girls Denmark (50) Austria (48) Czech Republic (46) Isle of Man (45) Netherlands (45) United Kingdom (43) Ireland (39) Russia (Moscow) (39) Lithuania (38) Germany (6 Bundesl.) (37) Belgium (36) Greece (35) Slovak Republic (34) Malta (33) Estonia (32) Faroe Islands (32) Bulgaria (27) Croatia (27) Poland (27) Switzerland (27) Latvia (26) Slovenia (25) Italy (24) France (22) Ukraine (22) Cyprus (21) Hungary (21) Finland (20) Romania (18) Sweden (17) Norway (15) Iceland (14) Portugal (14) Greenland (13) USA (12) Turkey (6 cities) (7)

57 53 54 45 55 47 42 44 45 43 46 43 42 41 38 34 33 38 36 33 30 32 33 30 24 34 27 20 26 21 17 16 20 17 15 10 % 60

40

20

0

42 41 40 44 35 39 36 34 31 31 27 28 28 27 26 30 21 16 18 20 23 18 16 15 19 12 14 20 12 14 14 12 8 9 10 4 0

20

40

60 %

Figure 32b. Lifetime use of any alcoholic beverage 40 times or more. Percentages among boys and girls. 2003. Values within brackets refer to all students. Data sorted by all students. Germany and Turkey: Limited geographical coverage. USA: Limited comparability.

The alcohol and drug situation 2003

135

; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ;

10 % 11 15 % 16 24 % 25 34 % 35

; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ;

; ; ; ; ; ; ; ; ; ;

% ; ; ; ; ;; ; ; ; ; ; ; ; ; ;

Data uncertain or not available Non-participating country

; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ;

; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ;

; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ;

; ; ; ; ; ; ; ; ; ; ; ; ; ;

; ; ; ; ; ; ; ; ; ; ; ; ; ;

Figure 33a. Use of any alcoholic beverage 20 times or more during the last 12 months. Percentages among all students. 2003. Germany and Turkey: Limited geographical coverage. USA: Limited comparability. Boys

Girls

48 46 47 31 42 38 39 32 38 35 26 33 31 29 29 27 28 24 26 27 24 23 26 19 20 16 13 17 13 18 10 11 16 10 11 8 % 50

40

30

20

10

36 36

Denmark (42) Austria (41) Netherlands (37) Ireland (35) Czech Republic (34) United Kingdom (34) Malta (32) Isle of Man (31) Belgium (30) Germany (6 Bundesl.) (29) Faroe Islands (27) Greece (27) Russia (Moscow) (26) Lithuania (23) Switzerland (23) Slovak Republic (22) Croatia (21) Estonia (21) Bulgaria (19) Italy (19) Poland (18) Slovenia (18) Cyprus (17) Latvia (17) Hungary (15) Ukraine (14) Finland (13) France (12) Norway (12) Romania (12) Greenland (11) Sweden (11) Portugal (10) Iceland (9) USA (8) Turkey (6 cities) (5) 0

28 39 28 31 26 30 22 23 26 23 23 17 16 16 13 19 14 12 13 12 9 15 10 11 13 8 12 7 11 9 6 8 7 3 0

10

20

30

40

50 %

Figure 33b. Use of any alcoholic beverage 20 times or more during the last 12 months. Percentages among boys and girls. 2003. Values within brackets refer to all students. Data sorted by all students. Germany and Turkey: Limited geographical coverage. USA: Limited comparability.

136

The alcohol and drug situation 2003

more during the last year gives a picture of a more frequent alcohol use. The highest numbers are found in Denmark and Austria (41–42%), the Netherlands (37%) and Ireland (35%) followed by the Czech Republic and the United Kingdom (34%). The lowest frequencies are found in Turkey (5%), Iceland and Portugal (9–10%). In most ESPAD countries, such frequent drinking during the last 12 months is a typical male behaviour. Thus, in about three fourths of the ESPAD countries a majority of those reporting this were boys. In only one country, Ireland, more girls than boys (39 vs. 31%) indicated this drinking behaviour. About equal proportions, however, are reported from almost only Nordic countries including the Faroe Islands, Finland, Greenland, Iceland, Isle of Man, Norway and Sweden.

Last 30 days Any alcohol (Tables 6a–c, figures 34a–b) The number of students who had been drinking any alcohol during the last 30 days varies quite a lot between the ESPAD countries. In Austria, Denmark and Isle of Man a vast majority of the students (79–82%) had been drinking alcohol during this period. Other countries where about three quarters of the students reported this include Germany (78%), the Czech Republic, Lithuania (77% each), Greece, Malta, and Switzerland (75% each). Much lower prevalence rates are reported from Turkey, where only 20% of the students reported any alcohol use during the last month, but also Iceland reports a rather low figure on this variable (37%). Countries where about half of the students had been drinking any alcohol during the last 30 days include Portugal (48%), Greenland, Norway, Sweden (51% each), Finland and Romania (54% each). A higher frequency of alcohol use is revealed in the number of students who had been consuming alcohol 10 times or more during the last 30 days, i.e. at least every third day if a drinking occasion is defined as a day when you drink alcohol. In the Netherlands one quarter of the students report this behaviour (25%), while about one fifth of the respondents in Austria, Belgium, Malta and the United Kingdom did so (17–21%). In some other countries, this drinking frequency is hardly reported at all. Proportions of 3% or less were found in Finland, Greenland, Iceland, Norway and Sweden. Thus, the very low prevalence rates were mainly found in the Nordic countries.

The alcohol and drug situation 2003

In a majority of the countries more boys than girls are reporting this frequency of drinking. With two exceptions this is clearly related to the prevalence rates, i.e. there are more males reporting this in the high prevalence countries. The two exceptions are the United Kingdom and Ireland where the sex distribution is about equal. No country is reporting more girls than boys with this behaviour, but in a number of countries the proportions are about equal between the sexes, especially so in the low prevalence countries. The students were asked what kind of beverage they had been drinking during the last 30 days. In the next three sections their choice of beverage is reported. The presentation is focused on the consumption of beer, wine and spirits.

Beer (Tables 7a–c, Figures 35a–b) The largest proportions that report having been drinking beer during last 30 days were found in Bulgaria (70%), Denmark (69%), Poland (68%), Romania (67%), the Czech Republic (63%) and Ukraine (61%). As a contrast, only 21% of the Turkish students had done so. In some other countries the prevalence rates are also rather low. They include Hungary, Norway and Portugal where about 35% had had beer. Many students report rather frequent beer consumption. The percentages of students who had been drinking beer 3 times or more during the last 30 days varies between 10 and 44%. The highest figures are found in Denmark (44%), Bulgaria (43%), Poland (41%) and the Netherlands (40%). Other countries with high levels include the Czech Republic (39%), Russia (Moscow) and the Slovak Republic (38% each). The smallest proportions were reported from Turkey (10%) and Norway (14%). Other countries where less than 20% had consumed beer that often include Finland, Hungary, Iceland, and Portugal. Beer drinking does not follow any geographical pattern, neither among the high nor the low prevalence countries. Drinking beer is a predominantly male behaviour. This is true also in relation to frequent consumption. In almost all ESPAD countries more boys than girls reported that they had been drinking beer 3 times or more often over the last 30 days. The only exceptions where the distributions are almost equal between the sexes are found in two countries in the North Atlantic, Greenland and Iceland.

137

; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ;

3% 4 7% 8 12 % 13 19 % 20

; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ;

; ; ; ; ; ; ; ; ; ;

% ; ; ; ; ;; ; ; ; ; ; ; ; ; ;

Data uncertain or not available Non-participating country

; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ;

; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ;

; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ;

; ; ; ; ; ; ; ; ; ; ; ; ; ;

; ; ; ; ; ; ; ; ; ; ; ; ; ;

Figure 34a. Use of any alcoholic beverage 10 times or more during the last 30 days. Percentages among all students. 2003. Germany and Turkey: Limited geographical coverage. USA: Limited comparability. Boys

Girls Netherlands (25) Austria (21) Belgium (20) Malta (20) United Kingdom (17) Ireland (16) Isle of Man (15) Croatia (13) Czech Republic (13) Denmark (13) Greece (13) Switzerland (13) Italy (12) Russia (Moscow) (12) Cyprus (11) Germany (6 Bundesl.) (11) Poland (10) Bulgaria (9) Slovak Republic (9) Lithuania (8) France (7) Portugal (7) Slovenia (7) Estonia (6) Hungary (6) Latvia (6) Romania (5) Ukraine (5) Faroe Islands (4) Turkey (6 cities) (4) USA (4) Greenland (3) Norway (3) Finland (2) Iceland (1) Sweden (1)

34 27 28 25 18 17 19 15 17 18 17 18 17 16 16 15 13 13 12 13 10 11 10 8 8 7 9 6 6 4 5 5 3 3 2 2 % 40

30

20

10

0

17 15 13 16 15 14 13 11 10 10 9 7 6 10 6 9 6 7 6 5 5 4 4 5 4 4 3 4 4 1 4 4 2 2 1 1 0

10

20

30

40 %

Figure 34b. Use of any alcoholic beverage 10 times or more during the last 30 days. Percentages among boys and girls. 2003. Values within brackets refer to all students. Data sorted by all students. Germany and Turkey: Limited geographical coverage. USA: Limited comparability.

138

The alcohol and drug situation 2003

; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ;

14 % 15 25 % 26 35 % 36 39 % 40

; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ;

; ; ; ; ; ; ; ; ; ;

% ; ; ; ; ;; ; ; ; ; ; ; ; ; ;

Data uncertain or not available Non-participating country

; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ;

; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ;

; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ;

; ; ; ; ; ; ; ; ; ; ; ; ; ;

; ; ; ; ; ; ; ; ; ; ; ; ; ;

Figure 35a. Beer consumption 3 times or more during the last 30 days. Percentages among all students. 2003. Germany and Turkey: Limited geographical coverage. USA: Limited comparability.

Boys

Girls

54 55 50 55 52 47 50 49 47 45 45 33 42 46 38 42 45 45 42 39 38 39 40 35 41 32 31 26 26 19 22 27 23 16 15 14 % 60

40

20

35 33 32

Denmark (44) Bulgaria (43) Poland (41) Netherlands (40) Czech Republic (39) Russia (Moscow) (38) Austria (36) Belgium (36) Ireland (36) Italy (34) Ukraine (34) Greenland (32) Latvia (32) Romania (32) Faroe Islands (31) United Kingdom (31) Germany (6 Bundesl.) (30) Malta (29) Croatia (28) Greece (28) Lithuania (28) Switzerland (28) Cyprus (27) Estonia (25) Isle of Man (25) Slovak Republic (22) Slovenia (21) France (20) Sweden (20) Iceland (19) Finland (18) Portugal (18) Hungary (17) Norway (14) USA (12) Turkey (6 cities) (10) 0

23 28 30 18 24 25 22 21 32 22 22 21 17 18 16 14 18 20 18 16 16 13 14 12 16 14 17 13 10 9 11 9 5 0

20

40

60 %

Figure 35b. Beer consumption 3 times or more during the last 30 days. Percentages among boys and girls. 2003. Values within brackets refer to all students. Data sorted by all students. Germany and Turkey: Limited geographical coverage. USA: Limited comparability.

The alcohol and drug situation 2003

139

Wine (Tables 8a–c, figures 36a–b) Much less students had been drinking wine during the last 30 days compared to beer. The country with the most outstanding figure reporting this behaviour is Malta where 68% had been drinking wine during the last 30 days. Other relatively high prevalence countries include Austria, the Czech Republic and Lithuania, where a little more than half of the students had consumed wine during this period. In some countries very few students had been drinking wine during the last month. The two with the lowest frequencies are Turkey (10%) and Portugal (15%). Other low prevalence countries include the Faroe Islands, Greenland, Iceland and Norway (around 19%). Apparently there is no clear geographical pattern in the distribution of low prevalence countries on wine consumption. They include one low alcohol prevalence country like Turkey, one wine producing country and four North Atlantic countries. The proportions of students reporting a wine consumption frequency of 3 times or more during the last 30 days are mainly lower than 20%. However, one country is outstanding in this respect since one third (35%) of the students in Malta reported this frequency of wine consumption. Other high prevalence countries include Austria and Italy (about 23%), the Czech Republic, Greece and Slovenia (21% each). Thus, all high prevalence countries are wine producing countries. The lowest proportions that reported this frequency of wine consumption are found in Norway (3%), Turkey (4%), Finland and Iceland (5% each). Other low prevalence countries (below 9%) are found in northern Europe such as Greenland, the Faroe Islands, Poland and Sweden, but also in “wine countries” like France and Portugal. In about one third of the ESPAD countries boys are in majority when it comes to drinking wine 3 times or more during the last 30 days. In about the same number of countries there are hardly any gender differences at all. In some countries, however, the girls are in majority among these consumers. Findings in this direction are mainly found in the Czech Republic, Germany, Ireland and the United Kingdom. Spirits (Tables 9a–c, figures 37a–b) The number of students who had been drinking spirits during the last 30 days varies considerably between the ESPAD countries. The highest per-

140

centages, around two thirds of the student population, are found in countries spread geographically all over Europe. They include Isle of Man (66%), Denmark, Malta (65% each), Greece, Switzerland (63% each), the United Kingdom (61%) and Ireland (60%). Much lower numbers are found in a few countries in different parts of Europe. They mainly include Turkey (11%) and Romania (23%). A similar picture of high and low prevalence countries is found when looking at the percentage of students that had been drinking spirits at least 3 times during the last 30 days. Again the British Isles are appearing at the top, but also the two Mediterranean countries. The highest proportion is found in Malta, where 43% of the students reported this frequency of spirits consumption. Next come the United Kingdom (39%), Ireland, Isle of Man (38% each), the Faroe Islands and Greece (37% each). Countries where rather few students reported this frequency of drinking spirits include Turkey (5%), Romania (6%) and Finland (10%). Other countries with low prevalence rates are Latvia, Iceland, Lithuania and Poland (12–14%). In about half of the countries there are more boys than girls reporting such a frequent consumption of spirits. However, in about the same number of countries the prevalence rates are equal or almost equal between the sexes. Only three countries reported proportions among the girls that exceeded those of the boys. These countries were all high frequency countries and they were all parts of the British Isles, i.e. Ireland, Isle of Man and the United Kingdom.

Last drinking occasion The questionnaire included five questions regarding the consumed quantities on the last drinking occasion, beverage by beverage. The students were asked: “The last time you had an alcoholic drink, did you drink any beer (/cider/alcopops/wine/spirits)? If so, how much?” The format of the response categories was set as fixed quantities relevant for each beverage in terms of centilitres. Since glasses, bottles and cans differ in size between countries, each ESPAD researcher described the fixed response categories in the best possible way. The question also included the response categories “I never drink beer (/cider/alcopops/wine/ spirits)” and “I did not drink beer (/cider/alcopops/ wine/spirits) on my last drinking occasion”. Countries in which cider or alcopops are virtually nonexistent did not include questions about these beverages.

The alcohol and drug situation 2003

; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ;

5% 6 9% 10 15 % 16 20 % 21

; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ;

; ; ; ; ; ; ; ; ; ;

% ; ; ; ; ;; ; ; ; ; ; ; ; ; ;

Data uncertain or not available Non-participating country

; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ;

; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ;

; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ;

; ; ; ; ; ; ; ; ; ; ; ; ; ;

; ; ; ; ; ; ; ; ; ; ; ; ; ;

Figure 36a. Wine consumption 3 times or more during the last 30 days. Percentages among all students. 2003. Germany and Turkey: Limited geographical coverage.

Boys

Girls

42 20 18 27 28 23 20 21 21 15 15 12 16 17 12 16 12 20 12 14 7 6 12 6 11 9 9 7 5 8 4 5 5 4 % 50

40

30

20

10

32

Malta (35) Italy (23) Austria (22) Czech Republic (21) Greece (21) Slovenia (21) Croatia (19) Hungary (19) Slovak Republic (19) Belgium (18) Isle of Man (18) United Kingdom (18) Germany (6 Bundesl.) (17) Russia (Moscow) (17) Ukraine (17) Estonia (15) Bulgaria (14) Lithuania (13) Romania (13) Latvia (12) Cyprus (11) Netherlands (11) Ireland (10) Switzerland (10) Denmark (9) France (8) Poland (8) Sweden (8) Faroe Islands (7) Greenland (6) Portugal (6) Finland (5) Iceland (5) Norway (4) Turkey (6 cities) (4)

29

0

16 23 24 15 17 15 16 17 15 20 22 21 19 18 15 9 16 9 14 7 12 13 8 10 4 7 8 4 6 3 6 5 4 3 0

10

20

30

40

50 %

Figure 36b. Wine consumption 3 times or more during the last 30 days. Percentages among boys and girls. 2003. Values within brackets refer to all students. Data sorted by all students. Germany and Turkey: Limited geographical coverage.

The alcohol and drug situation 2003

141

; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ;

10 % 11 17 % 18 27 % 28 35 % 36

; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ;

; ; ; ; ; ; ; ; ; ;

% ; ; ; ; ;; ; ; ; ; ; ; ; ; ;

Data uncertain or not available Non-participating country

; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ;

; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ;

; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ;

; ; ; ; ; ; ; ; ; ; ; ; ; ;

; ; ; ; ; ; ; ; ; ; ; ; ; ;

Figure 37a. Consumption of spirits 3 times or more during the last 30 days. Percentages among all students. 2003. Germany and Turkey: Limited geographical coverage.The Netherlands: Pre-mixed drinks not included. Boys

Girls

44 29 32 42 41 37 36 34 36 33 39 30 33 29 27 30 27 27 28 23 19 21 19 20 17 17 18 15 18 15 13 10 10 7 % 50

40

30

20

10

42 43

Malta (43) United Kingdom (39) Ireland (38) Isle of Man (38) Faroe Islands (37) Greece (37) Switzerland (35) Netherlands (34) Denmark (31) Austria (30) Belgium (29) Cyprus (29) Czech Republic (28) Greenland (26) Portugal (26) Bulgaria (25) Italy (25) Slovak Republic (25) Estonia (24) Germany (6 Bundesl.) (24) Hungary (22) Slovenia (20) France (19) Sweden (18) Croatia (17) Norway (17) Russia (Moscow) (16) Ukraine (16) Lithuania (14) Poland (14) Iceland (13) Latvia (12) Finland (10) Romania (7) Turkey (6 cities) (5)

33

0

46 46 34 32 33 30 31 23 26 21 27 20 24 25 21 22 21 23 21 21 16 17 18 18 14 12 12 11 13 10 10 5 1 0

10

20

30

40

50 %

Figure 37b. Consumption of spirits 3 times or more during the last 30 days. Percentages among boys and girls. 2003. Values within brackets refer to all students. Data sorted by all students. Germany and Turkey: Limited geographical coverage. The Netherlands: Pre-mixed drinks not included.

142

The alcohol and drug situation 2003

In a few countries (Austria and Germany) the response categories were changed into an open format to try to get as realistic and true alcohol volumes as possible. However, since this change of format is known to distort the distributions, the results of these countries are put below the line in the tables, and are excluded from the maps and bar graphs, in order to draw the reader’s attention to the limited comparability. In Switzerland the response categories for cider and alcopops consumption are different from the ESPAD format, which is the reason for putting these results under the bottom line in relevant tables. The results on these beverage specific questions are presented below. They include beer, cider, alcopops, wine and spirits.

Beer (Tables 10a–c, figures 38a–b) The proportions of students who had been drinking beer last time they had any alcohol vary between one third and two thirds. The highest percentage of students reporting this, are found in Poland and Romania (69% each), Denmark (65%), Lithuania (61%) and the Czech Republic (60%). The countries with relatively small proportions reporting this behaviour include Hungary and Turkey, where one third (33%) had been drinking beer on the last drinking occasion, but also in Norway (38%) and Croatia (37%). Some of the students who had been drinking beer had consumed quite large quantities. In some countries about one third of the students had consumed at least 101 cl beer on the last drinking occasion. These countries are Denmark (37%), Ireland (32%) and the Netherlands (28%). Other countries where quite large proportions report this level of consumption include Finland (25%), Greenland, Iceland (24% each), the Czech Republic, the United Kingdom (23% each) and the Faroe Islands (22%). Very few students reported this behaviour in Ukraine (4%), Turkey (5%), Portugal and Romania (6% each). Other low prevalence countries are Greece, Hungary and the Slovak Republic (7% each), Cyprus, Italy (8% each) and Slovenia (9%). An interesting detail is that the two countries which formerly were united as one country, the Czech Republic and the Slovak Republic, show completely different drinking pattern in relation to the consumption of beer. The gender pattern reveals that beer drinking is a predominantly male behaviour. In all countries, from high to low prevalencies, there are more boys

The alcohol and drug situation 2003

that reported this level of consumption. The only country with equal proportions between the sexes is Greenland.

Cider (Tables 11a–c) Not all ESPAD countries included the question about the consumption of cider in their questionnaire. The reason is that cider is not included in the alcoholic beverage assortments in these countries. However, the results show that not very many students had cider the last time they were drinking alcohol. The largest percentages of students reporting this are found in Romania (42%), Finland (38%), Sweden (35%), Estonia (34%) and Norway (32%). In certain countries very few students indicated cider consumption on the last alcohol occasion. Less than 10% had cider in Poland (5%), Turkey (6%) and Cyprus (8%). Also in the countries where the highest percentages of students had reported consumption of cider only rather few had been drinking large quantities. The highest figures in relation to a consumption of 101 cl or more are found in Ireland (14%), Sweden (9%), Finland (8%) and Norway (7%). Very small gender differences are observed. More boys than girls had been drinking cider at the last drinking occasion in Ireland and Isle of Man. The only country where the opposite is true, i.e. more girls than boys reporting this behaviour, is Finland. Alcopops (Tables 12a–c) Similar to the case of cider, not all countries have alcopops in the assortment of alcoholic beverages. In addition, there are mixtures of alcohol that may be considered as alcopops although it is labelled “mixed drinks” as is the case in the Netherlands. Since these beverages are very similar to alcopops and assumingly consumed in the same way, they are included in the alcopops category. The countries that report the highest percentages of students who had alcopops the last time they had any alcohol are mainly found in the western parts of Europe, but also in a few Mediterranean countries. Thus, the highest figures are noted for Cyprus, Isle of Man (62% each), Denmark (61%), the Netherlands (52%), the United Kingdom (50%), Greece (49%), Belgium (48%), Greenland (46%) and Norway (43%). The lowest figures are found in a few Baltic countries, but also in the Mediterranean country

143

6% 7 12 % 13 20 % 21 27 % 28

% ; ; ; ; ;; ; ; ; ; ; ; ; ; ;

Data uncertain or not available Non-participating country

; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ;

; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ;

; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ;

; ; ; ; ; ; ; ; ; ; ; ; ; ;

; ; ; ; ; ; ; ; ; ; ; ; ; ;

Figure 38a. Consumption of 101 cl beer or more on the last drinking occasion. Percentages among all students. 2003. Turkey: Limited geographical coverage.

Boys

Girls

51 43 35 23 28 37 34 32 32 23 27 26 23 23 23 22 23 21 19 18 17 15 13 10 11 13 10 11 15 8 6 % 60

40

20

22

Denmark (37) Ireland (32) Netherlands (28) Finland (25) Greenland (24) Iceland (24) Czech Republic (23) United Kingdom (23) Faroe Islands (22) Isle of Man (19) Norway (19) Sweden (18) Poland (16) Belgium (15) Croatia (14) Lithuania (14) Estonia (13) Malta (13) Switzerland (13) Latvia (11) Russia (Moscow) (11) Bulgaria (10) Slovenia (9) Italy (8) Greece (7) Hungary (7) Slovak Republic (7) Portugal (6) Romania (6) Cyprus (8) Turkey (6 cities) (5) Ukraine (4)

47

0

17 13 15 25 20 10 11 12 8 15 9 6 8 6 5 4 4 5 5 5 4 3 4 4 1 1 2 2 1 2 1 0

20

40

60 %

Figure 38b. Consumption of 101 cl beer or more on the last drinking occasion. Percentages among boys and girls. 2003. Values within brackets refer to all students. Data sorted by all students. Turkey: Limited geographical coverage.

144

The alcohol and drug situation 2003

Malta. Poland is the country that reports the smallest percentage of students who had alcopops to drink on the last drinking occasion (7%). Other countries with rather low figures include Sweden (14%), Latvia (17%) and, as mentioned above, Malta (18%). There are large differences in the number of students who report having been drinking rather large quantities on their last occasion with alcopops. In a number of countries only 1–2% report this, while in others 20–30% have been drinking considerable amounts. The highest numbers are reported from Isle of Man, where 35% of the students had been drinking 101 cl of alcopops or more. The countries next to Isle of Man are all on a somewhat lower level such as the United Kingdom (24%), Ireland, the Netherlands (20% each), Norway (17%), Denmark and Greenland (15%). The lowest percentages in this respect are reported from Latvia, Ukraine (1% each), Hungary, Poland, Romania, Slovenia, Sweden (2% each), Lithuania, Malta and Russia (Moscow) (3% each). The gender pattern is very homogenous. In a majority of the countries more girls than boys report having been drinking alcopops on their last drinking occasion. Deviant from this pattern are Cyprus, Greenland and Romania, where boys are in majority. On the other hand, in five countries no gender differences are observed. They include Greece, Malta, Poland, Slovenia and Sweden.

Wine (Tables 13a–c, figures 39a–b) There is a wide variation between countries in the consumption of wine on the last drinking occasion. In six countries half of the students or more had been drinking wine on this occasion. The highest numbers are observed in Malta where 61% of the students had wine. Other countries with high percentages are Slovenia (57%), Lithuania (54%), Estonia, the Slovak Republic (53% each) and the Czech Republic (50%). The lowest figures are found in the North Atlantic area: the Faroe Islands (13%), Greenland and Iceland (15% each). Another country with a low prevalence rate is found at the other end of the European continent, since 14% of the students in Turkey had been drinking wine on the last drinking occasion. Countries with the highest percentages reporting a consumption of 15 cl of wine or more on the last drinking occasion, slightly above one third of the study populations, are found both in the central

The alcohol and drug situation 2003

and southern parts of Europe. In Slovenia 39% reported this, in Malta 36% and in the Czech Republic 35%. Other countries with somewhat high numbers of students reporting this behaviour include Croatia and Estonia (31% each), Lithuania and the Slovak Republic (29% each). The lowest prevalence rates on this level of consumption are found in the Faroe Islands (4%), Iceland, Portugal (5% each), Turkey (7%), France and Greenland (8% each). Thus, there is no clear geographical pattern for the low prevalence countries. The gender pattern does not seem to be related to the prevalence rates in any systematic way. In a little less than half of the countries more boys than girls report a consumption of 15 cl wine or more. The opposite was found in somewhat fewer countries. Countries where no, or small gender differences were observed include most of the Nordic countries, which at the same time are among the low prevalence countries. They include Belgium, Denmark, the Faroe Islands, Finland, Greenland, Iceland, the Netherlands and Norway.

Spirits (Tables 14a–c, figures 40a–b) Also the consumption of spirits on the last drinking occasion differs substantially between countries. In eight countries at least half of the students had been drinking spirits the last time they had any alcohol including Malta (67%), the Faroe Islands (62%), Denmark, Greenland (61% each), Greece (58%), the Czech Republic (55%), Estonia (54%) and Ireland (50%). In some countries much smaller proportions of students report this behaviour. In Turkey 13% had indicated consumption of spirits on the last drinking occasion. The corresponding value for Romania is 15% but from Russia (Moscow) a somewhat higher percentage (24%) is reported. Of those who had been drinking at least 11 cl of spirits on the last occasion the value reported from the Faroe Islands (39%) is outreaching the percentages recorded in other countries. The next highest value is reported from another island at the other end of Europe since 27% in Malta indicated this consumption. Other countries with somewhat high figures include the Czech Republic, Estonia, Ireland (23% each), Greenland (22%), Poland (21%), Denmark and the Isle of Man (20% each). The lowest rates are reported from Romania (2%) and Turkey (3%), but also from Cyprus (6%) and Portugal (7%).

145

9% 10 15 % 16 25 % 26 34 % 35

% ; ; ; ; ;; ; ; ; ; ; ; ; ; ;

Data uncertain or not available Non-participating country

; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ;

; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ;

; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ;

; ; ; ; ; ; ; ; ; ; ; ; ; ;

; ; ; ; ; ; ; ; ; ; ; ; ; ;

Figure 39a. Consumption of 15 cl wine or more on the last drinking occasion. Percentages among all students. 2003. Turkey: Limited geographical coverage.

Boys

Girls Slovenia (39) Malta (36) Czech Republic (35) Croatia (31) Estonia (31) Lithuania (29) Slovak Republic (29) Hungary (24) Isle of Man (24) United Kingdom (24) Greece (23) Belgium (21) Poland (20) Italy (19) Russia (Moscow) (19) Denmark (18) Latvia (17) Romania (15) Ukraine (15) Bulgaria (14) Ireland (13) Netherlands (13) Cyprus (12) Finland (11) Sweden (11) Switzerland (11) Norway (10) France (8) Greenland (8) Turkey (6 cities) (7) Iceland (5) Portugal (5) Faroe Islands (4)

43 41 32 38 27 24 32 29 16 17 29 22 24 26 14 18 12 23 15 18 9 11 16 9 7 13 11 12 9 9 6 9 5 % 50

40

30

20

10

0

35 30 38 26 36 35 27 19 31 30 17 22 16 14 23 19 20 9 16 12 17 13 9 12 12 8 10 5 6 5 5 3 2 0

10

20

30

40 %

Figure 39b. Consumption of 15 cl wine or more on the last drinking occasion. Percentages among boys and girls. 2003. Values within brackets refer to all students. Data sorted by all students. Turkey: Limited geographical coverage.

146

The alcohol and drug situation 2003

4% 5 9% 10 15 % 16 25 % 26

% ; ; ; ; ;; ; ; ; ; ; ; ; ; ;

Data uncertain or not available Non-participating country

; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ;

; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ;

; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ;

; ; ; ; ; ; ; ; ; ; ; ; ; ;

; ; ; ; ; ; ; ; ; ; ; ; ; ;

Figure 40a. Consumption of 11 cl of spirits or more on the last drinking occasion. Percentages among all students. 2003. Turkey: Limited geographical coverage.

Boys

Girls Faroe Islands (39) Malta (27) Estonia (23) Ireland (23) Czech Republic (22) Greenland (22) Poland (21) Denmark (20) Isle of Man (20) Norway (19) Slovak Republic (18) Lithuania (17) United Kingdom (17) Iceland (15) Sweden (15) Finland (13) France (12) Greece (12) Hungary (12) Slovenia (12) Croatia (11) Ukraine (11) Italy (10) Russia (Moscow) (10) Bulgaria (9) Latvia (9) Belgium (8) Netherlands (8) Switzerland (8) Portugal (7) Cyprus (6) Turkey (6 Cities) (3) Romania (2)

42 31 29 16 26 29 30 21 14 20 22 22 14 16 17 15 15 15 15 12 12 15 13 13 10 11 11 10 11 9 10 5 3 % 50

40

30

20

10

0

36 24 17 30 19 15 13 19 26 19 14 12 21 13 13 12 9 9 9 11 10 6 8 7 7 7 5 6 5 6 3 1 1 0

10

20

30

40

50 %

Figure 40b. Consumption of 11 cl of spirits or more on the last drinking occasion. Percentages among boys and girls. 2003. Values within brackets refer to all students. Data sorted by all students. Turkey: Limited geographical coverage.

The alcohol and drug situation 2003

147

Among those who had been drinking 11 cl of spirits or more boys are dominating in a majority of the 35 ESPAD countries. In Ireland, Isle of Man and the United Kingdom, however, more girls than boys reported this level of consumption, i.e. on all the three British Isles. Countries where the figures are about the same among boys and girls include the Nordic countries Denmark, Finland, Iceland and Norway, but also Croatia, Portugal and Slovenia.

Beverages consumed (Tables 15a–c) Some of the information in tables 12–14 is summarised in table 15. It contains information about the proportions of students who consumed beer, wine and spirits on the last drinking occasion. The table also shows the proportion of students who drank relatively large quantities of beer (101 cl or more), wine (37 cl or more) or spirits (11 cl or more). As was pointed out above, data from Austria and Germany are not comparable on these variables and thus appearing under the bottom line in the tables. The most commonly consumed beverage on the last drinking occasion is beer, which was reported by half (49%) of the ESPAD students. The second most reported beverage is spirits (42% on average), while one third of the students had been drinking wine on this occasion. Beer is the dominating beverage in a little more than half of the countries, while spirits is the most common in six (the Faroe Islands, Greece, Greenland, Isle of Man, Malta and Portugal). In a few countries, however, about equal numbers of students have indicated both beer and spirits. These countries are Ireland, Norway, Sweden and the United Kingdom. Wine was the most commonly drunk beverage only in two countries, the Slovak Republic and Slovenia. In Estonia all three beverages were indicated in equal proportions and in Hungary the number of students who had been drinking wine was about the same as for spirits. The consumption of beer is on average a male behaviour. Almost twice as many boys as girls had beer on the last drinking occasion (61 vs. 37%). The gender difference is even more obvious in relation to the consumption of 101 cl or more. There are three times more boys than girls reporting this consumption (23 vs. 8%). Wine consumption on the last drinking occasion is more equally distributed between the sexes. About one third of both boys and girls had wine the last time they had any alcohol. A small minority (about 7%) had been drinking 37 cl wine or more

148

on that occasion and no gender difference was established. Also the percentages of students that had spirits on the last drinking occasion are very similar among boys and girls. Around 42% had been drinking spirits. There are, however, more boys (17%) than girls (12%) reporting a consumption of 11 cl spirits or more. The countries reporting the largest number of students who had 101 cl of beer or more on the last drinking occasion are Denmark (37%), Ireland (32%), the Netherlands (28%) and Finland (25%). On the other hand, very low figures on this level of consumption are reported from Ukraine (4%), Turkey (5%), Portugal, Romania (6% each), Greece, Hungary and the Slovak Republic (7% each). In no country more than one fifth of the students had been drinking 37 cl wine or more on their last drinking occasion. The largest numbers are found in Slovenia (19%), Croatia (16%), the Czech Republic and Malta (13% each). Very few students had indicated these amounts in the Faroe Islands (1%), France, Iceland, Portugal (2% each), Greenland, Switzerland and Turkey (3% each). However, relatively large number of students reported a consumption of spirits equalling to 11 cl or more on the last occasion they had alcohol. The highest number is noted for the Faroe Islands, where 39% of the students had been drinking these amounts. Other countries with high prevalence rates include Malta (27%), Estonia and Ireland (23% each). Also for this variable there are large discrepancies between the high and low prevalence countries. The lowest numbers of students who indicated this level of consumption are found in Romania (2%), Turkey (3%) and Portugal (7%). There are rather big discrepancies between boys and girls when it comes to the consumption of large quantities of beer. Among boys 23% on average have reported this consumption, compared to 8% among girls. For wine there is virtually no difference (8 vs. 6%), but for spirits the boys again are in majority (17 vs. 12%).

Estimated average consumption (Tables 16a–c, 17a–b, figures 41a–b) An attempt has been made to estimate the volumes consumed on the last drinking occasion in each country. For this purpose, the proportions in tables 10–14, indicating different volumes of alcohol, have been used. However, the questions on cider and/or alcopops are not relevant in all countries. The presentation begins with the calculated amounts of beer,

The alcohol and drug situation 2003

wine and spirits that the students had been drinking, based on their answers on the last drinking occasion. In the next section the beverages cider and alcopops have been added when appropriate. The calculations are based on the alcohol content for different beverage types and recalculated into pure alcohol. The alcohol content for alcopops it is assumed to be 4.5%, beer and cider 5%, wine 11% and spirits 40%. It should be noted, that in the fixed answering categories in the 2003 survey the indications given within brackets for volumes of wine have been changed since the 1999 survey. The explanatory bracket for “less than a glass” now says (<15 cl) and for “1–2 glasses” it is (15–30 cl). In the 1999 survey this was measured as (<10 cl) and (10–20 cl). Furthermore, the calculations are done only on students who had ever been drinking alcohol. This is different from the 1999 survey in which the calculations were based on the consumers of each beverage, which probably led to overestimations of the consumption. To help the reader who might want to compare the results of the two surveys, three additional tables (49 a–c) with recalculated figures for the 1999 data have been added. Beer, wine and spirits (Tables 16a–c, figures 41a–b) For the calculations the mid points of each response category’s range are used. For the last openended category the lowest value is used. This is most certainly a conservative estimate, since many of the students in this category probably had been drinking larger quantities. In some countries relatively large number of students indicated the highest category. They are often found in countries with the largest calculated quantities. This means in practice, that the calculated differences between the high consumption countries and the others probably are underestimations. It must be stressed that these kinds of calculations always are uncertain and build on a lot of assumptions. Thus, it is important not to overestimate the differences in the estimates. On the other hand, it seems reasonable to assume that substantial differences in consumption patterns between countries, as well as between boys and girls, most probably also reflect true differences since the calculations are done in exactly the same way in all countries. The total estimated average consumption of beer, wine and spirits that are calculated in tables 16a–c show that beer makes up almost half of the

The alcohol and drug situation 2003

consumed quantities (45%). The next most important beverage is spirits, which forms 37% of the total average for all countries. Wine is contributing to the consumed alcohol only to a relatively limited extent and makes up 17% of the total consumption. There are, however, rather large differences between countries in the distribution of beverages on the last drinking occasion. The consumption on the last occasion in the Faroe Islands is outreaching those of the other ESPAD countries. On average, the students in this country had consumed 8.3 cl of pure alcohol the last time they had an alcoholic beverage. The countries next at the top are Denmark (7.5 cl), Ireland (7.3 cl), Greenland, Malta (7.1 cl each) and the Czech Republic (7.0 cl). Countries where the students had been drinking rather small quantities include Romania, Portugal (3.3 cl each) and Ukraine (3.5 cl). Thus, students in the top countries had been drinking more than twice as much as students in the countries with the smallest consumption. There are of course also differences in the consumption pattern as regards beverage types. The largest proportion of beer, out of the total amount consumed, was found in the Netherlands where 65% of the consumption on the last drinking occasion was beer. Other countries with a large proportion of beer are Romania (61%), Iceland (59%), Turkey (58%) and Denmark (56%). Countries with the highest proportions of wine out of the total consumption include Slovenia (35% of the total amount consumed), Croatia, Hungary (28%) and the Slovak Republic (27%). Spirits make up the highest proportion in Greece (61%), the Faroe Islands (59%), Greenland (46%), the Slovak Republic and Portugal (45%). It is of course important to remember that these countries are rather different in total amounts consumed – the percentages mentioned only show the relations between the consumed volumes of different beverages on the last drinking occasion. The country that reports the highest average volume consumed among boys is the Faroe Islands. These boys had consumed 9.4 cl of pure alcohol on the last occasion. The countries next at the top are Malta (9.0 cl), the Czech Republic (8.8 cl), Denmark (8.7 cl) and Poland (8.4 cl). The top countries among girls are the Faroe Islands (6.7 cl), Greenland, Ireland (6.4 cl each) and Denmark (6.1 cl). There are, however, large discrepancies in consumed quantities between the genders in some countries, while in others the differences are smaller. In Greenland, Iceland, Ireland, the Isle of Man and

149

a) The estimates are based on the reported consumption of beer, wine and spirits only. It should be noted, that in some, but not all, countries alcopops and/or cider are available. The effect of their inclusion in the estimates is discussed in the section “Last drinking occasion”.

4,0 % 4,1 5,0 % 5,1 6,0 % 6,1 6,9 % 7,0

% ; ; ; ; ;; ; ; ; ; ; ; ; ; ;

Data uncertain or not available Non-participating country

; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ;

; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ;

; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ;

; ; ; ; ; ; ; ; ; ; ; ; ; ;

; ; ; ; ; ; ; ; ; ; ; ; ; ;

Figure 41a. Estimated average consumptiona) of beer, wine and spirits, in cl 100% alcohol, on the last drinking occasion. All students. 2003. Turkey: Limited geographical coverage.

Boys

Girls

9,4 8 7,9 9 8,8 7,2 8 8,5 8,3 5,7 6,7 7,2 6,8 7,5 6,7 7,1 6,2 6,1 6,4 5,9 5,9 5,8 5,6 5,3 5,1 5,2 5,1 5,1 4,8 4,3 4,4 % 10

8

6

6,7 6,1 6,4 6,4

Faroe Islands (8,3) Denmark (7,5) Ireland (7,3) Greenland (7,1) Malta (7,1) Czech Republic (7,0) Iceland (6,6) Estonia (6,5) Poland (6,4) United Kingdom (6,4) Greece (6,2) Isle of Man (5,9) Croatia (5,7) Finland (5,7) Netherlands (5,7) Norway (5,7) Lithuania (5,5) Slovenia (5,4) Sweden (5,1) Slovak Republic (4,9) Belgium (4,7) Italy (4,6) Bulgaria (4,5) Hungary (4,3) Russia (Moscow) (4,3) Latvia (4,0) Turkey (6 cities) (4,0) Cyprus (3,9) Switzerland (3,9) Ukraine (3,5) Portugal (3,3) Romania (3,3)

8,7

4

2

0

5,5 5,4 5,8 4,2 4,2 5,6 3,6 5,4 4,1 4,3 3,6 5,2 3,5 4,4 4,1 3,6 3,2 3,3 3,4 3 3,2 3,1 2,8 2,4 2,5 2,4 2,4 2,1 0

2

4

6

8

10 %

Figure 41b. Estimated average consumptiona) of beer, wine and spirits, in cl 100% alcohol, on the last drinking occasion. Boys and girls. 2003. Values within brackets refer to all students. Data sorted by all students. Turkey: Limited geographical coverage.

150

The alcohol and drug situation 2003

Norway the girls’ consumption on the last drinking occasion is about 80% of that of the boys. Other countries with relatively small differences are Denmark, the Faroe Islands, Slovenia, Sweden and the United Kingdom where the girls’ consumption is about 70% of that of the boys’. The largest differences are found in Romania, where the girls had been drinking alcohol to an amount of 47% of the boys’, and in Cyprus, Lithuania, the Netherlands, Poland and Switzerland where it makes up a little less than 50%. Beer, wine, alcopops, cider and spirits (Tables 17a–c) In many countries cider and alcopops are parts of the alcohol assortment available in shops. These beverages are sometimes important in relation to young peoples consumption, they are often sweet and tasty and they are promoted with flashy attractive labels. However, they are not available in all ESPAD countries, why the addition of them into the calculation of quantities consumed at the last drinking occasion makes it difficult to make comparisons. In tables 17 a–c the average alcohol consumption on the last drinking occasion is presented with the inclusion of alcopops and cider for countries in which these beverages are available. As mentioned in previous section, the calculations are made under the assumption that alcopops contain 4.5% alcohol and cider 5.0%. Questions on all five beverages were included in the questionnaires of 17 countries. However, in addition to Austria and Germany also Switzerland had changed the format for the questions on alcopops and cider, which puts a limit to the possibilities of doing comparisons. The question on alcopops was included in the questionnaires of 28 countries, out of which three countries used a format deviating from the ESPAD standard format (Austria, Germany and Switzerland). The results on cider and alcopops for these countries are presented below the line in the tables. The addition of the two beverages is important in those countries where this is appropriate. The average consumption rises from 7.3 to 11.2 cl pure alcohol in Ireland, from 5.9 to 10.3 cl in Isle of Man, from 6.4 to 10.2 cl in the United Kingdom and from 5.7 to 9.5 cl in Norway. On average the consumption increases with 1 cl pure alcohol per beverage, i.e. from 5cl on average for beer, wine and spirits to 6cl including alcopops and to 7cl if also cider is included. The effect on the distribution of beverages is

The alcohol and drug situation 2003

mainly affecting the girls’ consumption. The alcopops and cider proportions of the total alcohol consumed is overall more important in relation to girls’ alcohol consumption than to that of boys’, i.e. these beverages add usually more to the amounts consumed by girls.

Drunkenness Lifetime (Tables 18a–c, figures 42a–b) In 30 of the 35 countries studied the majority of the students have been drunk at least once. The countries with the highest figures in which three fourths or more of the student population have been drunk include Denmark (85%), Lithuania (81%), Estonia (80%), Isle of Man (79%), the Czech Republic, Ukraine (78% each), Austria, Ireland (76% each) and the United Kingdom (75%). The lowest proportions are reported from Turkey (21%) and Portugal (32%). Other countries where less than half of the students have experienced drunkenness include Cyprus (38%), France (43%) and Malta (47%). Some students who have been drunk have a rather limited experience of the phenomenon. Others, however, get intoxicated more frequently. The countries with the highest percentages indicating that students have been drunk 20 times or more in lifetime include Denmark (36%), Ireland (30%), Isle of Man (29%), the United Kingdom (27%), Estonia and Finland (26% each). In other countries only a few students report this frequency of drunkenness. In Turkey 1% had been drunk 20 times or more, in Cyprus 2% and in France, Greece, Portugal and Romania 3% gave this answer. In a majority of the countries there are more boys than girls reporting this frequency of intoxication. In no country are the girls in majority. However, in quite many countries the gender distribution is rather even. These countries include both the British Isles and most of the Nordic countries (Finland, the Faroe Islands, Iceland, Ireland, Isle of Man, Norway, Sweden and the United Kingdom). Last 12 months (Tables 19a–c, figures 43a–b) Many students who report lifetime experience of drunkenness probably refer to a rather recent event. Consequently the 12 months prevalence rates are rather close to the lifetime measures. In about 20 of the 35 ESPAD countries a majority of the students report having been drunk during the last 12 months.

151

; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ;

3% 4 9% 10 15 % 16 25 % 26

; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ;

; ; ; ; ; ; ; ; ; ;

% ; ; ; ; ;; ; ; ; ; ; ; ; ; ;

Data uncertain or not available Non-participating country

; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ;

; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ;

; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ;

; ; ; ; ; ; ; ; ; ; ; ; ; ;

; ; ; ; ; ; ; ; ; ; ; ; ; ;

Figure 42a. Proportion of all students who have been drunk 20 times or more in lifetime. 2003. Germany and Turkey: Limited geographical coverage. USA: Limited comparability.

Boys

Girls Denmark (36) Ireland (30) Isle of Man (29) United Kingdom (27) Estonia (26) Finland (26) Faroe Islands (24) Austria (21) Greenland (21) Lithuania (21) Czech Republic (18) Ukraine (18) Sweden (17) Iceland (16) Russia (Moscow) (15) Slovenia (15) Latvia (14) Norway (14) Slovak Republic (14) Germany (6 Bundesl.) (12) Hungary (11) Bulgaria (10) Poland (10) Switzerland (10) Croatia (9) Belgium (7) USA (7) Netherlands (6) Italy (5) Malta (4) France (3) Greece (3) Portugal (3) Romania (3) Cyprus (2) 0 0 Turkey (6 cities) (1)

41 32 28 27 33 25 26 27 24 29 25 24 18 16 18 20 19 14 20 16 16 15 15 14 14 11 10 9 8 7 4 4 5 7 2 3 % 50

40

30

20

10

0

0

31 29 29 27 19 28 23 13 18 12 13 11 15 15 13 10 10 13 10 8 5 7 5 6 5 3 6 4 3 3 2 3 2 2

10

20

30

40

50 %

Figure 42b. Proportion of boys and girls who have been drunk 20 times or more in lifetime. 2003. Values within brackets refer to all students. Data sorted by all students. Germany and Turkey: Limited geographical coverage. USA: Limited comparability.

152

The alcohol and drug situation 2003

; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ;

3% 4 9% 10 18 % 19 25 % 26

; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ;

; ; ; ; ; ; ; ; ; ;

% ; ; ; ; ;; ; ; ; ; ; ; ; ; ;

Data uncertain or not available Non-participating country

; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ;

; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ;

; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ;

; ; ; ; ; ; ; ; ; ; ; ; ; ;

; ; ; ; ; ; ; ; ; ; ; ; ; ;

Figure 43a. Proportion of all students who have been drunk 10 times or more during last 12 months. 2003. Germany and Turkey: Limited geographical coverage. USA: Limited comparability.

Boys

Girls

40

Denmark (34) Ireland (29) United Kingdom (24) Finland (23) Faroe Islands (22) Isle of Man (22) Estonia (21) Austria (20) Greenland (19) Czech Republic (16) Lithuania (15) Sweden (15) Iceland (14) Norway (14) Slovenia (13) Ukraine (13) Russia (Moscow) (12) Germany (6 Bundesl.) (11) Slovak Republic (11) Bulgaria (10) Latvia (10) Hungary (9) Switzerland (9) Croatia (8) Poland (8) Belgium (7) USA (7) Netherlands (6) 2 Italy (4) Malta (4) 1 Romania (4) 1 France (2) 2 Greece (2) 1 Portugal (2) 0 Turkey (6 cities) (2) 0 Cyprus (1)

30 24 21 23 19 26 26 19 21 20 15 14 14 15 17 14 15 16 15 12 12 12 12 12 10 8 8 7 6 4 4 3 4 2 2 % 40

30

20

10

0

0

29 28 25 25 22 25 17 13 19 12 9 13 15 14 10 9 9 8 7 6 7 5 6 5 4 4 6 4 4

10

20

30

40 %

Figure 43b. Proportion of boys and girls who have been drunk 10 times or more during last 12 months. 2003. Values within brackets refer to all students. Data sorted by all students. Germany and Turkey: Limited geographical coverage. USA: Limited comparability.

The alcohol and drug situation 2003

153

The highest 12 months prevalence rates were found in Denmark (82%), Ireland (72%), Isle of Man (71%), Greenland (70%), Austria (69%), the Czech Republic, Estonia, the United Kingdom (68% each), Lithuania and Ukraine (66% each). Much lower figures were recorded in Turkey (16%), Cyprus (25%), Portugal (28%) and France (29%). The number of students who have been drunk 10 times or more during the last 12 months are highest in Denmark and Ireland, where about one third of the students reported this (34 and 29% respectively). Other countries with high proportions on this variable include the United Kingdom (24%), Finland (23%), the Faroe Islands, Isle of Man (22% each), Estonia (21%), Austria (20%) and Greenland (19%). In half of the ESPAD countries, the number of students who had been drunk 10 times or more during the last 12 months make up one tenth of the populations. Very low percentages are reported from Cyprus, France, Greece, Portugal and Turkey (2% each). In a majority of the ESPAD countries there are more boys than girls who had been drunk as often as 10 times or more during the last 12 months. In only two countries there are slightly more girls than boys (Finland and Isle of Man). In the Faroe Islands, Greenland, Iceland, Ireland, Norway, Sweden and the United Kingdom, however, no gender differences are found. Hence, in all British Isles and all Nordic countries but Denmark, girls have been intoxicated rather frequently and to at least the same extent as boys. In addition, in some low prevalence countries it can be observed that the gender differences are small mainly because the prevalence rates are small.

Last 30 days (Tables 20a–c, figures 44a–b) The response categories in Austria and Germany were changed into an open format. Since this is expected to influence the comparability with other countries, the results from these countries are put below the line in the tables and are excluded from the graphs. The number of students who have been drunk as recently as during the last 30 days differs considerably between countries, from 8 to 60%. The highest figure is observed in Denmark, where 61% of the students had been drunk recently and which value is well above the second highest prevalence country, which is Ireland (53%). Other countries with high figures include Greenland, the Isle of Man (49% each) and the United Kingdom (46%).

154

On the other hand, in some countries this frequency of drunkenness is much less common. They are mainly found in the south, including Turkey (8%), Cyprus (10%), Portugal (14%), France, Romania (15% each) and Greece (16%). Looking at the number of students who have been drunk 3 times or more during the last 30 days implicates that the figures are smaller, but the pattern over the countries remain about the same. Thus, Denmark is still at the top together with Ireland, in which countries one fourth of the students had been drunk that often. Other countries with high prevalence rates include Isle of Man and the United Kingdom (23% each). In about half of the ESPAD countries the number of students reporting this frequency of intoxication is 10% or less. The lowest figures are reported from Turkey (1%), Cyprus (2%), France, Greece, Portugal (3% each) and Romania (4%). In a majority of the countries there are more boys than girls reporting this behaviour. A larger proportion of girls that report being drunk at least 3 times during the last 30 days is mainly found in Isle of Man. In others, the gender distribution is rather equal, which is the case in Finland, Greenland, Iceland, Ireland, Norway, Sweden and the United Kingdom. Thus in all British Isles and almost all Nordic countries there are at least as many girls as boys that had been drunk 3 times or more during the last 30 days.

Binge drinking (Tables 21a–c, figures 45a–b) Having five or more drinks in a row (binge drinking) would for most students of this age mean getting drunk. Thus, the distribution of responses in various countries to the question on how many times this amount had been consumed over the last 30 days would be expected to vary in about the same way as was the case in relation to drunkenness. This is also true to a large extent. The response categories in Austria and Germany were changed into an open format. Since this is expected to influence the comparability with other countries, the results from these countries are put below the line in the tables and are excluded from the graphs. The highest percentage of students who reported this is found in Denmark, where it was indicated by a majority of the students (60%). Other countries where more than half of the students had indicated this are the Netherlands (58%), Germany, Ireland, Isle of Man (57% each), the United Kingdom

The alcohol and drug situation 2003

4% 5 9% 10 15 % 16 20 % 21

% ; ; ; ; ;; ; ; ; ; ; ; ; ; ;

Data uncertain or not available Non-participating country

; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ;

; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ;

; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ;

; ; ; ; ; ; ; ; ; ; ; ; ; ;

; ; ; ; ; ; ; ; ; ; ; ; ; ;

Figure 44a. Proportion of all students who have been drunk 3 times or more during the last 30 days. 2003. Turkey: Limited geographical coverage. USA: Limited comparability.

Boys

Girls

30

Denmark (26) Ireland (26) Isle of Man (23) United Kingdom (23) Greenland (19) Faroe Islands (18) Estonia (17) Finland (16) Ukraine (16) Czech Republic (13) Lithuania (12) Norway (12) Slovenia (12) Russia (Moscow) (11) Slovak Republic (11) Bulgaria (10) Iceland (10) Poland (10) Sweden (10) Hungary (9) Switzerland (9) Belgium (8) Croatia (8) Latvia (8) Italy (7) Netherlands (7) USA (7) Malta (5) Romania (4) France (3) Greece (3) Portugal (3) Cyprus (2) Turkey (6 cities) (1)

27 20 22 20 20 23 15 18 17 17 12 16 13 14 17 9 13 11 11 12 12 12 12 9 10 8 7 6 5 3 6 3 4 % 30

20

10

0

21 25 25 25 19 16 13 17 10 10 8 10 8 9 8 8 9 5 9 5 6 4 5 7 3 4 6 4 3 2 3 2 0 1 0

10

20

30 %

Figure 44b. Proportion of boys and girls who have been drunk 3 times or more during the last 30 days. 2003. Values within brackets refer to all students. Data sorted by all students. Turkey: Limited geographical coverage. USA: Limited comparability.

The alcohol and drug situation 2003

155

9% 10 14 % 15 20 % 21 25 % 26

% ; ; ; ; ;; ; ; ; ; ; ; ; ; ;

Data uncertain or not available Non-participating country

; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ;

; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ;

; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ;

; ; ; ; ; ; ; ; ; ; ; ; ; ;

; ; ; ; ; ; ; ; ; ; ; ; ; ;

Figure 45a. Proportion of all students who reported “binge drinking” 3 times or more during the last 30 days. 2003. Turkey: Limited geographical coverage. USA: Limited comparability.

Boys

Girls

31 37 26 26 32 27 31 25 28 24 23 28 26 26 21 23 24 22 20 19 17 20 21 19 19 14 13 17 19 15 13 10 12 9 % 40

30

20

10

33

Ireland (32) Netherlands (28) Isle of Man (27) United Kingdom (27) Malta (25) Sweden (25) Denmark (24) Norway (24) Belgium (22) Latvia (22) Slovenia (22) Ukraine (22) Bulgaria (21) Estonia (20) Faroe Islands (19) Greenland (19) Czech Republic (18) Russia (Moscow) (17) Portugal (16) Croatia (15) Finland (15) Slovak Republic (15) Switzerland (15) Italy (13) Lithuania (13) Greece (11) Iceland (11) Poland (11) Romania (11) Cyprus (10) France (9) USA (9) Hungary (8) Turkey (6 cities) (5) 0

20 30 29 19 22 18 24 14 18 18 15 16 15 17 16 13 12 10 10 15 12 11 8 7 8 9 5 5 5 7 10 5 2 0

10

20

30

40 %

Figure 45b. Proportion of boys and girls who reported “binge drinking” 3 times or more during the last 30 days. 2003. Values within brackets refer to all students. Data sorted by all students. Turkey: Limited geographical coverage. USA: Limited comparability.

156

The alcohol and drug situation 2003

(54%), Belgium and Malta (50% each). As can be expected since many other alcohol variables in this country show low figures, very few students in Turkey (15%) had reported this. Other countries with rather few students indicating this behaviour include Romania (23%), Portugal (25%) and France (28%). A more frequent binge drinking, i.e. 3 times or more during the last 30 days, is reported by one fifth to one third of the students in about half of the ESPAD countries. The ranking order is not exactly the same as for the total prevalence, even if many of them are appearing in both groups. The highest numbers of students having been binge drinking 3 times or more during the last 30 days are found in Ireland (32%), Germany, the Netherlands (28% each), Isle of Man, the United Kingdom (27% each), Malta, Poland, Sweden (25% each), Denmark and Norway (24% each). Thus, there is a concentration of countries in the northern and western parts of Europe with Malta as the only exception. The countries with the lowest binge drinking figures are Turkey (5%), Hungary (8%), France (9%), Cyprus, Greece, Iceland and Romania (11% each). The value for USA on this variable (9%) is comparable to that of France.

Age at first use of alcohol and first drunkenness Beer, wine and spirits (Table 22) In a majority of the participating countries about half of the students or more have consumed at least one glass of beer or wine at the age of 13 years or younger. It is less common, however, to have tasted spirits (at least one glass) at this age – in about half of the countries this is reported by one third. In countries with the highest number of students that have tried beer at the age of 13, about two thirds of the students or more had done so. Many of these countries can be categorised as traditional “beer countries”, but this is not sufficient to explain the distribution over Europe. These high proportions are found in Latvia (72%), Slovenia (69%), Bulgaria, Denmark, Lithuania (67% each) and Ukraine (66%). Other countries with almost as high figures include Estonia (64%), Russia (Moscow) (62%), Isle of Man, the United Kingdom (61% each), Germany and the Slovak Republic (60% each). The lowest percentage in relation to beer drinking at an early age is found in Turkey where 19%

The alcohol and drug situation 2003

had this experience. Other countries with lower figures include Iceland (34%), Norway (39%), the Faroe Islands and Portugal (41% each). In four countries about two thirds of the students had been drinking wine at the age of 13 or younger. They include Lithuania (73%), Isle of Man, Slovenia (66% each) and the United Kingdom (65%). Much less students reported this behaviour in Turkey (11%), Norway (26%), Iceland, Portugal (27% each), the Faroe Islands (28%) and the Netherlands (29%). Rather few students had been drinking spirits at an early age. However, in four countries almost half of the students reported that they had done so. They include Denmark (48%), Isle of Man (47%), the United Kingdom (44%) and Malta (41%). Much lower figures were found in Turkey (7%), Romania (15%), Iceland and Norway (18% each). In all countries but one, there are more boys than girls that had been drinking beer at the age of 13. The only exception is Russia (Moscow) where the proportions were the same. The tendency with higher frequencies among boys is the same in most countries when it comes to wine consumption. However, in six countries spread all over Europe the gender distributions were about the same (Austria, Bulgaria, Germany, Greenland, the Netherlands and Norway). The same tendency with larger proportions among boys than girls are found for spirits in about two thirds of the countries. However, in about one third of the countries, spread all over Europe, there were rather equal proportions among boys and girls that had been drinking spirits at the age of 13. There are clear differences between different types of beverages in the proportion of students that have reported use at the age of 13 or younger. When looking at the averages of all ESPAD countries many more have indicated beer or wine (54 and 49% respectively) compared to spirits (30%). When looking at individual countries the number of students that have been drinking spirits at this young age is smallest in all countries, while beer is dominant in more countries than wine. However, in about half of the countries there are no big difference between beer and wine. Overall, the figures indicates that beer is the most common beverage among the youngest consumers (13 years or younger) in the ESPAD countries. In most countries the differences related to beverage types are about the same among boys as well as girls. However, the dominate role of beer is more visible among boys.

157

Drunkenness (Table 22, figures 46a–b) It is clear that many students in most ESPAD countries have tried alcohol at a fairly young age. The consumption has, however, not lead to intoxication to the same extent. The proportions of students that report having been drunk at the age of 13 or younger vary quite substantially between countries. About one fourth of the students in ten countries report that they experienced their first intoxication at the age of 13 or younger. In other countries the percentages are much lower, e.g. in two countries in which less than 10 percent reported this behaviour. The top country in relation to having been drunk at the age of 13 or younger is Isle of Man (38%) followed by Russia (Moscow) (37%). The figures are also high (33–36%) in Denmark, Estonia, Finland and the United Kingdom. The two countries with the lowest figures include Turkey (5%) and Cyprus (7%). In four countries this behaviour is reported by 10–11%, including Greece, Italy, Portugal and Switzerland. In a large majority of the countries there are more boys than girls reporting drunkenness at the age of 13. However, the proportions are rather similar in quite many countries including Austria, Finland, the Faroe Islands, Greenland, Iceland, Ireland, Norway, Malta and the United Kingdom. With the exception of Austria and Malta they all are British Isles and Nordic countries.

Drinking places (Tables 23a–c) To explore in which context the students usually consume alcohol, they were asked: “Think about the last day on which you drank alcohol. Where were you when you drank?” The response categories were “I never drink alcohol”, “At home”, At someone else’s home”, “Out on the street, in a park, beach or other open area”, “At a bar or a pub”, “In a disco”, “In a restaurant” and “Other place”. To be able to group the countries according to the most common answers, the two highest scores in each country has been counted. The response alternative “At someone else’s home” scored highest in comparison to the others. The countries with the highest proportions on this alternative are found in the Nordic countries and in the Baltic states. They include Denmark (66%), Greenland (61%), Norway (50%), Finland, Sweden

158

(43% each), Estonia (42%) and Lithuania (41%). The second most frequent choice was “At home”. Countries where most students have indicated this alternative include Romania (38%), Isle of Man (34%), Cyprus (33%) and the United Kingdom (30%). A disco is a place where many ESPAD students had been drinking alcohol on the last drinking occasion. Countries with most students indicating this alternative include Cyprus (48%), Austria (34%), Malta (32%), Greece (31%) and the Czech Republic (30%). A bar or a pub was almost as frequently indicated as a disco as the place where students had a drink at the last drinking occasion. The highest scores are observed in Austria, Italy (36% each), the Czech Republic (35%), Croatia (34%), Portugal (31%) and the Slovak Republic (30%), i.e. only countries in the central and southern parts of Europe. Outdoors, such as in the street, in a park or at a beach, was answered by 14% as an average. The highest proportions indicating this alternative were found in Russia, where 33% had said so, Latvia (31%) and Poland (30%). Very few answered that they had been drinking in a restaurant the last time they had alcohol, and this alternative are not among the two most frequent chosen by the students in any country. “Other places”, on the other hand, was frequently indicated in Greece (36%), Germany and Norway (22% each). There are only small differences between places where boys and girls drink alcohol. The most important differences are found for outdoor places (the street, a park or the beach), which have been reported by more boys than girls. To sum up, the places most frequently indicated by ESPAD students as the scene for their last drinking occasion are someone else’s home or their own home. It would be of interest to know to which extent existing alcohol regulations and laws might influence the choice of a bar, pub or a disco as a place at which young people drink alcohol. At least in some countries this option is rather limited, since the personnel would not be allowed to serve underage people. Finally, many students in Greece, Germany and Norway indicated “other places” to a rather high extent. It is difficult to see what the students in these countries might have in common when choosing this alternative.

The alcohol and drug situation 2003

; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ;

11 % 12 15 % 16 22 % 23 30 % 31

; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ;

; ; ; ; ; ; ; ; ; ;

% ; ; ; ; ;; ; ; ; ; ; ; ; ; ;

Data uncertain or not available Non-participating country

; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ;

; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ;

; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ;

; ; ; ; ; ; ; ; ; ; ; ; ; ;

; ; ; ; ; ; ; ; ; ; ; ; ; ;

Figure 46a. Proportion of all students who have been drunk at the age of 13 or younger. 2003. Germany and Turkey: Limited geographical coverage. USA: Limited comparability.

Boys

Girls

36 36 42 37 32 29 33 25 30 30 27 23 25 24 27 27 25 21 20 18 17 22 20 16 20 17 15 13 14 13 12 10 8 % 50

40

30

20

10

39

Isle of Man (38) Russia (Moscow) (37) United Kingdom (36) Estonia (35) Denmark (34) Finland (33) Greenland (29) Ukraine (26) Ireland (24) Croatia (23) Lithuania (23) Slovenia (23) Austria (22) Bulgaria (22) Germany (6 Bundesl.) (22) Latvia (22) Slovak Republic (22) Sweden (22) Czech Republic (19) Faroe Islands (19) Iceland (17) Norway (17) Romania (16) Belgium (15) Netherlands (14) Poland (14) Hungary (13) Malta (13) Switzerland (11) Greece (10) Italy (10) Portugal (10) USA (9) Cyprus (7) Turkey (6 cities) (5)

40

0

34 35 27 32 34 28 19 22 15 17 19 20 20 20 18 17 19 16 18 15 18 11 11 12 9 9 12 9 7 8 9 5 2 0

10

20

30

40

50 %

Figure 46b. Proportion of boys and girls who have been drunk at the age of 13 or younger. 2003. Values within brackets refer to all students. Data sorted by all students. Germany and Turkey: Limited geographical coverage. USA: Limited comparability.

The alcohol and drug situation 2003

159

Expected personal consequences (Table 24a–c, figure 47) The expected consequences of alcohol use vary considerable both between individuals and across countries. Different cultures promote different patterns of alcohol consumption as well as different psychosocial effects of intoxication. Also within countries, individuals adopt different drinking patterns and are experiencing the effects of alcohol in different ways. The students were asked to indicate how likely they thought that different positive and negative consequences would happen to them if they drink alcohol. The five proposed positive consequences included “Feel relaxed”, “Feel happy”, “Feel more friendly and outgoing”, “Have a lot of fun” and “Forget my problems”. The six proposed negative consequences included “Feel sick”, “Get a hangover”, “Not be able to stop drinking”, “Harm my health”, “Do something I would regret” and “Get into trouble with the police”. The proportions of students in each country responding “likely” or “very likely” to each question are presented in tables 24a–24c. Most students associate their alcohol consumption with having fun. A large majority (68% on average) anticipate this as a possible consequence.

Other positive consequences, which more than half of the students on average had indicated, included “feel more friendly and outgoing”, “feel happy”, and “feel relaxed”. Least support has the alternative “forget my problems” (45% on average). Among the negative consequences “harm my health” is the most anticipated, which 42% on average indicated. In regressing order the following alternatives are “get a hangover”, “do something I would regret”, “feel sick” and “get into trouble with the police”. The least expected consequence among these young people is “not be able to stop drinking”, which on average was indicated by 14%. Countries where most students on average had indicated positive consequences include Denmark, the Faroe Islands, Ireland, Isle of Man and the United Kingdom, which about three quarters of the students in these countries report. When looking at expected negative consequences the countries with the highest average proportions (around 45%) indicating any of those include Bulgaria, Croatia, the Faroe Islands, Romania, Slovenia and Italy. It seems as if the Faroese students to a high extent have anticipated positive as well as negative consequences when drinking.

"Negative" consequences

"Positive" consequences

1 1 2 2 2 3 3 3 4 4

No.

5 6

5

4

3

2

1

5

Finland Ireland United Kingdom Czech Republic Denmark Isle of Man Russia (Moscow) Ukraine Austria Estonia Germany (6 Bundesl.) Netherlands Slovak Republic Belgium Greece Sweden Switzerland Bulgaria Greenland Iceland Lithuania Malta Norway Faroe Islands Hungary Cyprus Latvia Slovenia Croatia Poland Turkey (6 cities) Italy Portugal Romania 0

4 4 3 3 3 3 3 2 2 2 2 2 1 1 1 1

No.

0

1

2

3

4

5

6

Figure 47. Anticipated positive and negative consequences of alcohol consumption. Number of statements for which the percentage of all students answering “likely” or “very likely” exceeds the average of all countries. Germany and Turkey: Limited geographical coverage.

160

The alcohol and drug situation 2003

To give an overview of the anticipated positive and negative consequences of alcohol use, figure 46 presents the sums of the proportions of students in each country that agreed with the different statements. Thus, for each of the five positive consequences, if the individual country’s proportion exceeds the average for all countries on this variable this country gets one point on this item. In the same way five of the negative consequences (the sixth, least anticipated consequence “not being able to stop drinking” was excluded to balance the scale) are used to summarise the negative side. To balance the positive and the negative consequences, each country’s positive points minus its negative points make up the value for this country. This means that the result might be a positive or a negative value, or it might be indifferent. In the figure all countries are presented with their summarised points. Thus, as can be seen in the figure students in Finland seem to be the most positive in their attitudes towards alcohol, with a total sum of +5 points. Other countries with most positive scores are Ireland, the United Kingdom (+4 points each) and the Czech Republic, Denmark, Isle of Man, Russia (Moscow) and Ukraine (+3 points each). In each of these countries students overall anticipate more positive and less negative consequences of their own alcohol consumption than in other ESPAD countries. It is notable that most of these countries are among those with highest drunkenness figures. On the negative side we mainly find the countries that most often are associated with low prevalence rates on alcohol consumption and drunkenness. They include Romania (-5 points), Italy, Portugal (-4 points each) as well as Croatia, Poland and Turkey (-3 points each). In these countries, students overall anticipate more negative and less positive consequences of their alcohol consumption than their counterparts in other participating countries.

Experienced problems caused by own alcohol use (Tables 25a:1 – 25c:2, figures 48–49) The students were also asked if they had encountered any problems related to alcohol use, drug use or related to some other reasons. The number of students who had experienced problems related to drug use was very low in almost all ESPAD countries, and is therefore not presented in this report. Rather many, however, had experienced various problems in relation to their own alcohol use. The fourteen problems listed in the question-

The alcohol and drug situation 2003

naire have been grouped into four categories. These categories are “Individual problems”, “Relationship problems”, “Sexual problems” and “Delinquency problems”. Included in “Individual problems” are the following items: “Performed poorly at school or at work”, “Damage to objects or clothing”, “Loss of money or other valuable items”, “Accident or injury” and “Hospitalised or admixed to an emergency room”. The problem most often indicated by the students in this group is “damage to objects or clothing” which on average had been indicated by 12%. The next in ranking are “loss of money or other valuable items” and “accident or injury”, which are indicated by about 8 and 6% respectively. The other two categories are only mentioned by 2–3% of the students. The highest average percentages of students indicating any of the individual problems are found in Lithuania (14%), Ireland, Isle of Man, the United Kingdom (13%) and Denmark (12%). The smallest proportions are found in Cyprus, France, Greece, Turkey (2%), Belgium, Italy, Malta, Portugal and Switzerland (3%). Included in “Relationship problems” are the following items: “Quarrel or argument”, “Problems in relationships with friends”, “Problems in relationships with parents”, “Problems in relationships with teachers”. The problem most indicated in this group is “quarrel or argument” which on average is indicated by 11%. The next most frequently indicated items are “problems in relationships with parents” (8%) and “problems in relationships with friends” (6%). Only 2% had indicated problems with teachers. The individual countries that for this group of problems have the highest average percentages include Lithuania (19%), Denmark (15%), Finland (12%), Greenland, Ireland and Isle of Man (10% each). Very few students have indicated these types of problems in Cyprus, Greece, Turkey (2%), Italy, the Netherlands and Portugal (3%). The problem group “Sexual problems” includes two items: “Engaged in sex you regretted the next day” and “Engaged in unprotected sex (without a condom)”. Both these alternatives are on average rather equally indicated (about 5%). One country (Ireland) had left out these two items in the questionnaire. Looking at the countries individually reveals that these problems are by far most experienced by

161

Denmark Isle of Man Finland Lithuania Ireland* Latvia United Kingdom Bulgaria Estonia Greenland Poland Russia (Moscow) Ukraine Faroe Islands Sweden Norway Slovak Republic Czech Republic Iceland Slovenia Austria Germany (6 Bundesl.) Belgium Croatia Cyprus France Greece Hungary Italy Malta Netherlands Portugal Romania Switzerland Turkey (6 cities)

Individual

Sexual

Relationship

Delinquency

5 5 5

3 3 3 4 3

4 5 3 4 4 2 1 3 2 3 3

1 2

2

3 3 0 1 0 1 1 0

2 2 1 0 1 0 1 0

3 3 2 3

1 3

4 0

2

0

2 2 2 3 2

2 4

2 1 1 3 0 1 3 2 1 4 0 3 3 0 2 4 1 1 2 2 0 2 1 2 0 2 0

0 0 0 0 0 0 0 0 0 0 0 0 0

No.

0

3

6

9

12

15

Figure 48. Experienced problems caused by alcohol. The number of vari* ables within each “problem group” for which a country’s percentage exceeds the average of all countries. All students. Germany and Turkey: Limited geographical coverage. * Not all alternatives were included.

Hospitalised

2

Problem with teachers

2

Robbed

2

29

40

Unprotected sex

4

Trouble with police

4

9 10

Accident or injury

5 6 6

Problem with friends

6

Regret sex

44 45

7

Fight

Other reasons

14

3

Poor school

Alcohol 35

Loss of valuables

8

Problem with parents

8

39 41 45 11

Argument

69

12

Damaged possessions 0

37

20

40

60

80

%

Figure 49. Cross-national average of students who report having each of 14 problems because of their alcohol use and because of other reasons.

162

The alcohol and drug situation 2003

the youth in Greenland and Isle of Man, where 17 and 13% respectively had indicated that they had experienced any of these two sexual experiences. Other countries with rather high figures on this variable are Denmark and the United Kingdom (9%) followed by Finland (8%). “Delinquency problems” included the items “Scuffle or fight”, “Victimised by robbery or theft” and “Trouble with police”. Of these the first one is the most often indicated, although the average proportion for all countries is relatively low (7%). The individual country that scores highest on this group of problems is Lithuania (10%), followed by Ireland, Isle of Man (9% each), Denmark and the United Kingdom (8% each). Very few students in Cyprus and Greece indicated this kind of problems (1% each), but also in Belgium, France, Italy, Malta, the Netherlands, Portugal, Switzerland and Turkey (2% each). For most of the problem groups the average scores do not indicate any clear gender pattern. The average scores on individual, relationships and sexual problems are the same or about the same for both boys and girls. The only group of experienced problems that reveals a gender difference is the delinquency problems group. On average more boys than girls indicated this (6 vs. 3%). The individual consequence that boys by far are more involved in is a scuffle or fight, which on average 10% of the boys had indicated compared to 5% of the girls. The pattern of rather small differences between boys and girls is also found in most individual countries. When there are differences the figure is usually higher among boys. However, in a few countries some of the problem types are mainly found among girls. This is the case in the Faroe Islands where more girls have reported sexual problems related to their alcohol consumption. Other countries include Finland (individual, relationship as well as sexual problems), Greenland (sexual problems), Iceland (sexual problems), Isle of Man (individual, relationship and sexual problems), Sweden (sexual problems) and the United Kingdom (individual, relationship and sexual problems). In all these seven countries, which only are found in the British Isles and among the Nordic countries, more girls have reported sexual problems related to their own alcohol consumption. In figure 48 the pattern of experienced problems in different countries is shown by counting for each

The alcohol and drug situation 2003

country the number of items on which the country scores higher than average. Thus, for each of the 14 problems and for each country, the number of items for which it scores above average are counted and summarised. The highest sum of items exceeding average was found in Denmark and Isle of Man (13 over average) followed by Finland, Lithuania (12 over), Ireland (11, however the two variables on sexual problems were omitted in the Irish questionnaire), Latvia and the United Kingdom (10). In other words, among the countries with most reported alcohol related problems are all the British Isles countries as well as Nordic countries and Baltic states. Countries that have no variable scores exceeding average are predominantly Mediterranean countries. They are also found in all other parts of Europe with the exception of the British Isles, the Nordic countries and the Baltic states. In order to assess the relative role of alcohol in different types of problems, the students were also asked about their experiences of the same problems for reasons other than their own alcohol use. Figure 49 shows the cross-national average of students who report having each of the 14 problems because of their alcohol use and because of other reasons. In most cases, the number of problems that the respondents specifically related to their own alcohol use was small in comparison with such problems caused by other factors. The exceptions from this pattern is the problem defined as “engaged in sexual intercourse you regretted the next day” (regret sex), which is indicated to the same extent because of alcohol as well as for other reasons. Two other variables that are only somewhat more connected with other reasons than alcohol are “unprotected sex” (engaged in sexual intercourses without a condom) (4 vs. 9%) and “trouble with police” (4 vs. 10%). A conclusion that can be drawn from the results on the two variables “expected consequences” and “problems because of alcohol use” is that many of the countries with students that report expected positive experiences from alcohol consumption are found at the top of the list of countries that report problems. It also seems as if young people in the south of Europe expect more problems to be asso ciated with alcohol consumption, but report less experience of such problems.

163

; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ;

; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ;

; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ;

; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ;                              

9% 10 16 % 17 25 % 26 39 % 40

                             

         

% ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ;

Data uncertain or not available Non-participating country

; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ;

; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ;

; ;

              ; ; ; ; ; ;

                    

                    

                    

             

             

Figure 50a. Lifetime experience of any illicit drug. Percentages among all students. 2003. Germany and Turkey: Limited geographical coverage. Spain and USA: Limited comparability.

Boys

Girls

48 45 42 41 42 43 42 39 37 33 32 31 33 29 32 28 25 24 27 24 26 29 25 21 21 18 21 15 11 13 9 9 10 8 7 8 5 % 50

40

30

20

10

40

Czech Republic (44) Switzerland (41) USA (41) Ireland (40) Isle of Man (40) France (38) United Kingdom (38) Spain (36) Belgium (33) Germany (6 Bundesl.) (30) Netherlands (29) Slovenia (29) Italy (28) Greenland (27) Slovak Republic (27) Estonia (24) Austria (23) Croatia (23) Denmark (23) Bulgaria (22) Russia (Moscow) (22) Ukraine (21) Poland (19) Portugal (18) Latvia (17) Hungary (16) Lithuania (16) Iceland (13) Finland (11) Malta (11) Faroe Islands (10) Norway (9) Sweden (8) Greece (6) Turkey (6 cities) (5) Cyprus (5) Romania (3) 0

37 40 40 39 34 35 34 28 27 24 27 24 26 22 19 21 22 19 19 19 12 14 15 13 14 10 11 12 9 10 10 7 5 3 3 2 0

10

20

30

40

50 %

Figure 50b. Lifetime experience of any illicit drug. Percentages among boys and girls. 2003. Values within brackets refer to all students. Data sorted by all students. Germany and Turkey: Limited geographical coverage. Spain and USA: Limited comparability.

164

The alcohol and drug situation 2003

Illicit drugs In this section the prevalence of use of illicit drugs, tranquillisers or sedatives (with and without a doctor’s prescription), anabolic steroids, alcohol in combination with pills and use of inhalants will be presented. Overall, the focus is on lifetime prevalence, except for illicit drugs for which also 12 months and 30 days prevalence rates are presented. The section begins with a presentation of the students’ knowledge about various illicit drugs.

Knowledge about drugs (Tables 26a–c) The prevalence of drug use differs widely across countries. In some countries both the knowledge of a drug and the use of it are rather widespread, while students in other countries have never heard the name, let alone having used it. To explore how well known certain substances are, also in low prevalence countries, and to be able to monitor possible changes over time, the students were asked if they had ever heard of certain drugs. The drugs included in this question are amphetamines, crack, cocaine, ecstasy, heroin, LSD, marijuana/hashish, methadone and tranquillisers or sedatives. On average, the most well known drugs are marijuana or hashish, cocaine and heroin, which a large majority (90% on average) indicated that they had heard of. The next substance in this hierarchy is ecstasy, which 83% on average had heard about. A group of drugs, including amphetamines, LSD and crack, were all known to about the same extent on average (60–66%) among the students. The least known substance was GHB which only 18% indicated knowledge about. Countries that score highest on average in relation to familiarity with the drugs listed are Isle of Man (79%) and the United Kingdom(78%). Other countries with high values (75–76%) include the Czech Republic, Denmark, Ireland, the Netherlands and Sweden. Countries where rather few students were familiar with these drug names include Turkey (34%), Greenland (39%) and Ukraine (43%). There are only small differences between boys and girls when averages are compared. However, it might be worth to note that, on average, there are more girls than boys that have heard about tranquillisers or sedatives (70 vs. 62%). The same tendency is also found in a vast majority of the countries. For some of the drugs there are substantial differences between countries in relation to the students’ knowledge. One example is LSD that only

The alcohol and drug situation 2003

17% of the Greenlandic and 20% of the Romanian students had heard of compared with 91% in Germany. Of the students in Turkey only 8% had heard about crack and in Romania only 19%. As typical countries at the other end of the scale, this was reported by about 90% in five countries (Germany, Ireland, Isle of Man, Sweden and the United Kingdom). The knowledge about GHB differs substantially between countries, from 4–5% in the Faroe Islands and Turkey to 55% in Iceland and 48% in Norway. The discrepancies are also large in relation to methadone. The smallest proportions that had heard about methadone were found in Turkey (7%) and Greenland (11%) while this was the case among 77% in Norway and 72% in Ireland. The range is wide also for magic mushrooms. For this drug the lowest figure was found in Turkey (11%) followed by Cyprus (13%). On the other hand, there are four countries in which around 90% of the students had heard about magic mushrooms (the Czech Republic, Ireland, Isle of Man and the United Kingdom).

Any illicit drug Lifetime (Tables 27a–c, figures 50a–b) The concept “any illicit drug” includes marijuana or hashish, amphetamines, LSD or other hallucinogens, crack, cocaine, ecstasy and heroin. The lifetime prevalence of any illicit drug varies considerably across the ESPAD countries. The highest prevalence rates of any illicit drug use are reported from the Czech Republic (44%), Switzerland (41%), Ireland and the Isle of Man (40% each). Other countries with high proportions include France, the United Kingdom (38% each), Belgium (33%), Germany (30%), the Netherlands, Slovenia (29% each), Italy (28%), Greenland and the Slovak Republic (27% each). A majority of these countries are found in the central and western parts of Europe, while only three are found in the eastern parts. Less than 10% have reported such drug use in Romania (3%), Cyprus, Turkey (5% each), Sweden (8%) and Norway (9%). Other countries with proportions around ten percent are the Faroe Islands (10%), Finland and Malta (11% each). In Spain 36% of the students have used an illicit drug. The corresponding figure for USA is 41%. Many of the students have only tried a drug once

165

; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ;

; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ;

; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ;

; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ;                              

7% 8 14 % 15 25 % 26 35 % 36

                             

         

% ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ;

Data uncertain or not available Non-participating country

; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ;

; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ;

; ;

              ; ; ; ; ; ;

                    

                    

                    

             

             

Figure 51a. Lifetime experience of marijuana or hashish. Percentages among all students. 2003. Germany and Turkey: Limited geographical coverage. Spain and USA: Limited comparability.

Boys

Girls Czech Republic (44) Switzerland (40) Ireland (39) Isle of Man (39) France (38) United Kingdom (38) Spain (36) USA (36) Belgium (32) Netherlands (28) Slovenia (28) Germany (6 Bundesl.) (27) Greenland (27) Italy (27) Slovak Republic (27) Denmark (23) Estonia (23) Croatia (22) Russia (Moscow) (22) Austria (21) Bulgaria (21) Ukraine (21) Poland (18) Hungary (16) Latvia (16) Portugal (15) Iceland (13) Lithuania (13) Finland (11) Malta (10) Faroe Islands (9) Norway (9) Sweden (7) Greece (6) Cyprus (4) Turkey (6 cities) (4) Romania (3)

48 44 38 41 42 41 39 38 37 32 31 31 29 31 32 27 28 24 26 23 23 29 23 18 20 18 14 18 11 13 9 9 9 7 7 6 4 % 50

40

30

20

10

0

40 36 39 38 35 35 33 34 28 24 26 24 26 23 22 18 18 20 18 18 19 12 13 13 12 12 11 9 11 8 10 9 6 5 2 2 2 0

10

20

30

40

50 %

Figure 51b. Lifetime experience of marijuana or hashish. Percentages among boys and girls. 2003. Values within brackets refer to all students. Data sorted by all students. Germany and Turkey: Limited geographical coverage. Spain and USA: Limited comparability.

166

The alcohol and drug situation 2003

; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ;

; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ;

; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ;

; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ;                              

2% 3 5% 6 10 % 11 17 % 18

                             

         

% ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ;

Data uncertain or not available Non-participating country

; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ;

; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ;

; ;

              ; ; ; ; ; ;

                    

                    

                    

             

             

Figure 52a. Proportion of all students who have used marijuana or hashish during the last 30 days. 2003. Germany and Turkey: Limited geographical coverage. Spain and USA: Limited comparability.

Boys

Girls

25 26 24 23 23

Spain (23) France (22) Isle of Man (21) Switzerland (20) United Kingdom (20) Czech Republic (19) Belgium (17) Ireland (17) USA (17) Italy (15) Slovenia (14) Netherlands (13) Germany (6 Bundesl.) (12) Greenland (11) 7 Austria (10) Slovak Republic (10) 7 Bulgaria (8) 7 Croatia (8) 5 Denmark (8) 5 Poland (8) 5 Portugal (8) 6 Russia (Moscow) (7) 4 Estonia (6) 5 Hungary (6) 3 Lithuania (6) 2 Ukraine (5) 4 Iceland (4) 2 Latvia (4) 3 Malta (4) 2 Finland (3) 2 Norway (3) 1 Cyprus (2) 2 Greece (2) 1 Turkey (6 cities) (2) 1 Faroe Islands (1) 0 Romania (1) 1 Sweden (1)

21 20 16 19 19 14 17 14 12 12 10 10 9 10 10 11 7 8 7 8 8 4 5 5 3 3 3 2 3 2 1 2 % 30

20

10

0

0

22 18 19 17 16 17 13 17 15 12 14 9 9 11 9

10

20

30 %

Figure 52b. Proportion of boys and girls who have used marijuana or hashish during the last 30 days.2003. Values within brackets refer to all students. Data sorted by all students. Germany and Turkey: Limited geographical coverage. Spain and USA: Limited comparability.

The alcohol and drug situation 2003

167

or twice, while others have had a more or less regular habit of drug taking. Countries where the highest percentages of students have used any drug 20 times or more include Switzerland (16%), France (15%), the United Kingdom (14%), the Czech Republic, Isle of Man (13%), Belgium (11%), Ireland, Italy and the Netherlands (10% each), i.e. about the same top countries as for lifetime prevalence. In contrast, only 1% or less report this in Cyprus, the Faroe Islands, Romania and Sweden. The gender pattern reveals that in a majority of the countries more boys than girls report that they have tried any illicit drug at least 20 times. In no country the opposite is true. On the other hand, in a number of countries the proportions are similar for boys and girls. If one exclude countries with only small percentages, this is mainly found in Croatia and Slovenia.

Marijuana or hashish Lifetime (Tables 28a–c, figures 51a–b) The vast majority of the students in all ESPAD countries that have tried any illicit drug have used marijuana or hashish. Thus, the number of students reporting experience with cannabis are almost identical with the total illicit drug prevalences. The top country in this respect is the Czech Republic where 44% of the students have used marijuana or hashish. Still high prevalence rates are reported from Switzerland (40%), Ireland, Isle of Man (39% each), France and the United Kingdom (38% each). Other countries where more than one fourth of the students have used cannabis include Belgium (32%), the Netherlands, Slovenia (28% each), Germany, Greenland, Italy and the Slovak Republic (27% each). The lowest levels of cannabis use are reported from Romania (3%), Cyprus, Turkey (4% each), Greece (6%) and Sweden (7%). Low prevalence rates are also found in the Faroe Islands, Norway (9% each) and Finland (10%). These low prevalence countries are either found in the south of Europe or among the Nordic countries. Data from Spain and the USA reveal that 36% of the students in both countries have ever used cannabis. In no country there are more girls than boys that have tried cannabis, and boys are in majority in about two thirds of the ESPAD countries. In some of them, on the other hand, there are no real gender differences. Those countries are mainly found in the British Isles or among the Nordic countries,

168

including the Faroe Islands, Finland, Greenland, Iceland, Ireland, Isle of Man, Norway and Sweden. However, included in the list is also a southern country (Greece). It may also be noted that the countries with about equal proportions between the sexes are both high and low prevalence countries.

Last 12 months and last 30 days (Tables 29a–c, figures 52a–b) Many of the students who have tried marijuana or hashish have apparently done so during the last year. Thus, the number of students indicating that they have used cannabis during the last 12 months is very similar to the lifetime prevalence of this drug. The highest number of students that had used cannabis during the last year is found in the Czech Republic (36%). Other high prevalence countries are Isle of Man (34%), France, Ireland, Switzerland and the United Kingdom (31% each). Countries where very few students have used cannabis during the last 12 months are to a large extent the same that reported low lifetime prevalence rates. Thus, the smallest number of students reporting this behaviour are found in Romania (2%), Cyprus, Turkey (3% each), the Faroe Islands (4%), Greece and Sweden (5% each). In Spain 32% of the students have used cannabis during the last 12 months. The corresponding value for the USA is 28%. Use of cannabis during the last 30 days usually indicates an active and ongoing habit. In some countries about one fifth of the students reports this, in others much lower prevalence rates are noted. The countries with the highest 30 days prevalence include France (22%), Isle of Man (21%), Switzerland, the United Kingdom (20% each) and the Czech Republic (19%). Other countries with somewhat high rates are Belgium, Ireland (17% each) and Italy (15%). In some countries however, very few report cannabis use during the last 30 days. The six countries with the lowest figures include the Faroe Islands, Romania, Sweden (1% each), Cyprus, Greece and Turkey (2% each). Data from Spain and USA reveals that 23% and 17% respectively of the students in these countries have used cannabis during the last 30 days. In many of the high prevalence countries there are more boys than girls indicating that they have used cannabis during the last 12 months. However, countries where no or only small gender differences can be seen include Ireland, Slovenia, Greenland, the Slovak Republic, Bulgaria, Croatia, Rus-

The alcohol and drug situation 2003

                             

2% 3 4% 5 6% 7 8% 9

                             

         

% ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ;

Data uncertain or not available Non-participating country

; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ;

; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ;

; ;

              ; ; ; ; ; ;

                    

                    

                    

             

             

Figure 53a. Lifetime experience of any illicit drug other than marijuana or hashish. Percentages among all students. 2003. Germany and Turkey: Limited geographical coverage. Spain: Limited comparability. Boys

Girls

11 9 10 8 9 11 8 9 11 8 8 9 8 6 7 7 8 6 6 5 5 4 5 3 4 5 4 2 2 3 4 1 3 2 3 % 15

10

5

12

Czech Republic (11) Estonia (10) Germany (6 Bundesl.) (10) Isle of Man (10) Ireland (9) United Kingdom (9) Spain (9) Austria (8) Belgium (8) Italy (8) France (7) Lithuania (7) Poland (7) Portugal (7) Croatia (6) Denmark (6) Iceland (6) Netherlands (6) Slovak Republic (6) Switzerland (6) Hungary (5) Latvia (5) Slovenia (5) Bulgaria (4) Greenland (4) Malta (4) Russia (Moscow) (4) Cyprus (3) Finland (3) Norway (3) Sweden (3) Turkey (6 cities) (3) Faroe Islands (2) Greece (2) Romania (2) Ukraine (2)

10

0

11 10 10 10 9 8 8 7 6 7 6 6 6 6 5 5 5 5 5 5 4 5 4 4 4 4 1 3 3 2 2 3 2 1 1 0

5

10

15 %

Figure 53b. Lifetime experience of any illicit drug other than marijuana or hashish. Percentages among boys and girls. 2003. Values within brackets refer to all students. Data sorted by all students. Germany and Turkey: Limited geographical coverage. Spain: Limited comparability.

The alcohol and drug situation 2003

169

sia (Moscow), Hungary and Iceland. Thus, countries with rather equal gender pattern do not seem to have any geographical concentration. However, in some countries the prevalence rates are so low that no gender pattern can be established.

Any illicit drug other than marijuana or hashish Lifetime, last 12 months and last 30 days (Tables 30a–c, 31a–c, 32a–c, 33a–c, 34a–c, figures 53a–b) As was established above, the most important and prevalent drug in all ESPAD countries is cannabis. Nevertheless, many students have also used other substances, and in some cases without any additional experience of cannabis. In tables 30 a–c and 31 a–c the lifetime, 12 months and 30 days prevalence rates of any other drug than cannabis are presented. In tables 32 a–c the lifetime prevalence of specific drugs such as amphetamines, LSD or other hallucinogens, crack, cocaine, heroin, ecstasy, magic mushrooms, GHB (gammahydroxybuturate), as well as any drug by injection are presented. Overall, the prevalence rates on these substances are relatively low. The ESPAD average is 6% with a range of 2–11%. Of those who have used any other drug than cannabis a majority have done so 1–5 times in their lives. Students, who have used any illicit drug other than marijuana or hashish, make up about one tenth of the total study population in countries with the highest prevalence rates. They include the Czech Republic 11%, Estonia, Germany, Isle of Man (10% each), Ireland and the United Kingdom (9% each). In nine ESPAD countries 3% or less report any experience of such drugs. The countries with the lowest prevalence rates include the Faroe Islands, Greece, Romania and Ukraine (2% each). A majority of those who have ever used any drug other than cannabis have done so rather recently. Therefore the 12 months prevalence rates are rather similar to the lifetime rates. The average for all countries on lifetime use is 6% and the average for 12 months 4%. The highest 12 months prevalence rates for these types of illicit drugs are found in Isle of Man (10%), Austria, the Czech Republic, Germany (7% each), Estonia, Ireland and Italy (6% each). Very few students had used such a drug during the last 12 months in the Faroe Islands, Finland, Romania and Turkey (1% each). The 30 days prevalence is on average 2% for all countries. The highest figures are found in Austria

170

(4%), Belgium, the Czech Republic, Germany, Ireland, the Netherlands and the United Kingdom (3% each). Very low prevalence rates are observed in some countries. Values of only 1% or below are reported from the Faroe Islands, Finland, Greece, Norway, Romania, Russia (Moscow), the Slovak Republic, Sweden, Turkey and Ukraine. The gender pattern is rather homogeneous both for the lifetime, 12 months and 30 days prevalence rates. Tables 32a–c show the prevalence rates for individual drugs. Overall the rates are low, but in a few individual countries they are higher. Besides cannabis, the most commonly used illicit drug is ecstasy, which 3% on average have indicated. The average rates for amphetamines, LSD or other hallucinogens, cocaine and magic mushrooms are all the same (2%). Crack, heroin and any drug by injection was on average mentioned by 1% of the students. Very few (0%) had indicated experience of GHB. The countries with the highest percentages of students reporting use of amphetamines are Estonia (7%), Germany, Iceland, Lithuania and Poland (5% each). On the other hand, in 13 countries 1% or less reported such use. Very few students have used LSD or other hallucinogens. The highest percentages are found in the Czech Republic and Isle of Man where 5–6% reported this. The use of crack or cocaine is also very limited. The highest value is observed in relation to cocaine and this is found in the Isle of Man, Italy and the United Kingdom, where 4% reported use. Around 1% on average had ever used heroin. The single highest value is found in Italy where 4% gave this answer. Ecstasy is, apart from cannabis, the most used drug of those included in the questionnaire. In the Czech Republic 8% had used it, followed by Isle of Man (7%), Croatia, Estonia, Ireland, the Netherlands and the United Kingdom (5% each). Magic mushrooms are not very frequently used in the majority of the countries. However, a few countries are more outstanding in reported use, such as the Czech Republic (8%), Isle of Man (7%), Belgium, France, Germany and the Netherlands (5% each). The lifetime use of GHB is limited to 1% of the students or less in all ESPAD countries. Another practically non-existent habit is drug taking with use of a needle (drugs by injection).

The alcohol and drug situation 2003

; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ;

2% 3% 4 6% 7 9%

; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ;

; ; ; ; ; ; ; ; ; ;

10 % ; ; ; ; ;; ; ; ; ; ; ; ; ; ;

Data uncertain or not available Non-participating country

; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ;

; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ;

; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ;

; ; ; ; ; ; ; ; ; ; ; ; ; ;

; ; ; ; ; ; ; ; ; ; ; ; ; ;

Figure 54a. Lifetime experience of tranquillisers or sedatives without a doctors prescription. Percentages among all students. 2003. Germany and Turkey: Limited geographical coverage. Boys

Girls

12 8 7 9 5 8 7 4 4 5 5 4 5 6 4 3 3 4 3 3 3 2 2 3 2 3 1 2 1 2 3 2 1 % 25

20

15

10

5

22

Poland (17) Lithuania (14) France (13) Czech Republic (11) Hungary (10) Belgium (9) Estonia (9) Iceland (9) Netherlands (8) Finland (7) Croatia (6) Italy (6) Sweden (6) Switzerland (6) Faroe Islands (5) Isle of Man (5) Portugal (5) Romania (5) Slovenia (5) Denmark (4) Greece (4) Slovak Republic (4) Greenland (3) Latvia (3) Malta (3) Norway (3) Russia (Moscow) (3) Turkey (6 cities) (3) Austria (2) Bulgaria (2) Germany (6 Bundesl.) (2) Ireland (2) Ukraine (2) United Kingdom (2) Cyprus (1)

10 10

0

18 15 14 13 10 13 10 10 9 9 7 7 7 5 4 7 7 8 5 5 5 4 4 3 3 3 3 2 2 2 2 1 1 1 0

5

10

15

20

25 %

Figure 54b. Lifetime experience of tranquillisers or sedatives without a doctors prescription. Percentages among boys and girls. 2003. Values within brackets refer to all students. Data sorted by all students. Germany and Turkey: Limited geographical coverage.

The alcohol and drug situation 2003

171

; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ;

3% 4 5% 6 8% 9 11 % 12

; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ;

; ; ; ; ; ; ; ; ; ;

% ; ; ; ; ;; ; ; ; ; ; ; ; ; ;

Data uncertain or not available Non-participating country

; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ;

; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ;

; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ;

; ; ; ; ; ; ; ; ; ; ; ; ; ;

; ; ; ; ; ; ; ; ; ; ; ; ; ;

Figure 55a. Lifetime experience of alcohol together with pills. Percentages among all students. 2003. Germany and Turkey: Limited geographical coverage.

Boys

Girls

10 11

Germany (6 Bundesl.) (16) Slovak Republic (15) Austria (13) Czech Republic (12) Finland (12) Hungary (11) Faroe Islands (10) Isle of Man (10) Croatia (9) Ireland (9) Malta (9) Poland (9) Iceland (8) Sweden (8) Denmark (7) France (7) Lithuania (7) United Kingdom (7) Belgium (6) Estonia (6) Latvia (6) Russia (Moscow) (6) Slovenia (6) Norway (5) Bulgaria (4) Netherlands (4) Portugal (4) Switzerland (4) Ukraine (4) 2 Italy (3) Romania (3) 3 Greece (2) 2 Greenland (2) 1 Turkey (6 cities) (2) Cyprus (0) 0

8 7 5 8 4 9 7 6 7 6 6 5 6 5 6 6 6 4 5 6 4 3 3 5 3 4 4 4 2 2 2 3 1 % 25

20

15

10

5

0

0

22 18 20 15 18 13 16 11 12 13 11 11 11 12 8 10 8 8 6 8 7 6 9 6 5 4 4 5 4 4

5

10

15

20

25 %

Figure 55b. Lifetime experience of alcohol together with pills. Percentages among boys and girls. 2003. Values within brackets refer to all students. Data sorted by all students. Germany and Turkey: Limited geographical coverage.

172

The alcohol and drug situation 2003

; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ;

5% 6 7% 8 12 % 13 16 % 17

; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ;

; ; ; ; ; ; ; ; ; ;

% ; ; ; ; ;; ; ; ; ; ; ; ; ; ;

Data uncertain or not available Non-participating country

; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ;

; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ;

; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ;

; ; ; ; ; ; ; ; ; ; ; ; ; ;

; ; ; ; ; ; ; ; ; ; ; ; ; ;

Figure 56a. Lifetime experience of inhalants. Percentages among all students. 2003. Germany and Turkey: Limited geographical coverage.

Boys

Girls Greenland (22) Isle of Man (19) Cypern (18) Ireland (18) Malta (16) Greece (15) Slovenia (15) Austria (14) Croatia (14) Iceland (12) United Kingdom (12) Faroe Islands (11) France (11) Germany (6 Bundesl.) (11) Czech Republic (9) Poland (9) Slovak Republic (9) Denmark (8) Estonia (8) Finland (8) Portugal (8) Sweden (8) Belgium (7) Latvia (7) Russia (Moscow) (7) Switzerland (7) Italy (6) Netherlands (6) Ukraine (6) Hungary (5) Lithuania (5) Norway (5) Turkey (6 cities) (4) Bulgaria (3) Romania (2)

23 18 19 14 16 17 15 14 14 12 12 10 12 12 9 10 10 9 9 8 10 8 9 8 7 9 8 7 9 6 6 6 5 4 2 % 25

20

15

10

5

0

22 20 16 21 15 13 15 14 14 11 13 13 10 11 9 8 7 7 7 8 6 8 5 7 6 6 5 5 4 4 4 4 3 3 1 0

5

10

15

20

25 %

Figure 56b. Lifetime experience of inhalants. Percentages among boys and girls. 2003. Values within brackets refer to all students. Data sorted by all students. Germany and Turkey: Limited geographical coverage.

The alcohol and drug situation 2003

173

The 12 months and 30 days prevalence of use of different drugs other than cannabis are overall very low in a majority of the countries; 1–2% or less report any use. However, in a few countries the 12 months figures mounts to 3–4%. Use of amphetamines during the last 12 months is reported by 4% in Austria and by 3% in Denmark, Estonia, Germany, Iceland, Lithuania and Poland. LSD or other hallucinogens during the last 12 months are reported by 3% in the Czech Republic, cocaine by 3% in the United Kingdom, heroin by 3% in Italy, ecstasy by 5% in the Czech Republic and 3% in Belgium, the Netherlands and the United Kingdom. Magic mushrooms were used by 4% in the Czech Republic and by 3% in Belgium, Germany, Italy and the Netherlands. The highest 30 days prevalence is noted for amphetamines in Austria (3%) and ecstasy in Croatia (3%).

Tranquillisers, anabolic steroids, alcohol together with pills Lifetime (Tables 35a–c, figures 54a–b, 55a–b) Tranquillisers or sedatives can be used both as a legally prescribed medicine and as an illicit drug. The majority of the students that have used any such drug have used a prescribed medicine, with an average of 8% for prescribed drugs and 4% when not prescribed. The prevalence rates differ however rather much over the countries. The highest percentages of students that have used tranquillisers or sedatives prescribed by a doctor are found in the Czech Republic (20%), France (17%), Belgium, Croatia, Iceland, Portugal and the Slovak Republic (14–15%). Rather low figures, on the other hand, are found in Cyprus (1%), the Faroe Islands (3%), Austria, Bulgaria, Greece, Isle of Man and the United Kingdom (4% each). Use of tranquillisers or sedatives without prescription is most common in Poland (17%) followed by Lithuania (14%), France (13%) and the Czech Republic (11%). Similar to the legally prescribed use, the lowest prevalence rates are found in Cyprus (1%), Austria, Bulgaria, Germany, Ireland, Ukraine and the United Kingdom (2% each). Very few students in most ESPAD countries have ever used anabolic steroids. The use of these substances is mainly associated with athletic training and bodybuilding. Only few students in the ESPAD countries reported such use. The highest number of students is found in Poland and Turkey (3% each).

174

It is well known that young people sometimes combine the use of pills with alcohol with the anticipation of getting a synergetic effect. The prevalence rates of “alcohol together with pills” are highest in Germany (16%), the Slovak Republic (15%), Austria (13%), the Czech Republic and Finland (12% each). Low prevalence countries for this variable are Cyprus (0%), Greece, Greenland and Turkey (2% each). The ESPAD students were also asked if they used to combine alcohol and cannabis. This behaviour is much more frequent than to combine alcohol with a pill. Almost one third of the students in the Czech Republic, Ireland, Isle of Man, Switzerland and the United Kingdom reported use of alcohol and cannabis at the same time. As a contrast, only 1% of the students in Cyprus and Romania had experienced this. Looking at the distributions by gender reveals that, on average, there are more girls that report having used tranquillisers or sedatives without prescription as well as alcohol together with pills. On the other hand, there are more boys than girls that have used alcohol and cannabis at the same time. A more frequent use of alcohol together with pills among girls is reported from about half of the countries. In the remaining countries the figures are to a large extent the same for both sexes. However, no country reports that more boys than girls have done this. The situation is similar for the use of alcohol and cannabis at the same time, but with boys in the majority. In about half of the countries there are more boys than girls that have tried this, while no country reported the opposite. In about half of them there are only small or no gender differences in the reported figures.

Use of inhalants (Tables 36a–c, figures 56a–b) The students were asked: “On how many occasions (if any) have you sniffed a substance (sniffing glue, aerosols etc.) to get high?” The highest lifetime prevalence rates are reported from countries in very different parts of Europe. The top country on lifetime prevalence is Greenland, where 22% had done so. Other countries with high levels of inhalants use include Isle of Man (19%), Cyprus, Ireland (18% each), Malta (16%), Greece and Slovenia (15% each). In Romania as well as Bulgaria the figures are as low as 2–3%. Other low prevalence countries include Turkey (4%), Hungary, Lithuania and Norway (5%).

The alcohol and drug situation 2003

Some of those who declared experience of inhalants may have tried it rather long time ago and is perhaps no longer using it. The last 12 months prevalence rates are lower, but the highest figures are found in about the same countries as for lifetime prevalence. The highest rates of use of inhalants during the last 12 months are reported from Greenland (16%), Cyprus, Isle of Man (11% each), Ireland and Malta (10% each). As can be expected the 30 days prevalence rates are lower. The highest values are found, again, in about the same countries as for lifetime and 12 months prevalence figures. The highest percentages of students who have used inhalants during the last 30 days are found in Cyprus (6%), Greece and Malta (5% each). Very small gender differences are found in the use of inhalants. In a majority of the countries there are no differences, but in Belgium, Cyprus, Greece, Portugal and Ukraine more boys than girls reported this behaviour. In one country only, Ireland, more girls than boys have used inhalants. It is striking that the high prevalence countries to a large extent are islands. It is difficult to see why this is so. A possible explanation might be that the social control in smaller societies might make it more difficult for young people to get hold of other illegal substances.

Onset First drug used (Tables 37a–c) The students were asked about the first illicit drug they ever used. The drugs listed were tranquillisers or sedatives, marijuana or hashish, LSD, amphetamines, crack, cocaine, heroin, ecstasy, magic mushrooms and GHB. The most important illegal substance as a debut drug is cannabis. This was on average answered by 18% of all students, which corresponds to about 80% of all students that have tried any illicit drug. The “dominance” of cannabis is also found in all, but three, of the ESPAD countries. In more than half of them cannabis was mentioned as the first illicit drug by 80% or more of the students that had tried any such drug. Second to cannabis, but with much lower figures, are tranquillisers or sedatives. This was reported by 2% of all students which is about 9% of all students that have tried any illicit drug. Rather high prevalence rates for tranquillisers and sedatives are mainly reported from Poland and Lithuania in which 35–40% of the “drug users” gave this answer.

The alcohol and drug situation 2003

A comparison between the sexes shows that more boys than girls used cannabis as their first drug. However, the opposite is true for tranquillisers and sedatives which, on average, was more common among girls, and this especially true in Lithuania and Poland.

How the first drug was obtained (Tables 38a–c) The students were asked how they obtained the drug on the first occasion. The responses were given in a fixed format including 13 alternatives. The results in the tables are summarised in nine groups, one of which is “I have never used any illicit drug”. There are three alternatives that seem to apply for most of the students. They are: “It was shared in a group”, “Given by an older friend” and “Given by a friend of the same age or younger”. Each of them was on average indicated by 5% of the student population, which corresponds to about 20– 36

Czech Republic

82

33

France

77

12

Netherlands

77 30

Ireland

73 43

Italy

72

29

Slovenia

71 34

Belgium Germany (6 Bundesl.)

28

United Kingdom

27

70 69 68

19

Slovak Republic

68

21

Switzerland

63

15

Austria

63 20

Croatia

62

12

Denmark

62 26

Isle of Man

61

12

Norway

61

7

Lithuania

60

3

Faroe Islands

58 7

Finland

55 23

Poland

53

5

Malta

15

Bulgaria 11

Greece

47

6

Latvia

46 11

Hungary Iceland

6

Cyprus

6

44 43 41

3

Greenland

41 7

Sweden

35

4

Russia (Moscow)

31

6

Romania Ukraine

3

Turkey (6 cities)

3

0

Any place

48

9

Estonia

Schools

50

14

Portugal

53 52

27 20 17

20

40

60

80

100 %

Figure 57. The proportion of students who reported that cannabis was easily available in schools and any other place respectively. Germany and Turkey: Limited geographical coverage.

175

25% of those who have tried any illicit drug. This means that about 70% of all “drug users” had mentioned any of these three answers. In individual countries one of these categories is more important than others. One example is “shared in a group”, which was especially dominant in the Czech Republic, Estonia, Russia (Moscow) and Slovenia. Another, and even more striking example, is that “given by older friend” was by far the most frequent answer in Greenland. It means, that this is how about 60% of the Greenlandic “drug users” were introduced to illegal drugs. Rather few students answered that they bought the first drug they used, either from a friend or someone else. Taken together, these answers are on average given by 13% of those that have tried any illicit drug. With one exception, this way of getting the first drug is uncommon in all countries. The only exception is Malta in which about one third of the those who had used any drug answered that their first drug was bought from a friend. No specific gender pattern is observed for this variable.

Reasons for first use By mistake the answers to the question about the reason for the first drug use were calculated on all students and not limited to “drug users” only. This was not realised until it was too late to ask for recalculated figures to be put in this report. However, in spite of this some written comments will be made. The main results from the 2003 data collection reveals that the pattern of responses is very similar to the answers given in 1999. The dominant reason for the first drug use is that the students were curious. On average this was answered by about two thirds of all students that had tried any illicit drug. With one exception, this is the outstanding reason in all countries. The major exception is Greenland in which the most important single reason for the first drug use was a wish to feel high. This was answered by about one third of the “drug users”. There was also relatively many Greenlandic “drug users” (about 30%) that answered “other reasons”. The second most important reason for the first drug use was “wanted to get high”, which on average was answered by about 20% of the “drug users”. Other reasons were given to a much smaller extent than curiosity and a wish to feel high. This also includes “wanted to forget my problems”, a category that was mentioned by a little more than

176

10% of all students that had used any illicit drug. The gender differences are small. However, in some countries there are slightly more boys than girls that answered that they wanted to feel high the first time they tried an illegal drug.

Age at first use (Table 39) The two most common drugs that have been used at the age of 13 or younger are cannabis and inhalants. On average 4% had used cannabis and 3% inhalants at this very young age. The highest figures on early consumption of cannabis are found in the United Kingdom and the Isle of Man, where about 13% answered this. Other countries with relatively high numbers reporting this are Switzerland (11%), Germany (9%), Ireland and the Netherlands (8% each). Marijuana or hashish is also the drug that most students in USA have used at the age of 13. Students who reported that they used inhalants at the age of 13 are predominantly found in Cyprus (10%), but also in Croatia, Greenland and the Isle of Man (7%), followed by Austria, Greece, Ireland and Slovenia (6%). The other drugs listed are only occasionally mentioned as debut drugs. No country reports that more than 1% indicated LSD or ecstasy as their first drug (with the exception of Isle of Man with 2% for LSD). Tranquillisers or sedatives are indicated by 2% on average. The country with the highest percentage indicating this is Poland (4%), followed by Belgium, Estonia, Greenland, Lithuania and the Netherlands (3%). Very small gender differences are observed. As an average boys tend to indicate cannabis or inhalants at a somewhat higher degree than girls, but the differences are very small.

Places to buy cannabis (Tables 40a–c, figure 57) The students were asked: “In which of the following places do you think you could easily buy marijuana or hashish if you wanted to?” The results show that there are rather large differences between countries in the extent to which the students thought they had knowledge of any such place. The European countries where most students think they know of any place to buy cannabis include the Czech Republic (82%), the Netherlands (77%), Ireland (73%), Italy (72%), Slovenia (71%) and Belgium (70%). In other countries, however, rather few students could specify a place where

The alcohol and drug situation 2003

they would be able to buy cannabis. They include Turkey (17%), Ukraine (20%), Romania (27%), Russia (31%) and Sweden (35%). It is obvious that the awareness of any possibility to buy drugs is closely related to the prevalence rates in a country. Among the alternatives given, the place that on average is most frequently indicated is a disco or a bar. This was on average answered by 27% of the students. On second “ranking place” is “street, park etc.” (23%). Within the group indicating “disco/bar” the Czech Republic students are those who most frequently gave this answer (55%), followed by the Slovak Republic (46%), Germany (44%), Belgium (43%), Austria (42%) and Denmark (40%). Least common was this category among students in Ukraine (7%), Sweden (8%), Turkey (10%), Russia (12%) and Greenland (13%). Students who indicated “street/park etc.” are mainly found in Italy (45%), Slovenia (39%), Belgium (38%), Norway (37%) and Ireland (36%). Very few students have given this answer in Turkey (4%), Cyprus (5%), Ukraine (6%), Russia (8%) and Romania (9%). To have a possibility to buy cannabis at the house of a dealer was on average indicated by 21% of the students. Countries with rather high percentages of students giving this answer are Italy (43%), France (41%), the United Kingdom (39%) and Denmark (36%).

Schools are least indicated on this question (apart from “other places”). Despite the quite low average of 16%, rather high proportions gave this answer in a number of countries. They include Italy (43%), the Czech Republic (36%), Belgium (34%), France (33%) and Ireland (30%). In some countries many students answered “other places”. The highest figures are found in the Netherlands (64%), Norway (48%) and Belgium (38%). A major reason for this high figure in Belgium and the Netherlands was that an extra answering category “coffee shop” was used in tables 40a–c. These answers are included in the category “Other places”. The high figure for “Other places” in Norway include to a large extent names of places or streets where Norwegian students think that they can buy cannabis. There are on average more girls than boys that think that they can buy cannabis at a disco or a bar (30 vs. 24%), while it is the other way around for “school” (14 vs. 17%). When there are differences between boys and girls within countries, they usually follow this general pattern. Even though the averages are about the same for boys and girls when it comes to the category “house of a dealer”, this is more frequently indicated by girls than boys in a few individual countries, including Finland, Ireland, Isle of Man, Malta, Norway and the United Kingdom. On the other hand, more boys than girls gave this answer in Greenland.

Lifetime abstinence from various substances (Tables 41a–c) In tables 41a–c the rates of lifetime abstainers are given for each of the following substances: cigarettes, alcohol, illicit drugs, tranquillisers or sedatives and inhalants. In addition four calculated variables are presented in the table, which reflect the proportion of those who abstained from using different combinations of the previously listed substances. The average percentage of lifetime non-smokers is 34%. The highest rates of abstainers are found in Iceland (54%), Malta (52%), Greece and Turkey (50% each). On the other hand, the smallest numbers of lifetime non-smokers are found in the Faroe Islands (17%), Austria, the Czech Republic, Lithuania (20%), Greenland (21%), Latvia (22%) and Germany (23%).

The alcohol and drug situation 2003

In most countries rather few students reported lifetime abstinence of drinking alcohol. The average for all ESPAD countries is 11%. The highest value in this respect for an individual country is found in Turkey, which by far outreach most other countries, since more than half of the students (55%) never had been drinking alcohol. Other countries with relatively high percentages of alcohol abstainers are Iceland (25%), Portugal (22%) and Greenland (20%). The lowest rates are found in ten countries where less than 5% of the students had never used any alcohol. They include the Czech Republic, Lithuania (2%), the Slovak Republic (3%), Austria, Denmark, Estonia, Germany, Greece, Isle of Man and Latvia (4%). The average abstinence figure for illicit drug use (including marijuana or hashish, LSD, amphetamines, crack, cocaine, heroin and ecstasy) is 78%

177

for all ESPAD countries. The highest percentages of abstainers from these drugs in the individual countries are found in Cyprus, Turkey (95%), Greece (93%), Sweden (92%), Faroe Islands, Norway (91% each) and Romania (90%). The lowest rates are observed in the Czech Republic (56%), Ireland, Isle of Man (60% each), France and the United Kingdom (62% each). The high abstinence countries are found in the south of Europe and among the Nordic countries while the low abstinence countries include all countries of the British Isles. A large majority (95%) of the students in the ESPAD countries have never used tranquillisers or sedatives. There are, however, differences between individual countries, but they are not dramatic. The highest value, 98%, is observed in seven countries, including Austria, Germany, Ireland, Norway, the Slovak Republic, Ukraine and the United Kingdom. The lowest rate of non-users of these substances is found in Poland where 83% had never used it. Other countries with relatively low percentages are Lithuania (87%), France (88%) and the Czech Republic (89%). The average rate of abstinence from inhalants is 90%. The variation around this value ranges from 78% (Greenland) to 97% (Bulgaria and Romania). Other countries with low percentages of abstainers from inhalants use also include Isle of Man (81%), Ireland (82%), Malta (84%), Greece and Slovenia (85%). Apart from Bulgaria and Romania high percentages of abstainers are found also in Turkey (96%), Hungary, Lithuania and Norway (95%). Tables 41a–c also include figures representing abstinence rates for combinations of drugs. Thus, the a-category represents those that are abstainers from cigarettes as well as alcohol, b) cigarettes, alcohol and illicit drugs, c) cigarettes, alcohol, illicit drugs and tranquillisers/sedatives, d) cigarettes, alcohol, illicit drugs, tranquillisers/sedatives and inhalants. The countries vary in the proportions of students who are abstainers from any of the drugs included. Analysis of the sequence of figures for the four substance combinations reveals no difference in most countries or a change of only one percentage

178

point. This means that if students neither smoked nor used alcohol, they usually did not use any other substance either. Looking closer at the data reveals that the only thing that differs between countries is if, and when, the possible change occurs. For example in Malta, the Netherlands and Ukraine the small difference occurs between a) and b), i.e. the value decreases when illicit drugs are added. This means that some students, which not already are among those who use cigarettes or drink alcohol, have used illicit drugs, thus making the group who did neither of this a little smaller. In Cyprus (10, 10, 9, 9), Greenland (9, 9, 8, 8) and Lithuania (2, 2, 1, 1) the change happens when tranquillisers or sedatives are added, while in Portugal (14, 14, 14, 13) the inclusion of inhalants makes the total abstainers fewer. The gender pattern is of course the opposite of the gender pattern of the prevalence figures for these drugs. The average number of abstainers from cigarettes or alcohol seems to be very similar between boys and girls. However, there are lesser abstainers from illicit drugs among boys (75% on average) than among girls (81%). For tranquillisers/sedatives and inhalants the gender differences are on average very small. Changes in relation to the combinations are somewhat different between boys and girls. Among boys the changes occurred between a) and b) in Romania, between b) and c) in Italy and Malta and between c) and d) in Turkey. Among girls there was a larger variation than among boys as to the extent the students had used a drug without first “starting” with alcohol or cigarettes. Thus, in Croatia, Cyprus and Slovenia the changes were observed between c) and d), in Greenland between both a) and b) and c). In four countries, Norway, Portugal, Russia and Switzerland the change occurred between b) and c) and in Poland, Romania and Slovenia it happened between c) and d). These results indicate that in most ESPAD countries those who are abstainers from cigarettes and alcohol most probably also are abstainers from illicit drugs, tranquillisers/sedatives and inhalants.

The alcohol and drug situation 2003

; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ;

19 % 20 34 % 35 49 % 50 56 % 57

; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ;

; ; ; ; ; ; ; ; ; ;

% ; ; ; ; ;; ; ; ; ; ; ; ; ; ;

Data uncertain or not available Non-participating country

; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ;

; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ;

; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ;

; ; ; ; ; ; ; ; ; ; ; ; ; ;

; ; ; ; ; ; ; ; ; ; ; ; ; ;

Figure 58a. Proportion of all students who perceive inhalants “very easy” or “fairly easy” to obtain. 2003. Germany and Turkey: Limited geographical coverage.

Boys

Girls Ireland (77) Isle of Man (62) Slovenia (61) Germany (6 Bundesl.) (60) Finland (58) Austria (56) United Kingdom (53) Cyprus (52) Denmark (52) Faroe Islands (51) Sweden (51) Croatia (50) Belgium (48) Czech Republic (47) Greece (46) Iceland (45) Poland (44) Switzerland (44) Malta (39) Hungary (37) Norway (37) Slovak Republic (37) France (36) Netherlands (36) Estonia (35) Bulgaria (32) Russia (Moscow) (30) Lithuani (28) Greenland (27) Latvia (26) Ukraine (19) Turkey (6 cities) (17) Italy (16) Portugal (15) Romania (13)

83 66 57 60 56 52 51 53 52 47 51 47 48 54 47 45 44 48 36 37 38 44 36 38 36 30 30 28 31 24 22 18 17 16 15 % 100

80

60

40

20

0

71 57 65 61 59 60 55 51 52 54 51 53 47 42 45 44 44 41 42 38 35 31 36 33 35 34 30 29 22 28 16 16 15 13 12 0

20

40

60

80

100%

Figure 58b. Proportion of boys and girls who perceive inhalants “very easy” or “fairly easy” to obtain. 2003. Values within brackets refer to all students. Data sorted by all students. Germany and Turkey: Limited geographical coverage.

The alcohol and drug situation 2003

179

; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ;

15 % 16 20 % 21 45 % 46 55 % 56

; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ;

; ; ; ; ; ; ; ; ; ;

% ; ; ; ; ;; ; ; ; ; ; ; ; ; ;

Data uncertain or not available Non-participating country

; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ;

; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ;

; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ;

; ; ; ; ; ; ; ; ; ; ; ; ; ;

; ; ; ; ; ; ; ; ; ; ; ; ; ;

Figure 59a. Proportion of all students who perceive marijuana or hashish “very easy” or “fairly easy” to obtain. 2003. Germany and Turkey: Limited geographical coverage.

Boys

Girls

86

55 57 53 55 55 56 53 44 48 48 44 39 35 34 34 34 25 25 26 22 22 21 25 21 21 21 17 14 18 12 9 %100

80

60

40

20

80

Faroe Islands (83) Ireland (60) Czech Republic (58) United Kingdom (58) Isle of Man (55) Slovenia (55) Denmark (52) Switzerland (51) Belgium (49) Slovak Republic (49) France (47) Croatia (45) Italy (44) Netherlands (42) Germany (6 Bundesl.) (41) Poland (37) Bulgaria (36) Iceland (36) Austria (33) Portugal (29) Norway (26) Russia (Moscow) (24) Estonia (23) Sweden (23) Latvia (22) Greece (20) Greenland (20) Hungary (20) Lithuani (20) Malta (20) Finland (19) Cyprus (13) Ukraine (13) Romania (11) Turkey (6 cities) (7)

60 60 61

0

60 56 54 55 53 50 47 44 43 42 46 40 35 38 35 37 39 33 25 27 23 20 25 22 19 16 19 19 19 20 11 9 10 5 0

20

40

60

80

100%

Figure 59b. Proportion of all students who perceive marijuana or hashish “very easy” or “fairly easy” to obtain. 2003. Values within brackets refer to all students. Data sorted by all students. Germany and Turkey: Limited geographical coverage.

180

The alcohol and drug situation 2003

; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ;

5% 6 10 % 11 14 % 15 19 % 20

; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ;

; ; ; ; ; ; ; ; ; ;

% ; ; ; ; ;; ; ; ; ; ; ; ; ; ;

Data uncertain or not available Non-participating country

; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ;

; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ;

; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ;

; ; ; ; ; ; ; ; ; ; ; ; ; ;

; ; ; ; ; ; ; ; ; ; ; ; ; ;

Figure 60a. Proportion of all students who perceive LSD or other hallucinogen “very easy” or “fairly easy” to obtain. 2003. Germany and Turkey: Limited geographical coverage.

Boys

Girls Croatia (21) Poland (21) Slovenia (18) United Kingdom (18) Czech Republic (17) Denmark (16) Ireland (16) Isle of Man (15) Slovak Republic (15) Belgium (14) Bulgaria (14) Germany (6 Bundesl.) (14) Austria (13) Estonia (13) Iceland (13) Sweden (13) Norway (12) Portugal (12) Hungary (11) Italy (11) Latvia (11) Lithuani (11) Greece (10) Russia (Moscow) (10) Switzerland (10) Netherlands (9) France (8) Cyprus (6) Finland (6) Malta (6) Romania (5) Faroe Islands (4) Greenland (4) Turkey (6 cities) (4) Ukraine (4)

21 22 18 18 17 18 16 16 17 16 13 14 13 14 12 14 13 14 11 13 12 12 12 11 11 11 9 8 4 5 6 4 5 5 5 % 25

20

15

10

5

0

22 21 19 18 18 15 16 14 13 12 15 13 12 12 14 13 11 11 11 10 10 10 9 9 10 7 7 5 7 7 4 4 3 4 3 0

5

10

15

20

25 %

Figure 60b. Proportion of all students who perceive LSD or other hallucinogen “very easy” or “fairly easy” to obtain. 2003. Values within brackets refer to all students. Data sorted by all students. Germany and Turkey: Limited geographical coverage.

The alcohol and drug situation 2003

181

Attitudes towards drugs Perceived availability of substances (Tables 42a–c, figures 58a–b, 59a–b, 60a–b) The students were asked: “How difficult do you think it would be for you to get each of the following?” For each of the listed substances the response categories were: “Impossible”, “Very difficult”, “Fairly difficult”, “Fairly easy”, “Very easy” and “Don’t know”. The proportions of students who indicated “very easy” or “fairly easy” to this question are discussed in this section. There are considerable differences in the availability of alcohol compared to illegal drugs. However, there are also substantial differences within the group of illegal substances. Considering the averages, beer is perceived slightly more available than wine (87% and 82% respectively answering “very easy” or “fairly easy”), with spirits a little behind (72%)). In all countries except Cyprus, beer is estimated to be the easiest alcoholic beverage to obtain compared to wine and spirits, although the differences are very small in some of the countries. Spirits is, in comparison, estimated to be most difficult (i.e. least easy) to obtain in virtually all countries. However, in some countries there are hardly any differences in the perception of the availability between beer, wine and spirits. On average, the largest proportions of students who claim that it is “very” or fairly” easy to get beer, wine and spirits are found in Denmark (96% on average), Greece (93%), the Czech Republic (92%) and the Slovak Republic (91%). These beverages seem to be least easy to obtain in Greenland (30%) and Turkey (46%). The lowest single figures are found for sprits and wine in Greenland (30 and 42% respectively) and for the same beverages in Turkey (34 and 46% respectively). For other drugs the availability varies considerably across both countries and substances. Looking at the average figures, inhalants and cannabis are the two most mentioned substances (41 and 35% respectively). Inhalants seem to be easiest to get in Ireland (77%) followed by Slovenia (61%) and Germany (60%). Least easy to find are inhalants in Italy, Portugal, Romania and Turkey in which 13–17% gave this answer. In most countries anabolic steroids are perceived as less easy to get. The largest proportions answering “very” or “fairly” easy are found in Poland (27%), Bulgaria (24%) and Greece (20%). Smallest proportions were reported from the Faroe Islands (3%) followed by Finland, France, Greenland

182

and Ukraine (4–5%). The average of the ESPAD countries was 14% and the corresponding figure for USA 30%. Marijuana or hashish is somewhat easier to get than all other drugs but inhalants. The average percentage of students who reported that cannabis was “very” or “fairly” easy to obtain was 35%. The largest figure was found in the Faroe Islands (83%). Other countries with rather many students giving these answers are the Czech Republic, Ireland and the United Kingdom (58–60%). The smallest proportions were found in Turkey (7%) and Romania (10%). The corresponding figures in Spain and USA were 67 and 74% respectively, i.e. higher than in all but one of the ESPAD countries. The perceived availability for amphetamines is highest in Poland (27%), followed by Croatia and Denmark (22–23%) and Austria, Germany, Iceland and the United Kingdom (18–19%). The availability is judged to be much lower in some countries, including the Faroe Islands, Greenland, Turkey and Ukraine (4–5%) as well as Cyprus, Finland and Romania with 6–7%. In Spain this was reported by 43% and in USA by 36%, i.e. by more students than in any of the ESPAD countries. On average, LSD or other hallucinogens are thought to be “very” or “fairly” easy to obtain by 12% of the ESPAD students. These answers were given by 21% in Croatia and Poland. Next comes a group of countries with 17–18%, including the Czech Republic, Slovenia and the United Kingdom. Very few students (4–5%) thought so in the Faroe Islands, Greenland, Romania, Turkey and Ukraine. Again, the figures for Spain (43%) and USA (23%) were higher than in any of the ESPAD countries. Crack seems to be most available in Denmark, Ireland, Isle of Man, Poland, Slovenia and the United Kingdom in which 16–18% answered that this was “very” or “fairly” easy to obtain. Countries with the lowest figures (2–5%) include Cyprus, Finland, Greenland, Turkey and Ukraine. The corresponding figure in USA is much higher (30%). The figures about the perceived availability of cocaine are in most countries very similar to those of crack. The highest figures are found in Denmark, Ireland, Poland, Slovenia and the United Kingdom in which 18–22% answered “very” or “fairly” easy. The low prevalence countries include Finland, Greenland, Turkey and Ukraine with 2–5% giving this answer. However, the mean value is 12%,

The alcohol and drug situation 2003

which means that also for cocaine the corresponding figures are higher in Spain (40%) and USA (30%). There are big differences between countries in the perceived availability of ecstasy. Countries where the highest number of students answer that ecstasy is “very” or “fairly” easy to obtain include the Czech Republic, Ireland and Slovenia with 32–34%. Much lower figures (3–5%) are found in Greenland, Turkey and Ukraine. The average for all ESPAD countries is 17%, but the corresponding figures are much higher in Spain (48%) and USA (36%). For heroin the largest percentages of students who think that this substance is easy to find are reported from Poland (20%) together with Croatia, Denmark, Ireland and Slovenia (15–17%). Very few students (2–5%) thought so in Finland, Greenland, Turkey and Ukraine. In Spain 31% gave this answer and in USA 19%. On average 13% of the ESPAD students answered that magic mushrooms were “very” or “fairly” easy to obtain. It was most easily available in the Czech Republic and Isle of Man (28% each) followed by Ireland, Poland and the United Kingdom (22–24%). Much lower figures (3–4%) were found in Cyprus, Greenland, Turkey and Ukraine. GHB has the lowest ESPAD average of all drugs (7%). However, there are big variations between countries and the figure was twice this high (14– 15%) in Denmark and Poland. The lowest prevalences (2–4%) were reported from the Faroe Islands, Finland, Greenland and Ukraine. Tranquillisers and sedatives is the “third easiest” drug to obtain with an ESPAD average of 21%. The country with the highest figure is Cyprus (42%) followed by Greece and Poland (39–40%). Only 4% of the students in Ukraine answered that tranquillisers and sedatives were “very” or “fairly” easy to obtain. The figures were also low in Greenland, Russia (Moscow) and Turkey (with 9–10%). The corresponding figure in Spain (66%) is much higher than in any ESPAD country. To sum up, alcohol is considered to be “very” or” fairly” easy to obtain by a large majority of the students in most countries. Inhalants is the most available substance among other drugs. On average this was mentioned by 41% of the ESPAD students. Marijuana or hashish come next (35%) followed by tranquillisers or sedatives (27%). Among the remaining drugs listed, ecstasy is on average perceived most easy to obtain (17%). For all other drugs the corresponding figures vary between 7 and 13%.

The alcohol and drug situation 2003

The perceived availability of illegal drugs differs between countries. Among the ESPAD countries it seems to be highest in the Czech Republic, Denmark, Ireland, Poland, Slovenia and the United Kingdom. However, with a few exceptions, all illegal drugs (for which comparable figures are available) the perceived availability is higher in Spain and USA than in any of the ESPAD countries. Also the lowest perceived availability of illegal drugs is mainly concentrated to a limited number of countries. These include Greenland, Romania and Ukraine. Looking at the ESPAD average figures there are very few gender differences in the perceived availability of illegal drugs. There are more boys than girls answering that anabolic steroids, cannabis and magic mushrooms are “fairly” or “very” easy to obtain, while the opposite is true for tranquillisers and sedatives.

Perceived risks of substance use (Tables 43a–c) The students were asked: “How much do you think people risk harming themselves (physically or in other ways) if they a) smoke cigarettes occasionally, b) smoke one or more packs of cigarettes per day, etc”. Eighteen items regarding cigarette smoking, alcohol consumption and illicit drug use suggesting different intensity of use were listed. The response categories were “no risk”, “slight risk”, “moderate risk”, “great risk” and “don’t know”. The comments in this section is concentrated to answers indicating “great risk” for each of the items. Many of the drugs included in the question is not known by students in Greenland, which makes the Greenlandic data less comparable with data from other countries. To stress this Greenland is put below the line in tables 43a–c. The average values of risk assessment vary substantially between different substances. The highest average value is denoted for regular injections of drugs, which 81% of the ESPAD students would associate with a great risk. A little lesser students thought that regular use of cocaine/crack (76%) or regular use of ecstasy (73%) would put people at risk. The behaviours that rather few students indicate as risky are use of marijuana or hashish once or twice (32%), use of inhalants once or twice (35%) or use of amphetamine or GHB once or twice (37% each). A majority of the students (69%) think that smoking a pack of cigarettes or more per day would mean

183

a health risk. The individual countries where the highest percentage of students indicated this include the Faroe Islands (86%), Denmark, Romania (77% each), France, Isle of Man, Switzerland (76% each). Countries where least students considered this as a great risk are Ukraine (47%), Russia (Moscow) (51%), Slovenia (56%), Croatia and Portugal (59% each). In USA 72% thought that smoking one or more packs of cigarettes per day would mean a great risk. Five or more drinks each weekend is not considered to be a great risk, and on average only 37% thought so. About half of the students in the countries that scored highest had indicated this to be associated with great risk. They include Turkey (52%), France (51%) and Poland (49%). However, in five countries only one fifth of the students or less would consider 5 or more drinks each weekend to be a risky behaviour. These countries are Ireland (15%), the Netherlands, Norway (19% each), the United Kingdom (21%), the Isle of Man (22%) and Belgium (23%). A majority of these countries are among the top countries as regards frequent alcohol use among students. In USA 53% indicated this to be a risky behaviour, thus being on the same level as the European countries scoring high on this variable. Taking marijuana or hashish once or twice is on average not seen as a very risky behaviour. Only one third of the student think so, which is the lowest rate compared to all other variables included in this question. There are variations, but in only two countries (Lithuania with 58 and Romania with 51%) more than half of the students answered this. Other high prevalence rates are observed in Greece and Poland (48% each). In nine countries 15% or less considered use of cannabis once or twice as a risky behaviour. The lowest figure is found in Isle of Man (11%), followed by the Netherlands (12%), the Czech Republic, Switzerland, the United Kingdom (13% each), Belgium, Germany (14% each), Denmark and Ireland (15% each). The figure for USA (22%) is also rather low. Regular use of cannabis is viewed upon quite differently compared to use on single occasions. On average 70% of the students thought such use would implicate great risk. In the Faroe Islands and Greece 87% of the students thought this would be risky. Other countries where rather many students thought so include Iceland, Sweden (83% each), Finland, Latvia and Poland (81%). On the other hand, less than half of the students in Isle of Man (44%), the United Kingdom (46%) and the Nether-

184

lands (47%) gave this answer. It is obviously so, that the students in low prevalence areas like the Nordic countries tend to have a stricter view on this than those in the high prevalence parts of Europe, such as the British Isles. In USA the corresponding figure is 66%. In somewhat more than one third of the ESPAD countries occasional use, such as once or twice, of LSD was indicated as risky. The highest values are found in Iceland (70%), Lithuania (57%) and Poland (54%). Much less strict attitudes seem to be prevalent in the Netherlands (25%), the Czech Republic (26%), the Slovak Republic (27%) and Denmark (29%). Of the American students 54% thought that using LSD occasionally was associated with great risk. Regular use of LSD is overall considered as a greater risk than occasional use, but the average is not higher than for regular use of cannabis (69%). The countries where most students thought that regular use of LSD would be risky include Finland, Iceland (86% each), Poland (81%), the Czech Republic and Lithuania (78%). The lowest number of students who agreed with this statement is found in Turkey with 44%. Other countries with somewhat low percentages are the Netherlands (55%) and Romania (58%). The corresponding figure for USA is 83%. About one third of the ESPAD students thought that using amphetamines once or twice would be risky. In countries with the highest rates giving the answer “great risk” only somewhat more than half of the students thought so. In Iceland 60% gave this answer, in Lithuania 56% and in Poland 55%. About one-fifth in the Slovak Republic (22%) and Switzerland (23%) answered this and about one-fourth in Germany (25%), Austria and the Netherlands (26% each). In some countries rather many students thought that using amphetamines regularly would mean a great risk. These countries are mainly found in the north or by the Baltic sea and include Finland (87%), the Czech Republic, Poland (85% each), Iceland (84%) and the Faroe Islands (80%). A much lower figures is found in Turkey (45)%. Other countries with rather low values all represent more than half of the students, e.g. Greece, the Netherlands, Romania, Switzerland and Ukraine (53–58%). Many students seem to consider occasional use of cocaine or crack as a minor danger of personal harm. The highest percentages of students who think that using these substances once or twice would mean a great risk are found in Iceland (63%),

The alcohol and drug situation 2003

Lithuania (60%) and Poland (58%). Other countries where more than half of the students gave this answer include Croatia and Russia (Moscow) (51– 54%). In some countries about one third of the students thought that occasional use of cocaine/ crack would be risky. They include the Netherlands (30%), Denmark (31%), Norway (34%) and Belgium (35%). In USA this item regarded cocaine powder only, but 55% of the students thought that occasional use would implicate a great risk. Regular use of cocaine or crack was considered to be a great risk by about 85% in the Czech Republic, the Faroe Islands, Iceland and Poland (85–87%). The smallest figure in this respect was reported from Turkey (52%). Occasional use (once or twice) of ecstasy was considered as a great risk by 42% on average. The highest numbers of students who indicate this are found in Iceland (68%) and Ireland (63%). In only five other countries more than half of the students answered this. They include the Faroe Islands, Isle of Man, Lithuania, Poland and the United Kingdom (52–54%). The lowest figures are found in the Czech Republic, the Slovak Republic (23% each) and the Netherlands (28%). In comparison, 55% of the American students indicated this. Regular use of ecstasy is viewed upon in a different way than occasional use. On average 73% of the ESPAD students regard such use as a great risk. The highest numbers indicating this are found in Iceland (86%), the Faroe Islands (85%), Finland, France, Ireland, Malta, Poland (82% each) and Denmark (81%). Rather few students in Turkey (44%) and Ukraine (58%) thought this to be a great risk. Occasional use of inhalants (once or twice) was on average considered as a great risk factor by 35% of the students. The highest numbers in individual countries were found in Iceland, Lithuania and Poland, where 55% in each country indicated this. In three countries, however, only one fifth of the students gave this answer, including Germany (19%), Austria, the Netherlands (21% each) and the Slovak Republic (23%). The corresponding figure in USA is 50%. Regular use of inhalants was indicated as a risky behaviour by 68% on average. The highest rates were reported from the Czech Republic (85%), Iceland and Poland (82% each). Countries where only slightly more than half of the students thought that regular use of inhalants would be risky include Turkey (52%) and Malta (55% each). In USA 76% of the students thought that such use would impli-

The alcohol and drug situation 2003

cate a great risk. The use of GHB is not spread in all countries and it was therefore not included in all questionnaires. On average the occasional use (once or twice) was considered a great risk by 37% of the students. The highest values are found in Iceland (66%), Lithuania (55%) and Poland (54%). Small percentages are reported from the Slovak Republic (23%), Belgium, the Netherlands (25% each), Germany and Switzerland (26% each). Regular use of GHB was on average judged as a risky behaviour by 62% of the ESPAD students. The highest figure is found in Iceland with 82%. Around three quarters of the students in Denmark, Lithuania, Poland and Sweden reported this (72– 78%). Lower number of students gave this answer in Turkey (43%), Belgium (47%) and Ireland (49%). Use of drugs by injection is rare in most countries in this age group. Thus, it might be expected that most students would associate such use with great risks. The average percentage of students viewing occasional use as a great risk is 62%. The highest proportion in the individual countries was found in Iceland (80%). Somewhat lower levels were found in Ireland, Lithuania (73% each), France, Poland (72% each) and Latvia (71%). The lowest figures are reported from Turkey (42%), the Netherlands (44%) and Sweden (45%). Regular injections of drugs are on average thought to be a great risk by 81% of the ESPAD students. The highest values are reported from France (92%), the Czech Republic (91%) and Iceland (90%). A much lower figure is reported from Turkey (51%). Overall more girls than boys perceive the different behaviours to be associated with great risks. However very small differences can be seen in relation to the occasional (once or twice) use, for all the included substances. It might also be of interest to notice that the lowest risk perceptions to a large extent are found in a limited number of countries. This is mainly the case in the Netherlands, in which rather few students associated the different behaviours with risks. In other countries, however, it was the other way around. High percentages of students in Iceland and Poland considered the listed behaviours to implicate great health risks.

Perceived risks of heavy drinking (Tables 44a–c) The role of alcohol and the way that alcohol consumption is perceived differs between countries. However, all societies are concerned about drunk-

185

enness and problems that follow out of this. There is also a main general concern about risks related to alcohol consumption and especially problems related to heavy drinking. The awareness of possible consequences of heavy drinking differ between countries, probably both among adults and young people. To learn more about the perception of heavy drinking among young people in different countries the students were asked the following question: “Do you think that heavy drinking influences the following problems?”. The problems listed were “traffic accidents”, “other accidents”, “violent crime”, “family problems”, “health problems”, “relationship problems” and “financial problems”. Tables 44a–c show the percentages that have answered “Yes, considerably” and “Yes, quite a lot”. The problem that most ESPAD students relate to heavy drinking is traffic accidents, which on average was indicated by 85% of the students. Next to that come other accidents and health problems (74% each), closely followed by violent crime (70%). The corresponding figures are a little lower for family problems (69%), financial problems (66%) and relationship problem (63%). In nearly half of the countries 90% of the students or more have related heavy drinking to traffic accidents. The highest figures are found in a group of countries with 93–96%, including Austria, Croatia, France, Greece, Italy, Poland, Russia (Moscow) and Turkey. Figures below 80% are found in Ukraine (74%), Hungary (77%) and in Norway (79%). In Italy, Poland and Turkey a majority of the students (84–86%) thought that heavy drinking is related to other accidents. This was also indicated by 81–82% of the students in Austria, Croatia, Romania and Russia (Moscow). Percentages below 65% are found in Belgium, Hungary (56% each), Ukraine (62%) and the Netherlands (64%). The figures related to heavy drinking are similar to those of other accidents. Six countries report figures above 80% of which the highest are found in Russia (Moscow) and Turkey (87–89%). Other countries with high figures (82–83%) include Croatia, Greece, Italy and Romania. Four countries have reported figures below 65%, including Belgium and Hungary (58–59%) as well as Iceland, the Netherlands and Norway (60–63%). In Turkey 87% of the students relate heavy drinking to violent crime. Next to that come Croatia and Iceland (82–83%) followed by the Faroe Islands and Poland (80%). Belgium is the country with the lowest figure (47%) followed by Estonia and France (58–

186

60%). Other countries with low figures (61–62%) include Germany, Italy, Latvia and Ukraine. The country in which the vast majority of the students relate family problems to heavy drinking is Turkey where 88% gave this answer. In a group of four countries the corresponding figure was 79– 81% (Croatia, Poland, Romania and the Slovak Republic). Low figures are mainly reported from the Netherlands (50%) and Belgium (54%) but also from a group of countries including France, Germany, Isle of Man, Norway, Sweden, Ukraine and the United Kingdom, in which 60–62% of the students indicated this. A supposed influence of heavy drinking on financial problems is mainly reported from Turkey in which 84% of the students answered this. Next come two countries with 79% (the Czech Republic and the Slovak Republic) followed by Austria and Poland with 75–76%. At the other end of the scale is Denmark (41%) and France (44%). A little bit behind follow Belgium with 53% and five countries with 56–59% (Bulgaria, Greece, Italy, the Netherlands and Ukraine). Relationship problems is the category that the students consider being least related to heavy drinking. The ESPAD average is 63% but, like for all other variables, there is a considerable difference between the countries with the highest and lowest figures (83 and 49% respectively). Turkey is the country with the highest figure (83%). Second to this, but with substantially lower figures, follows a group of countries (Austria, the Czech Republic, Malta, Romania and the Slovak Republic), in which 70–74% had indicated that heavy drinking is related to problems with relations. The lowest figure (49%) is reported from three countries (Belgium, Lithuania and the Netherlands) closely followed by the Faroe Islands, Ukraine (51% each) and Norway (54%). Some countries are repeatedly appearing in the comments above, either as a country in which many students relate most of the problem categories to heavy drinking, or the other way around, i.e. rather few students agree. Countries in which many students relate heavy drinking to many of the problems mentioned include Croatia, Poland and Turkey. In another group relatively few students relate heavy drinking to the different problems. Examples of countries in this group include Belgium, the Netherlands and Ukraine. For all categories of problems but family problems, there are more girls than boys who think that they either “considerably” or “quite a lot” are re-

The alcohol and drug situation 2003

lated to heavy drinking. This is especially true for violent crime, which on average was answered by 73% of the girls and 67% of the boys. The outcome on this variable show that the students’ opinions vary over the countries. It is reason-

able to think that this might reflect more aspects of the drinking cultures than just personal attitudes. Important variables that would need separate analyses are drinking cultures, traffic legislation etc.

Purchase of alcoholic beverages (Tables 45a–c) The legal drinking age differ between the ESPAD countries. In some countries you need to be 18 years old to drink alcohol in a restaurant or a pub and 20 to buy wine or spirits in a store. In other countries there are lower age limits while some countries do not have restrictions at all. Another difference in the availability of alcoholic beverages is that some countries have state owned monopoly stores or other specific outlets, while beer, wine and spirits in other countries are available in grocery stores as any other provisions. Hence, at least from a legal point of view, alcoholic beverages should be differently available for 16 year old persons in the ESPAD countries. The students were asked the following question: “Think back over the LAST 30 DAYS. How many times (if any) have you bought beer, wine or spirits in a store (grocery store, liquor store, kiosk or gas station) for your own consumption?”. The answers to such a question mirror two things. One is of course the availability of beer, wine and spirits and the other is how common it is to drink each of the three different beverages. The more common it is to drink a beverage the more common it might be that it is bought in a store. Beer is the beverage that most students have bought for their own consumption during the last 30 days. On average this was answered by 25% of the ESPAD students. A little less than one fifth said that they had bought spirits (19%) while wine was the beverage less commonly bought for own consumption (11%). There are large differences in the number of students that during the last 30 days had bought beer for their own consumption in a store. This was answered by slightly more than half of the students in Poland (53%). Next came a group of four countries in which 46–47% of the students gave this answer (Bulgaria, Denmark, Russia (Moscow) and Ukraine). The lowest figure is reported from a group of countries in which 10–12% had done so

The alcohol and drug situation 2003

(Greenland, Isle of Man, Portugal and Sweden). Other countries with low figures (14–16%) include Hungary, Norway and Turkey. The ranking of countries appear to be about the same when it comes to a purchase frequency of 3 times or more often. This was most common in Poland and Russia (Moscow) (28% each) followed by Bulgaria and Denmark (23–24%). Three countries report that this was done by only 4% of the students (Hungary, Portugal and Sweden) while another two reported 6% (France and Isle of Man). The second most popular beverage to buy (and to drink) is spirits. It is first and foremost in Denmark that the students have bought spirits for their own consumption during the last 30 days. This was the case with as much as 45% of the students. Next in prevalence rate come the Faroe Islands and Malta (29–31%) followed by Belgium, Bulgaria and the United Kingdom (25% each). The lowest number of students that have done so are found in Sweden and Turkey (5–6%), followed by Finland (7%), Romania (9%) and Croatia, Iceland and Norway (11–13%). The high and low prevalence countries are about the same when it comes to the purchase of spirits for own use at least 3 times during the last 30 days. This had been done by 16% of the students in Denmark and Malta and by 12–13% in Estonia, the Faroe Islands and the United Kingdom. This was least common in Finland, Romania, Sweden and Turkey (2% each). The least commonly bought alcoholic beverage for own consumption is wine and the variations between countries is also smaller (2–26%). It is the Maltese students that have bought wine to the largest extent (26%). Next to them come the students in Russia (Moscow) (21%) followed by five countries in which 18% gave this answer (Austria, Estonia, Hungary, Slovenia and Ukraine). The countries where it is least common to buy wine include France, Sweden (2% each), Iceland, Portugal (3% each), the Faroe Islands, Finland, Greenland, the

187

Netherlands and Norway (4% each). It is unusual that students have bought wine in a store more often than twice during the last 30 days. The range goes from 0% in the Faroe Islands and the Netherlands to 10% in Malta. Overall it is more common among boys than among girls to have bought alcoholic beverages in

a store during the last 30 days. This is true for beer as well as for wine and spirits, even if it is most pronounced in relation to beer. Whenever there is a discrepancy between boys and girls in a single country it usually follows this general trend. However, there are some very few exceptions, mainly for spirits.

Perceived cigarette, alcohol and drug use among friends (Tables 46, 47a–b) It would be reasonable to think that in countries with high prevalence rates on e.g. smoking, there should also be high percentages reporting that most or all friends are doing the same. The students were asked: “How many of your friends would you estimate smoke cigarettes?” as well as similar questions for alcohol consumption and the use of different illicit substances. The response categories were: “None”, “A few”, “Some”, “Most” and “All”. In the next paragraphs about cigarette smoking and alcohol use, the proportions who answered “most” or “all” friends will be presented. Looking at the ESPAD averages the most common is that the students have friends that drink alcohol (60%) or are smoking cigarettes (47%). There are much fewer who have friends that get drunk at least once a week (17%). Countries with high percentages reporting that most or all friends smoke cigarettes include Cyprus and Finland (88–89%) followed by Bulgaria and Russia (Moscow) (67–71%). The lowest figures are found in Iceland (17%) and Sweden (20%) but also in Denmark, Ireland, Norway, Portugal and Turkey with 27–29%. Overall, there are more girls than boys reporting that their friends smoke. This holds true in more than two thirds of the participating countries. Although drinking alcoholic beverages is a widespread behaviour in most of the ESPAD countries it is only in a little more than half of the countries that 50% or more report that most or all of their friends drink alcohol. The largest figure is to be found in Denmark where 89% of the students reported this. Next follow Ireland and Isle of Man with 80–81% but also Austria and Germany (75– 77%). The distance to Turkey is huge with only 19% in that country giving this answer. Second lowest is Hungary (26%) followed by Portugal (38%).

188

In a majority of the countries the gender differences in this respect are very small or non-existent. Only in nine countries notably higher proportions of girls than of boys answered that most or all of their friends were drinking alcohol (Bulgaria, Estonia, the Faroe Islands, Finland, Iceland, Ireland, Isle of Man, Latvia and Norway), while the opposite was true in only two (Romania and the Slovak Republic). Overall, there are rather few students that reported that most or all of their friends get drunk once a week or more often. There are, however, a few countries where one fourth or more of the students reported this. These include Isle of Man (39%), Denmark, Ireland and the United Kingdom (32–36%) as well as Bulgaria, Croatia and Estonia (27% each). This answer has only been given by 5–6% of the students in Cyprus, Greece, Portugal, Turkey and by 8–9% in Hungary, Iceland and Poland. In a very large majority of the countries there are no substantial differences between the sexes in relation to possible drunkenness among friends. This section also includes information about how common it is that students think that “some”, “most” or “all” of their friends are using cannabis, LSD or other hallucinogens, amphetamines, tranquillisers or sedatives, cocaine or crack, ecstasy, heroin, inhalants, alcohol together with pills and anabolic steroids. As expected, the highest proportion giving this answer is found for marijuana or hashish with an ESPAD average of 21%. However, the range is wide and goes from 3 to 46%. The highest figures (44–46%) are found in Isle of Man, Italy and the United Kingdom closely followed by Belgium, the Czech Republic and Switzerland (42–43%). The smallest figure is found in Romania (3%) followed by the Faroe Islands, Greece, Hungary, Malta, Sweden and Turkey (5–6%). Even though there are huge differences between

The alcohol and drug situation 2003

countries there are practically no gender differences within countries. With the exception of cannabis there are few countries in which as many as one tenth of the students report that their friends use any of the suggested drugs. One exception is inhalants for which two countries report rather high figures. One is Cyprus with 15% and the other Greenland with 12%. In three countries at least 10% of the students report that friends are using alcohol together with pills. In Isle of Man 14% of the students answered that at least some of their friends do this. The second country was the United Kingdom with 12% and the third Croatia with 10%. When it comes to ecstasy 10% of the students have answered this in Croatia, the Czech Republic and Isle of Man. The same figure is reported from two countries about the use of magic mushrooms. These countries are the Czech Republic and Isle of Man.

For all other drugs the figures are smaller. They range from 1–7% for LSD or other hallucinogens, from 1–8% for amphetamines, from 1–8% for tranquillisers and sedatives without a doctors’ prescription, from 1–8% for cocaine or crack, from 1–5% for heroin, from 1–4% for GHB and from 1–6% for anabolic steroids. If the countries with the highest figures on each drug are counted some kind of a pattern is showing. Three countries belong to the “top countries” on five out of the twelve drugs on the list. They include Croatia, Isle of Man and Italy. In general, there are hardly any gender differences in the student’ perceived drug use among friends. However, in some countries there are more girls than boys estimating that their friends take alcohol together with pills. On the other hand, in some countries there are more boys than girls reporting that they have friends that use anabolic steroids.

Cigarette, alcohol and drug consumption among elder siblings (Tables 48a–c) Students who have any elder sibling were asked whether the sibling(s) ever smoke cigarettes, drink alcohol, get drunk, smoke marijuana or hashish, take tranquillisers or sedatives or take ecstasy. This information is perhaps most interesting in relation to the students’ own behaviour and will be discussed from this perspective in the next chapter. However, it might also be of interest to see the findings as they are, and the number of students who indicated any of the listed behaviours are presented below. The most common behaviour among the elder siblings is that they drink alcohol. On average this was answered by 62% of the ESPAD students. To have elder siblings who smoke cigarettes and who get drunk was equally common. Both alternatives were answered by 42%. Elder siblings smoking marijuana or hashish was on the average mentioned by 10% of the students while only 3% said that the elder siblings either took ecstasy or tranquillisers or sedatives without a doctors’ prescription. In about one third of the countries 50% or more of the elder siblings smoke cigarettes. This was reported to the highest extent in Greenland (68%) and the Faroe Islands (60%) followed by Austria,

The alcohol and drug situation 2003

Belgium, the Czech Republic, Germany and Norway (52–53%). The lowest figures (26–28%) are found in Cyprus, Isle of Man, Italy, Malta, Romania and the Slovak Republic. There are much more students reporting alcohol consumption among elder siblings. The top country is Ireland in which nearly nine out of ten elder siblings drink alcohol (89%). High figures are also reported from Denmark, Iceland and Norway (84– 85%) as well as from the Czech Republic, France and the United Kingdom (80–81%). The lowest figures are found in Turkey (18%) and Romania (24%), but to some extent also in Italy (31%), Cyprus (34%) and the Slovak Republic (37%). It is rather obvious that there are fewer students reporting that elder siblings get drunk than that they drink alcohol. However, high figures are also found for this variable with 76–79% in Denmark, Ireland and Norway and 72–74% in Greenland, Iceland and the United Kingdom. The discrepancy to the countries with the lowest figures is remarkable with 9% in Cyprus and Romania and 12% in Greece and Turkey. Other countries with low figures include Hungary and the Slovak Republic(15% each). Even if the ESPAD average for elder siblings

189

who smoke cannabis is 10%, figures that are more than twice as high are found in some countries. Three countries report that this is the case for 24– 25% of the students (Belgium, Ireland and the United Kingdom) while the corresponding figure was 22% in the Czech Republic and Switzerland. The lowest number of students that gave this answer (2–3%) are found in Cyprus, Finland, Greece, Lithuania, Romania and Sweden. As mentioned above, there are much fewer students that have answered that their elder siblings

190

take ecstasy or that they take tranquillisers and sedatives without a doctors’ prescription. For the latter drug the range goes from 1 to 5% and for ecstasy from 1 to 7%. On average there are more girls than boys who have elder siblings that smoke cigarettes, drink alcohol and get drunk. These kinds of differences in relation to alcohol are found in a majority of countries, in about half of the countries for cigarette smoking and in about one third of the countries for drunkenness.

The alcohol and drug situation 2003

Correlates of adolescent substance use

Introduction The literature on adolescent substance use has identified a wide range of attitudinal, behavioural and structural factors that have a significant correlation with some types of substance use in some countries at some point in time. However, such statistical associations are far from deterministic, and there is hardly any such correlate of adolescent substance use that has not been found to be non-significant in some study. Furthermore, certain factors appear to have a positive association with substance use in some studies, but a negative association with such use in other studies. Given the methodological dif ferences between studies, it is in most cases difficult to determine if such inconsistent patterns in the correlates of substance use reflect substantive or methodological differences. The ESPAD study provides a unique opportunity to examine the patterns of association between substance use and various other factors. The ESPAD data is collected according to a single research protocol and employs strictly comparable variables for cross-cultural comparisons. As in earlier studies, each country performs the statistical analysis needed for cross-national comparisons and files a standard country report with its results. This procedure limits the scope for this analysis somewhat, but the current comparison of raw correlations nev-

ertheless offers a first glimpse of what could be achieved with a common database in future waves of the ESPAD project. The following analysis examines the correlation between adolescent use of cigarettes, alcohol and cannabis on one hand, and various background factors on the other. The ESPAD study offers a wide variety of indicators of each type of substance use and patterns of correlation differ somewhat for different indicators. In the interest of simplicity, the correlations reported below are all based on frequency of lifetime use. As discussed in the methodological chapter, the confidence intervals of ESPAD data do not take into account the clustered nature of the samples. Tests of statistical significance based on the assumption of simple random sampling will therefore provide a higher level of precision than would be obtained under the assumption of cluster sampling. In other words, associations that are found to be statistically significant with standard t-tests might not be significant if intracluster correlations were taken into account. In this section tests of statistical significance based on the assumption of simple random sampling are provided for general guidance, but they should be interpreted with considerable caution.

Parental education Research has shown that educational attainment is associated with decreased smoking in particular, and somewhat less consistently with decreased alcohol consumption (Bjarnason, 2000). To the extent that educated parents are more knowledgeable about the dangers of adolescent substance use and communicate such information more effectively to their children, the educational attainment of parents should also be associated with less adolescent substance use. Interestingly, however, a number of earlier studies in various European countries (Glend-

Correlates of adolescent substance use

inning, Shucksmith and Hendry, 1997; Morgan and Grube, 1989; Parker and Measham, 1994; Pedersen, 1990; Thorlindsson and Vilhjalmsson, 1991; Tuinstra et al., 1998) have failed to find any effects of parental education on adolescent substance use. Contrary to these findings, the results of the 2003 ESPAD study suggest that there is some association between parental education and adolescent substance use, but this association is far from simple. As figures 61 and 62 show, the correlation between parental education and the use of ciga-

191

Cannabis Isle of Man United Kingdom Sweden Norway Ireland Faroe Islands Czech Republic

Alcohol

-0,13 -0,1 -0,12 -0,11

Faroe Islands

-0,03

-0,09 -0,12

-0,04 -0,07 -0,03 -0,03

Malta Ukraine Greece

-0,06

Czech Republic Sweden 0,03

-0,01

Greenland

Lithuania 0,04

-0,03

Austria Estonia

0,05 0,06

0,01 0 0,03 0,02 0,02 0,05 0,07

Belgium 0,11

Hungary

0,13

Netherlands

0,13

0,18 0,15

-0,04

Slovenia

0,01

Croatia

0,02

0,01

-0,05 -0,08 -0,02 -0,03 -0,03 -0,02 -0,06

0,02

-0,07

0,08

-0,06 0,03 0,02

-0,05 -0,01 -0,01

0,02 0,05

-0,01 -0,04

0,01 0,03

-0,04

0,05

-0,01 -0,03

0,12

-0,04 -0,05

0,01

0,1

0,02

0,01

0,1

0,2

0,01

Figure 61. Use of cigarettes, alcohol and cannabis by father’s education. Filled bars: Significant correlations. Bars marked with lines: Non-significant correlations.

0,13

-0,04 0,04

0,12

-0,05 0,06 0,04 0,04 0,04 0,08 0,04 0,03

Cyprus

0,15

-0,01 0,09 0,09

0,01

0,06

0,13

0,01 0,02

0,12 0,1 0,11

Bulgaria -0,2

0,12

-0,06

Croatia 0,19

0,05

0,05 0,03

-0,01

Romania 0,13

0,06

0,02 0,04

-0,02 -0,01

Slovenia

0,09 0,07

-0,01

0

0,02 -0,04

Turkey (6 cities)

0,07 0,09

0,07

Bulgaria

192

-0,06 -0,09 -0,04 -0,04 -0,06

Switzerland

0 0,01 0,04

-0,1

-0,01

Greece

-0,06

Romania

0,03

-0,13

Ukraine

-0,03 -0,01

-0,03 -0,07 -0,08 -0,06

Italy

0,01

Austria

0,01

-0,13

0,01

-0,03

Hungary

-0,08 -0,03 -0,1 -0,06 -0,06 -0,08 -0,07

Malta

0,05 0,02 0,03 0,02

Belgium

Cigarettes

0,07

0

Greenland

-0,2

Denmark

0,01 0,04

-0,01

Cyprus

Germany (6 Bundesl.)

0,02 0,04

-0,02 -0,03

Slovak Republic

0,06

-0,04

Switzerland

Netherlands

Russia (Moscow) 0,02

0,03 0,01

Russia (Moscow)

Turkey (6 cities)

Ireland

0

-0,02

Slovak Republic

Finland

0,02

-0,04 -0,02 -0,01

Germany (6 Bundesl.)

Norway

0,01

-0,08

Italy

Alcohol

Iceland

-0,1

Finland

Estonia

United Kingdom

-0,02 -0,06 -0,08 -0,03 -0,06 -0,05 -0,02 -0,02 -0,1 -0,04

-0,05 -0,06

Denmark

Isle of Man

-0,03

-0,13

Iceland

Lithuania

Cannabis

Cigarettes

0,04

-0,1

0

0,1

0,18

0,2

Figure 62. Use of cigarettes, alcohol and cannabis by mother’s education. Filled bars: Significant correlations. Bars marked with lines: Non-significant correlations.

Correlates of adolescent substance use

rettes, alcohol, and cannabis ranges from being significantly negative to being significantly positive across the 31 countries providing data on this association. Furthermore, none of the countries report a significant association between the education of both parents and all three substances. The education of the father is only significantly associated with all three types of substance use in the Isle of Man and the education of the mother only has such an association with all three types of substance use in Cyprus, the Netherlands and Switzerland. Certain patterns do nevertheless emerge in these graphs. Cigarette use has a statistically significant negative correlation with parental education in 19 of the 31 reporting countries. A significant negative association was found with both mother’s education and father’s education in ten countries. Such a correlation was found with father’s education only in five countries and with mother’s education only in four countries. In contrast a significant positive correlation between cigarette smoking and father’s education was only found in one country and between smoking and mother’s education in two countries. The preponderance of the evidence thus points to a general, yet far from universal, pattern of parental education being associated with less smoking among European youths. The evidence regarding parental education and adolescent alcohol use is much less clear. On the one hand, a significant negative correlation was found in eight of the 31 countries. In four of these countries the effect was found for both parents, in three for father’s education only and in one for mother’s education only. On the other hand, alcohol use has a significant positive correlation with parental education in 14 countries. This effect was found for both parents in nine countries, for father’s education only in two countries and mother’s education only in three countries. Six of the eight countries with a significant negative association were located in the northern part of Europe. In contrast seven of the 14 countries where a signifi-

cant positive association was found are located in the eastern part of Europe, and the remaining seven are divided between Mediterranean countries and countries in the western part of Europe. The reason why parental education should operate in different ways in different countries and regions of Europe is unclear and warrants further research. In the case of cannabis use a negative association with parental education was found in five countries. In two of these countries the negative association was found for the education of both parents, in two countries with father’s education only and in one country with mother’s education only. In contrast, a positive correlation was found between parental education and cannabis use in 12 countries. This correlation was found for the education of both parents in seven countries, with father’s education only in one country and with mother’s education only in four countries. No clear geographical patterns emerged in this context. The five countries with negative correlations include three countries in the northern part of Europe and two countries in the eastern part. The 12 countries with positive correlations include three Mediterranean countries, three in the western part of Europe, and five countries in the eastern part. Again, the reason for this inconsistent pattern of correlations between parental education and cannabis use calls for further research. Overall, parental education has positive or nonsignificant associations only with different types of substance use in eight countries and negative or non-significant associations only in 10 countries. In 10 countries a mix of positive and negative associations was found, and in three no effects whatsoever were found for parental education. Tobacco use has the most consistent negative association with parental education. In the case of alcohol use and cannabis use there seems to be a certain tendency for negative effects to emerge in countries in the northern part of Europe and positive effects in countries in the eastern part.

Family structure A large body of research in Europe and NorthAmerica has found that adolescents who reside with both biological parents are less likely to smoke cigarettes, drink alcohol, or use cannabis (see Bjarnason, 2000). While this research generally finds

Correlates of adolescent substance use

all types of substance use to be more prevalent among adolescents who live with a single parent, the evidence regarding the effect of living with one biological parent and a stepparent is somewhat less conclusive.

193

Research among adolescents in several European countries has found substance use among adolescents that live with one parent and a stepparent to be similar to such use among adolescents that live with a single parent (Adalbjarnardottir and Blondal, 1996; Bjarnason, Anderson, et al, 2003; Bjarnason, Davidaviciene, et al, 2003; Glendinning, Shucksmith and Hendry, 1997; Irgens-Jensen, 1991). However, some studies in North-America have found that the presence of a stepparent may counteract the effect of not living with both biological parents (Adlaf and Ivis, 1996; Amey and Albrecht, 1998; Thomas, Farrell and Barnes, 1996). It is not clear to what extent these differences in research findings reflect underlying causal or structural differences in the position of single or divorced parents in Europe and North-America. The association between family structure and adolescent substance use in 29 reporting countries is shown in figures 63 and 64. These results shown are standardised regression coefficients where living with both biological parents serves as a reference category. For clarity of presentation the coefficients for living with a single parent and living with a stepparent are shown in separate graphs. The results illustrate a rather consistent picture of the association between family structure and substance use among European youth. There were no countries where living with a single parent or a stepparent was associated with significantly lower use of tobacco, alcohol or cannabis. Living with a single parent was significantly and positively associated with increased tobacco use in 25 of the 29 reporting countries. Similarly, living with one parent and a stepparent was significantly and positively associated with increased tobacco use in 23 of the 29 reporting countries. There were only three countries where cigarette smoking was neither related to living with a single parent nor living with one parent and a stepparent. These results thus show a clear and consistent pattern of increased smoking among European adolescents that do not live with both biological parents, regardless of the presence of a stepparent. The association between family structure and adolescent alcohol use was consistent with the association for cigarette smoking, but fewer significant coefficients were observed. Living with a single parent was associated with significantly more alcohol use in 12 of the 29 reporting countries. Living with one parent and a stepparent was associated with significantly more alcohol use in 15 of the 29 countries. Of the 10 countries that report no

194

Cannabis

Bulgaria Romania Greenland

Alcohol 0,01

-0,01

0,04 0,01

-0,01

0,04 0,04 0,04

-0,01

0,07

Netherlands -0,02

0,05

Croatia

0,07

0,01 0,02

Slovenia

0

Hungary

0

0,07 0,06 0,09 0,04 0,01

Estonia Slovak Republic

0,04

0,02

Russia (Moscow)

0,03

Ukraine

0,03

Germany (6 Bundesl.)

0,08 0,07

0,05 0,05 0,06 0,05 0,07

0,04

Turkey (6 cities)

0,06 0,06 0,09

0,01

0,07 0,04 0,04

Greece United Kingdom

0,09

0,05

Lithuania

0,03

0,07 0,08 0,07 0,1

0,07 0,06 0,08

Italy Austria France

Cigarettes

0,04

0,11 0,06 0,12

0

0,11

Malta

0,09

0,05

Denmark

0,09 0,1

0,02

0,11 0,06 0,05

Isle of Man

0,12

0,08 0,09 0,08 0,07

Iceland Norway

0,08 0,07 0,08

Sweden Belgium

0,05

Czech Republic

0,12 0,14

0,08

0,06

Switzerland

0,11

0,1

0,12 0,12

0,12 0,12 0,13 0,11

Finland -0,04

0,11

0

0,04

0,08

0,12

0,16

Figure 63. Use of cigarettes, alcohol and cannabis by living with a single parent. Filled bars: Significant correlations. Bars marked with lines: Non-significant correlations.

Correlates of adolescent substance use

Cannabis -0,01

Romania Turkey (6 cities)

-0,02

Alcohol

0

Cigarettes

0,03

0

0,05 0,02 0,01

Croatia

0,01

Malta

0,04 0,03 0,03 0,06

Greece -0,04

0,07

Bulgaria

0,04

0,01

Lithuania

0,02

Italy

0,02

Ukraine

0,02

Slovak Republic

0,02

0,05 0,04 0,07 0,05 0,06 0,07 0,08 0,08 0,1

Netherlands

0,05

Slovenia

0,05

0,07 0,08 0,1 0,08

0,04

Estonia Hungary

0,08 0,06

Czech Republic

0,09 0,11

0,03

Greenland

0,08 0,08

Germany (6 Bundesl.)

0,1 0,1 0,11

0,04

France

0,11 0,11

0,04

0,12

0,05

Sweden Switzerland

0,1

0,13 0,13

0,05

Denmark

0,08

0,11 0,1 0,12

0,08

Norway Belgium

0,13

0,1

0,06

Austria

0,11 0,12 0,11

0,07

0,13 0,1 0,11 0,11 0,11 0,09

Isle of Man Russia (Moscow) United Kingdom

0,09

0,15 0,12

0,1

Iceland 0

0,04

0,08

0,13

0,11

0,09

Finland

-0,04

0,11

0,09

significant association between family structure and alcohol use, seven countries were in the eastern part of Europe, and one each in the northern, southern and western parts. The association between family structure and alcohol use thus appears to be variable between countries, and less likely to be observed in the eastern part of Europe than in other regions. Finally, significantly more cannabis use was found among those who live with a single parent in 21 of the 29 countries. Similarly, such use was significantly higher among those who live with one parent and a stepparent in 22 of the 29 countries. There were only four countries where an association between family structure and cannabis use were not observed. These countries are all in the eastern part of Europe. Cannabis use thus appears to be higher among European adolescents that do not live with both biological parents. These results seem to suggest that the association between family structure is dependent upon the type of family and type substance in question. Tobacco use was most clearly and consistently associated with living with either a single parent or a parent and a stepparent. Alcohol use was associated with living with a single parent in some countries, and with living with a parent and a stepparent in other countries. It was only associated with both in eight countries, while no association between family structure and alcohol use was found in ten countries, most of the latter hailing from the eastern part of Europe. Cannabis use falls somewhat in between the two previously discussed substances, with about two-thirds of the countries reporting a significant association with each type of family structure other than living with both parents, and all but four reporting an association with at least one type of non-intact family structure.

0,12

0,15 0,14 0,14

0,16

Figure 64. Use of cigarettes, alcohol and cannabis by living with one parent and a stepparent. Filled bars: Significant correlations. Bars marked with lines: Non-significant correlations.

Correlates of adolescent substance use

195

Economic situation In most industrialised countries, lower economic and occupational status is associated with more smoking and alcohol use among adults (see Bjarnason, 2000). However, similar to parental education, the economic status of the family has generally not been found to be associated with adolescent substance use. There is nevertheless somewhat inconsistent evidence relating the socio-economic characteristics of residential neighbourhoods to adolescent substance use. Thus, research in the United States has tended to find substance use to be more prevalent in affluent, predominantly white towns and suburban neighbourhoods (Cronk and Sarvela, 1997; Ennett et al., 1997; Skager and Fisher, 1989), while research in England (Measham, 1996), Scotland (Glendinning, Shucksmith and Hendry, 1997) and Sweden (Hagquist, 1997) has found adolescent substance use to be positively associated with neighbourhood deprivation and proportion of blue-collar workers. In the ESPAD project the socio-economic background of students was measured by asking how well off they think their families are compared to other families. In 24 of the 31 reporting countries this measure of a poor economic status had no relationship whatsoever with adolescent cigarette use (figure 65). Furthermore, a poor economic situation as suggested by this measure was not significantly related to less smoking in any of the 31 reporting countries. However, in the remaining seven countries adolescents who reported that their families were worse off economically compared to other families were also more likely to smoke cigarettes. Five of these countries were in the north of Europe and two were Mediterranean countries. The findings were more mixed in relation to alcohol use. A poor economic status of the family was associated with significantly less drinking in eight of the reporting countries, and with significantly more drinking in six countries. In the remaining 17 countries there was no significant relationship between the reported economic status of the family and alcohol use among adolescents. There was no clear geographical pattern to these inconsistent results. Of the eight countries where poor economic status was associated with less drinking, two were located in the western, four in the eastern, one in the southern and one in the northern part of Europe. Of the six countries where the opposite correlation was observed, three were in the eastern part of Europe, one in the southern and two in the northern parts.

196

Cannabis Netherlands Greenland Turkey (6 cities) Russia (Moscow) Lithuania Czech Republic Slovenia Germany (6 Bundesl.) Estonia

Alcohol

-0,09 -0,01 -0,12

-0,05

0,06 -0,06 -0,04 -0,03 -0,06 -0,01 -0,03 -0,06

-0,01 -0,03 -0,03

-0,06

0,03

-0,04

Austria

-0,03 -0,01

Croatia

0

-0,01

Switzerland

Greece

0,01

0

-0,06

-0,02 -0,02 -0,01

Slovak Republic

0

-0,02 -0,03 -0,02

Faroe Islands

Belgium

0

-0,03 -0,05 -0,01 -0,04 -0,03 -0,02 -0,06 -0,04

Italy Hungary

0

-0,05

Malta Denmark

Cigarettes

0,03

0 0,01 0,03

-0,08

0,03 -0,04

0,02 0,01

-0,03

0,03

-0,01 -0,01

0,01 0,03 0,02 0,01 0,02 0,03 0,01 0,02 0,01 0,04 0,02

Ireland Sweden Romania 0

Bulgaria Norway Ukraine

0,05 0,02 0,03 0,02 0,03

-0,01

Cyprus Isle of Man

0,07

0,02 0

0,07 0,06 0,06

0

0,08 0,03 0,05

Finland United Kingdom

0,03

Iceland -0,15 -0,1 -0,05

0

0,05

0,08 0,08

0,09 0,08 0,08 0,07

0,1

0,15

Figure 65. Use of cigarettes, alcohol and cannabis by poor economic situation of the family. Filled bars: Significant correlations. Bars marked with lines: Non-significant correlations.

Correlates of adolescent substance use

Finally, a poor economic status of the family was associated with less cannabis use in seven countries. Two of these countries were in the western part of Europe, two in the eastern, two in the southern, and one in the northern part of Europe. Poor economic status was associated with more cannabis use in two countries, both of them in the northern part of Europe. In the remaining 22 countries there was no significant association between the economic status of the family and cannabis use. The preponderance of the evidence thus suggests that cannabis use is more prevalent in more affluent families in some European countries, but in the majority of countries there was no such association. In general these findings suggest that cigarette use is either unrelated to economic status or more common in poorer families, while cannabis use shows the opposite tendency. Alcohol use showed an inconsistent pattern that warrants further investigation.

Cannabis Greenland

Alcohol

0,03

Faroe Islands Cyprus

0,09 0,08 0,11 0,14 0,09 0,12

Turkey (6 cities) 0,11

Romania

Germany (6 Bundesl.) 0,14

Greece

Correlates of adolescent substance use

0,24 0,19

Malta Netherlands

0,2

0,3

0,25 0,25 0,28

0,24 0,26 0,25 0,26 0,21 0,3 0,3 0,17 0,32 0,33 0,29 0,25 0,3 0,22 0,33 0,3 0,28 0,26 0,31 0,22 0,33 0,32 0,29 0,3 0,28 0,27 0,28 0,32 0,27 0,29 0,32 0,31 0,28 0,31 0,32 0,26 0,32 0,29 0,28 0,35 0,34 0,26 0,33 0,34 0,25 0,35 0,23 0,35 0,37 0,35 0,29 0,32 0,35 0,3 0,32 0,35 0,29 0,33 0,18 0,41 0,41 0,27 0,39 0,35

Bulgaria Russia (Moscow) Sweden

Estonia

Research on the effects of parenting styles on adolescent substance use has frequently distinguished between parental support, parental monitoring and parental rule-setting. In general, strong parental support has been found to be associated with less substance use among European youth (Foxcroft and Lowe, 1995; Shucksmith, Glendinning and Hendry, 1997; Thorlindsson and Vilhjalmsson, 1991). Similarly, research has generally found parental monitoring to be associated with less adolescent substance use (Adlaf and Ivis, 1996; Barnes and Farrell, 1992; Beck et al., 1999; Glendinning, Shucksmith and Hendry, 1997; Jackson, Hendriksen and Dickinson, 1999; Krohn et al., 1993; Mulhall, Stone and Stone, 1996; Reifman et al., 1998). In contrast, studies of parental rule-setting have either found no such association, net of other factors (Barnes and Farrell, 1992; Beck et al., 1999; Jackson, Hendriksen and Dickinson, 1999), or a positive association between rule-setting and substance use (Bjarnason, 200; Hundleby and Mercer, 1987; Reifman et al., 1998). In the ESPAD study, students were asked if their parents know where they spend Saturday nights. The correlation between this single-item proxy measure of parental control and adolescent substance use is shown in figure 66. In 30 of the 31 reporting coun-

0,21 0,24 0,19 0,19 0,21 0,23 0,27 0,22 0,2 0,24 0,22 0,21 0,24

Austria

Belgium

Parental control

Cigarettes

Ukraine Norway Ireland Hungary Croatia Italy Denmark Slovak Republic Czech Republic United Kingdom Lithuania Switzerland Isle of Man Slovenia Finland Iceland 0

0,1

0,2

0,3

0,4

0,5

Figure 66. Use of cigarettes, alcohol and cannabis by parents not knowing where students spend Saturday night. Filled bars: Significant correlations. Bars marked with lines: Non-significant correlations.

197

Cannabis

Alcohol

tries, adolescents used substantially and significantly more tobacco, alcohol and cannabis when their parents did not know where they spent Saturday nights. The only exception to this pattern was Greenland, where this association was weaker and the population smaller, resulting in non-significant findings. These results overwhelmingly support the conclusion that parental control is strongly associated with all types of substance use among European youth.

Cigarettes

0,08 0,11 0,08

Estonia

0,2 0,14 0,18 0,22 0,13 0,21 0,21 0,24 0,13 0,26 0,12 0,24 0,28 0,12 0,22 0,14 0,2 0,28 0,19 0,2 0,24 0,13 0,19 0,32 0,18 0,2 0,26 0,2 0,16 0,28 0,23 0,18 0,24 0,26 0,18 0,27 0,3 0,13 0,28 0,26 0,19 0,28 0,17 0,24 0,33 0,28 0,19 0,28 0,31 0,16 0,29 0,31 0,2 0,26 0,23 0,26 0,29 0,28 0,24 0,27 0,29 0,22 0,3 0,26 0,22 0,34 0,3 0,25 0,27 0,24 0,27 0,33 0,32 0,22 0,3 0,31 0,25 0,29 0,34 0,23 0,29 0,33 0,22 0,32 0,22 0,29 0,37 0,3 0,33 0,29 0,27 0,32 0,33

Austria Germany (6 Bundesl.) Greenland Netherlands Belgium Greece Malta Romania Turkey (6 cities) Bulgaria Switzerland Italy France Hungary Cyprus Denmark United Kingdom Czech REpublic Faroe Islands Norway Slovak Republic Ukraine Russia (Moscow) Sweden Croatia Iceland Isle of Man Slovenia Lithuania Ireland Finland 0

0,1

0,2

0,3

0,4

Truancy Research in a variety of countries has rather consistently found adolescent substance use to be associated with higher levels of truancy and other measures of poor school performance (e.g. Arellano, Chaves and Deffenbacher, 1998; Costa, Jessor and Turbin, 1999; Ellickson et al., 1998; Thorlindsson et al., 1998). Furthermore, individual students are more likely to initiate substance use in schools where truancy is high and student commitment to school is low (Ennett et al., 1997; Hagquist, 1997). Figure 67 shows the correlation between the number of days a student has skipped school in the past 30 days and the number of times he or she has used different types of substances. In each and every one of the 32 reporting countries a positive correlation was found between truancy and use of cigarettes, alcohol and cannabis. The strength of this association varies between substances and across countries, but it was statistically significant in all cases. It can therefore be concluded with considerable confidence that truancy is associated with increased use of cigarettes, alcohol and cannabis among European students.

Sibling substance use

0,5

Figure 67. Use of cigarettes, alcohol and cannabis by truancy.

198

Finally, substance use by siblings has been argued to be among the strongest predictors of adolescent substance use (Stormshak et al., 2004). The ESPAD study provides an opportunity to examine the strength of this predictor across 31 reporting European countries. In the ESPAD questionnaire, students were asked if their elder siblings use various substances. The response categories were “yes”, “no”, “don’t know” and “don’t have any older siblings”. In the analysis

Correlates of adolescent substance use

Cannabis

Alcohol

shown in figure 68, only those students with older siblings were included. Following the argument that only sibling substance use known to the respondent can increase the probability of substance use initiation, this variable was coded 1: “yes”, 2: “no” or “don’t know”. The correlations were calculated as standardised regression coefficients with sibling use of each substance as a binary independent variable, and the respondent’s use as a continuos dependent variable. The results show that having an elder sibling who uses a particular substance was associated with more use by the younger sibling. This significant positive association was found for cigarettes, alcohol and cannabis in all 31 countries, with the single exception of cannabis use not attaining statistical significance in the Faroe Islands. It can therefore be concluded that having an elder sibling who uses tobacco, alcohol or cannabis is associated with an increased use of those substances among European students.

Cigarettes

0,07

Turkey (6 cities)

0,13 0,1 0,04 0,07

Romania

0,04

Lithuania

0,16 0,18

0,12 0,08

Slovak Republic Cyprus

0,07

0,19 0,17 0,2

0,13 0,1

Greece

0,2

Estonia

0,22 0,2

0,14 0,16 0,17 0,15 0,18

Czech Republic 0,05

Faroe Islands

0,18 0,17 0,18 0,19 0,23 0,13 0,18 0,17 0,2 0,19 0,22 0,14 0,21 0,15 0,21 0,24

Ireland Bulgaria Denmark Hungary Norway Greenland

0,15 0,14

0,3

0,31

Summary

0,27 0,16 0,18 0,22 0,17 0,25 0,22 0,22 0,2 0,25 0,17 0,22 0,3 0,16 0,19 0,25 0,22 0,19 0,27 0,19 0,22 0,31 0,18 0,21 0,3 0,19 0,21 0,22 0,28 0,22 0,25 0,24 0,24 0,17 0,27 0,29 0,29 0,2 0,25 0,28 0,23 0,24 0,28 0,22 0,26

Ukraine Malta Iceland Croatia Germany (6 Bunesl,) Switzerland Slovenia Italy Austria Finland Russia (Moscow) Sweden United Kingdom Netherlands Belgium Isle of Man

0,17

0

0,1

0,2

The association between adolescent substance use and family background is complex and dependent upon the type of substance, the element of family background and the country under study. Parental education and the economic status of the family have a positive association with substance use in some countries, but a negative association in other countries. Living with both biological parents is not found to be associated with increased substance use in any country, but the relative impact of living with a single parent or a parent and a stepparent differs between countries. In some countries, there is no significant association between family background and some types of substance use. In contrast, such factors as lack of parental control, skip ping school, and having a sibling that uses various substances are almost universally associated with increased use of tobacco, alcohol, and cannabis.

0,36 0,29

0,3

0,4

0,5

Figure 68. Use of cigarettes, alcohol and cannabis by sibling use.

Correlates of adolescent substance use

199

Key results country by country

erage during the last 12 months, been drunk during the last 12 months, lifetime use of cigarettes, cigarette smoking during the last 30 days, lifetime use of marijuana or hashish, lifetime use of any illicit drug other than marijuana or hashish, lifetime use of inhalants, lifetime use of tranquillisers or sedatives without a doctor’s prescription and lifetime use of alcohol together with pills. The results of each country are summarized in a graph, together with the unweighted averages of all participating ESPAD countries. This is done in order to facilitate the interpretation of the results, i.e. to compare each country’s prevalence rates with the mean of the ESPAD countries.

In the previous chapter one variable at a time has been presented and the results from all participating counties were compared in tables and figures. It is, however, also of interest to look at the results country by country. In this chapter some of the most important findings from each participating country are presented and briefly commented. For more detailed information on each variable, please see the tables (Appendix II). The methodology of each country’s study is presented in Appendix I, “Sampling and data collection in participating countries”. Nine variables were chosen to give an overview of the results: Consumption of any alcoholic bev-

Austria the average for all ESPAD countries. A slightly higher percentage (8%) than the average (6%) had reported use of any other illicit drug than cannabis. Inhalants were used by 14% compared to 10% on average. Very few (2%) Austrian students had used tranquillisers or sedatives without a doctor’s prescription (average 6%). Rather many students had used alcohol in combination with pills (13%), which is about double the ESPAD average for this variable (7%).

The Austrian figures for selected variables are overall higher than the average. The proportion that had been drinking alcohol during last 12 months is higher (93%) than the average (83%). The Austrian students had also been drunk during last 12 months to a higher degree (69%) than the ESPAD average (53%). In Austria 80% of the students had ever smoked which is higher than the ESPAD average (66%). Smoking during the last 30 days was reported by 49%, compared to an average of 35%. Cannabis use was reported by 21%, which is exactly

% 100

All countries

93 83

80

80 69 60

Austria

66

53

49 35

40

21

21

20 6

8

10

14

13 6

2

7

0 Any alc. last 12 months

Drunk last 12 months

Key results country by country

Ever smoked

Smoked last 30 days

Cannabis lifetime

Any drug but cannabis

Inhalants lifetime

Tranq./ sedatives

Alcohol +pills

201

Belgium in Belgium than the average for ESPAD countries (32 compared to 21%). On the other hand was use of any other drug than cannabis very similar to that of other ESPAD countries (8 and 6% respectively). Somewhat less students had been using inhalants in Belgium (7%) compared to the average (10%). Use of tranquillisers or sedatives was reported by 9% and use of pills in combination with alcohol by 6%. The average among the ESPAD countries was 6 and 7% respectively.

The Belgian students reported about the same prevalence of alcohol use during the last 12 months (86%) as the average for all ESPAD countries (83%). However, the proportion that had been drunk during the same period was lower (47 compared to 53%). Somewhat less students in Belgium had ever smoked (61%) compared to the average (66%). Also the proportion that had smoked during last 30 days was close to the average (32 and 35% respectively). Lifetime use of cannabis was more frequent %

All countries

100 83

Belgium

86

80 66 61 60

53 47 35

40

32

32 21

20 8

6

10

7

9

6

7

6

0 Any alc. last 12 months

Drunk last 12 months

Ever smoked

Smoked last 30 days

Cannabis lifetime

Any drug but cannabis

Inhalants lifetime

Tranq./ sedatives

Alcohol +pills

Bulgaria The Bulgarian students drink alcohol to about the same extent as the average in other ESPAD countries (86 and 83% respectively), and the same can be said about the frequency of intoxication during the last 12 months (56 and 53%). Somewhat higher proportions reported to have ever smoked (71%) compared to the average (66%) and the 30-days prevalence was higher (46%) than the average of all ESPAD countries (35%). The Bulgarian students had used cannabis to the same degree as the

average for all countries (21%), and the proportion that had ever tried any other drug was also close to the average (4 versus 6%). Very few students in Bulgaria had used inhalants (3%), which is much lower than the average (10%), and the same goes with the use of tranquillisers or sedatives (2 compared to 6%). Somewhat lower proportions than the average reported use of alcohol together with pills (4 versus 7%).

% All countries 83

Bulgaria

86

80 66 60

53

71

56 46

40

35 21 21

20 6

10 4

3

6

2

7

4

0 Any alc. last 12 months

202

Drunk last 12 months

Ever smoked

Smoked last 30 days

Cannabis lifetime

Any drug but cannabis

Inhalants lifetime

Tranq./ sedatives

Alcohol +pills

Key results country by country

Croatia The Croatian outcome on selected variables are very close to the average of all ESPAD countries. Thus, the proportion that had been drinking alcohol during the last 12 months was 82% (83% on average) and the proportion who had been drunk during the same period was only slightly lower (48%) than the average (53%). In Croatia 70% reported that they have ever been smoking (average 66%), and the proportion that had been smoking last 30 days

was 36%, with 35% as the average. Cannabis use was reported by 22% and any illicit drug use other than cannabis by 6%, which are the same levels as for all countries. Slightly more students in Croatia had been using inhalants (14 versus 10%), but use of tranquillisers or sedatives without a doctor’s prescription was the same as the average (6%). Use of alcohol together with pills was reported by 9% compared to 7% as the average.

% All countries

100 83

Croatia

82

80 66 60

53

70

48 35

40

36 21

22

20 6

10

6

14 6

6

9

7

0 Any alc. last 12 months

Drunk last 12 months

Ever smoked

Smoked last 30 days

Cannabis lifetime

Any drug but cannabis

Inhalants lifetime

Tranq./ sedatives

Alcohol +pills

Cyprus Only 4% reported use of cannabis and 3% experience with other illicit drugs, compared to 21 and 6% respectively for all ESPAD countries. However, experience with inhalants were more common in Cyprus (17%) than in the average of the ESPAD countries (10%). The proportion of students who reported use of tranquillisers or sedatives without a doctor’s prescription is the same as the average 6%. Very few students in Cyprus reported use of alcohol together with pills (2%) in comparison to all countries (7%).

There is a substantial difference between any alcohol consumption during the last 12 months and drunkenness experience during the same period in Cyprus. The former variable was broadly the same as the average (79 versus 83%), while the latter was about half the average (25% compared to 53%). Lifetime smoking was also less reported in Cyprus (54%) than the average (66%), and the difference is even more pronounced in the last 30 days prevalence of smoking (22 compared to 35%). Experience of illicit drugs is very uncommon in Cyprus.

% All countries

100 83 80

Cyprus

79 66

60

54

53

35

40 25

22

21

17

20 4

6

10 3

6

6

7

2

0 Any alc. last 12 months

Drunk last 12 months

Key results country by country

Ever smoked

Smoked last 30 days

Cannabis lifetime

Any drug but cannabis

Inhalants lifetime

Tranq./ sedatives

Alcohol +pills

203

The Czech Republic the average for all countries (21%). The use of any other illicit drug than cannabis is also higher than the average (12 compared to 6%). Use of inhalants, however, is about the same in the Czech Republic (9%) as the average (10%). Rather large proportions have used tranquillisers or sedatives without a doctor’s prescription (11%) compared to all countries (6%). Also alcohol in combination with pills is more common in the Czech Republic (12%) than the average (7%).

Almost all students in the Czech Republic had used alcohol during the last 12 months (95%), which is higher than the average (83%). Also the proportion of students who had been drunk during the last 12 months is higher (68%) than the average (53%). More students than the average had been smoking in lifetime (80 compared to 66%), while the proportion having smoked during the last 30 days (43%) is closer to the proportions in all countries (35%). About twice as many students in the Czech Republic had used marijuana or hashish (44%) as %

All countries

95

100 83

Czech Republic

80

80 68 60

66

53 44

43 35

40

21 20

12

10

6

9

11

6

7

12

0 Any alc. last 12 months

Drunk last 12 months

Ever smoked

Smoked last 30 days

Cannabis lifetime

Any drug but cannabis

Inhalants lifetime

Tranq./ sedatives

Alcohol +pills

Denmark marijuana or hashish in lifetime (23 compared to 21%). However, the experience of any other illicit drug than cannabis is on the same level as the average (6%). The use of inhalants (8 versus 10%) as well as the use of tranquillisers or sedatives without a doctor’s prescription (4 versus 6%) are rather close to the mean proportions for all ESPAD countries. The proportions reporting alcohol use in combination with pills is the same as the ESPAD average (7%).

The proportion of students in Denmark who had been drinking alcohol during the last 12 months is higher (95%) than the average (83%). The difference is, however, more pronounced when comparing the proportions of students who had been drunk during the same period (82% compared to 53%). The proportion of students who had ever smoked is about the same (64%) as the average (66%) and the figure of the 30 days prevalence is somewhat lower (30%) than the average (35%). It is slightly more common in Denmark than the average to have used

% All countries

95

100 83

Denmark

82

80 66 60

64

53 35

40

30 21

23

20 6

6

10

8

6

4

7

7

0 Any alc. last 12 months

204

Drunk last 12 months

Ever smoked

Smoked last 30 days

Cannabis lifetime

Any drug but cannabis

Inhalants lifetime

Tranq./ sedatives

Alcohol +pills

Key results country by country

Estonia tively). The prevalence rates of cannabis use is slightly higher than the average (23 compared to 21%). There are more Estonian students than the average that have used any illicit drug than cannabis (10 compared to 6%) and the same is true for tranquillisers and sedatives without a doctor’s prescription (9 and 6% respectively). Inhalants had been used by 8% in Estonia compared to 10% as the ESPAD average. The corresponding figures for alcohol together with pills are 6 and 7% respectively.

A somewhat higher proportion than the average for all ESPAD countries had been drinking alcohol during the last 12 months in Estonia (87 compared to 83%). The number reporting having been drunk during the same period is, however, higher than the average (68 compared to 53%). The proportion of students who reported to have ever been smoking was also higher than the average for all countries (77 compared to 67%), while the proportion who had been smoking during the last 30 days was about the same as the average (37 and 35% respec%

All countries

100 83

Estonia

87 77

80 68 60

66

53 35

40

37 21

23

20

10

6

10

8

9

6

7

6

0 Any alc. last 12 months

Drunk last 12 months

Ever smoked

Smoked last 30 days

Cannabis lifetime

Any drug but cannabis

Inhalants lifetime

Tranq./ sedatives

Alcohol +pills

The Faroe Islands used any illicit drug. The proportion of students who had used marijuana or hashish was less than half the average (9% versus 21%) and the same can be said about any other illicit drug than cannabis (2 compared to 6%). The proportion reporting use of inhalants was about the same as the average (11 and 10% respectively) and this is also true regard ing the use of tranquillisers or sedatives without a doctor’s prescription (5 versus 6%). The use of alcohol together with pills is slightly higher than the average (10 compared to 7%).

The proportion of students in the Faroe Islands who had been drinking alcohol during the last 12 months was lower than the average (76 compared to 83%), while the proportion of students who had been drunk during the same period was slightly above average (57 versus 53%). However, the lifetime smoking prevalence is substantially higher in the Faroe Islands (83%) than the average for all ESPAD countries (66%) and the 30 days prevalence of smoking slightly higher (41 compared to 35%). Very few students in the Faroe Islands had

% All countries

100 83 80

Faroe Islands

83 76 66

60

53

57 41

40

35 21

20 9

6

10

11

2

6

5

7

10

0 Any alc. last 12 months

Drunk last 12 months

Key results country by country

Ever smoked

Smoked last 30 days

Cannabis lifetime

Any drug but cannabis

Inhalants lifetime

Tranq./ sedatives

Alcohol +pills

205

Finland (38 versus 35%). Smaller proportions than average reported use of marijuana or hashish (11 versus 21%) as well as use of illicit drugs other than cannabis (3 versus 6%). The proportions reporting use of inhalants is about the same as the average (8 compared to 10%) and the same is true for tranquillisers and sedatives without a doctor’s prescription (7 versus 6%). In Finland it is more common to have used alcohol in combination with pills (12%) than the average for all ESPAD countries (7%).

In Finland the proportion of students who had been drinking any alcohol during the last 12 months is broadly the same as the average for all countries (80 compared to 83%). The 12 months prevalence of being drunk is, however, substantially higher than average (64 compared to 53%). The proportion of students who had ever smoked cigarettes is somewhat higher in Finland than the average for all ESPAD countries (70 compared to 66%) and the same holds true regarding the 30 days prevalence %

All countries

100 83 80

Finland

80 64

60

66

70

53

40

35

38 21

20

11

6

10

8

3

6

7

7

12

0 Any alc. last 12 months

Drunk last 12 months

Ever smoked

Smoked last 30 days

Cannabis lifetime

Any drug but cannabis

Inhalants lifetime

Tranq./ sedatives

Alcohol +pills

France who had used marijuana or hashish is about twice the average of all countries (38 versus 21%), but the proportion reporting use of any other illicit drug but cannabis is about the same as the average figure (7 compared to 6%). Use of inhalants is also about the same as the average (11 versus 10%), while the use of tranquillisers or sedatives without a doctor’s prescription is above average (13 versus 6%). Use of alcohol together with pills is reported by a proportion equal to the average (7%).

The proportion of students in France who had consumed any alcohol during the previous 12 months is about the same as the average for all ESPAD countries (80 compared to 83%). Moreover, the proportion reporting having been drunk during the same period is substantially smaller than the average (29 versus 53%). The lifetime prevalence rates of smoking cigarettes is about average (68%) and this is also true for the 30 days prevalence of smoking (33%). The proportion of students in France

% All countries

100 83 80

France

80 66

60

68

53 35

40 29

38

33 21

20

6

7

10

11

13 7

6

7

0 Any alc. last 12 months

206

Drunk last 12 months

Ever smoked

Smoked last 30 days

Cannabis lifetime

Any drug but cannabis

Inhalants lifetime

Tranq./ sedatives

Alcohol +pills

Key results country by country

Germany (six Bundesländer) hashish (27%) than the average for all countries (21%). The use of any other illicit drug than cannabis is also higher than the average (10 compared to 6%). Use of inhalants, however, is about the same in Germany (11%) as the average (10%). Rather small proportions have used tranquillisers or sedatives without a doctor’s prescription (2%) compared to all countries (6%). On the other hand, alcohol in combination with pills is more common in Germany (16%) than the average (7%).

Almost all students in Germany had used alcohol during the last 12 months (93%), which is higher than the average (83%). Also the proportion of students who had been drunk during the last 12 months is higher (61%) than the average (53%). More students than the average had been smoking in lifetime (77 compared to 66%) and the tendency is the same about the proportion that have smoked during the last 30 days (45 compared to 35%). More students in Germany had used marijuana or % 100

All countries

93

Germany (6 Bundesl.)

83 77

80 66 61 60

53 45 35

40

27 21 20

10

10

6

11

16 6

2

7

0 Any alc. last 12 months

Drunk last 12 months

Ever smoked

Smoked last 30 days

Cannabis lifetime

Any drug but cannabis

Inhalants lifetime

Tranq./ sedatives

Alcohol +pills

Greece and the use of any illicit drug but cannabis shows a difference in the same direction (3 versus 6%). The proportion of students who had used inhalants is higher in Greece (15%) than the average for all ESPAD countries (10%), while the use of tranquillisers or sedatives without a doctor’s prescription is closer to the average (4 versus 6%). The proportion reporting use of alcohol in combination with pills is much smaller than the average for all countries (2 compared to 7%).

A vast majority of the students in Greece had been drinking an alcoholic beverage during the last 12 months (91%), which is above average (83%). In contrast, less than the average had been drunk during the same period (37% compared to 53%). Lifetime smoking among the Greek students is also below average (50 versus 66%) and the 30 days prevalence of smoking has the same tendency (28 compared to 35%). The use of marijuana or hashish is much lower than average (6 compared to 21%)

% All countries

100

Greece

91 83

80 66 60

53

50 37

40

35 28 21

20 6

6

10

15

3

6

4

7

2

0 Any alc. last 12 months

Drunk last 12 months

Key results country by country

Ever smoked

Smoked last 30 days

Cannabis lifetime

Any drug but cannabis

Inhalants lifetime

Tranq./ sedatives

Alcohol +pills

207

Greenland tion of students who report having used marijuana or hashish is also higher than the average (27 versus 21%), while the use of any other illicit drugs is less frequent (4 compared to 6%). The use of inhalants is much more common in Greenland (22%) than the average of the ESPAD students (10%). However, the figures are in the opposite direction for tranquillisers and sedatives without a doctor’s prescription (3 versus 6%) as well as for alcohol together with pills (2 compared to 7%).

The proportion of students in Greenland who had any alcohol consumption during the previous 12 months is lower than the average for all countries (73 versus 83%). In contrast, the proportion reporting having been drunk during the same period is substantially higher (70%) than average (53%). There are also relatively more students in Greenland who had ever been smoking (79%) and who had smoked during the last 30 days (60%) than the averages (66 and 35% respectively). The propor%

All countries

100 83 80

Greenland

79 73

70

66 60

60

53

40

35 27

22

21

20

6

10

6

4

3

7

2

0 Any alc. last 12 months

Drunk last 12 months

Ever smoked

Smoked last 30 days

Cannabis lifetime

Any drug but cannabis

Inhalants lifetime

Tranq./ sedatives

Alcohol +pills

Hungary marijuana or hashish is lower than average (16 compared to 21%), while the use of any illicit drug other than cannabis is about average (5%). The use of inhalants is less common in Hungary than the average of all ESPAD countries (5 versus 10%). The proportion of students who ever used tranquillisers or sedatives without a doctor’s prescription is above average (10 compared to 6%) and the tendency is the same for alcohol together with pills (11 and 7% respectively).

The proportion of students in Hungary who had consumed alcohol during the last 12 months is about the same as the average for all countries (84 compared to 83%). However, the proportion reporting having been drunk during the last 12 months is lower than the average (46 versus 53%). The proportion of students who had ever smoked is slightly higher than the average for all countries (72 and 66% respectively) and this holds true also regarding the 30 days prevalence (39 versus 35%). The proportion of Hungarian students who have used

% All countries

100

Hungary

83 84 80 66 60

72

53 46 35

40

39 21

20

16 6

5

10

5

6

10

7

11

0 Any alc. last 12 months

208

Drunk last 12 months

Ever smoked

Smoked last 30 days

Cannabis lifetime

Any drug but cannabis

Inhalants lifetime

Tranq./ sedatives

Alcohol +pills

Key results country by country

Iceland hashish is also less frequent than the average (13 compared to 21%). However, the use of any illicit drug other than cannabis is equal to the ESPAD average (6%). Lifetime use of inhalants is only slightly higher than the average (12 versus 10%). Also the use of tranquillisers or sedatives without a doctor’s prescription and alcohol in combination with pills show the same tendency (9 versus 6% and 8 versus 7% respectively).

The proportion of Icelandic students who had consumed any alcohol during the last 12 months is lower than the average for all ESPAD countries (64 compared to 83%), and the same is true for the proportion that reported having been drunk during the same period (47 versus 53%). Smoking is less common in Iceland than in most other countries; the lifetime prevalence is 46% compared to 66% on average, and 30 days prevalence is 20% compared to the average of 35%. The use of marijuana or %

All countries

100

Iceland

83 80 66

64 60

53 47

46 35

40

20

20

21 13 6

10

6

12

9

6

7

8

0 Any alc. last 12 months

Drunk last 12 months

Ever smoked

Smoked last 30 days

Cannabis lifetime

Any drug but cannabis

Inhalants lifetime

Tranq./ sedatives

Alcohol +pills

Ireland the average for all ESPAD countries (39 versus 21%), while the use of illicit drugs other than cannabis only is slightly above average (9 versus 6%). Use if inhalants, however, is about twice the average (18 compared to 10%). There are fewer Irish students than the ESPAD average that have used tranquillisers or sedatives without a doctor’s prescription (2 compared to 6%). A slightly higher proportion than average reported use of alcohol in combination with pills (9 versus 7%).

The proportion of Irish students who had been drinking any alcohol during the last 12 months is a little higher than average (88 compared to 83%). However, the proportion that had been drunk during the same period is substantially higher than the average (72 versus 53%). The lifetime smoking prevalence is about the same as the average (67 compared to 66%) and the same is true for the 30 days prevalence (33 versus 35%). The use of marijuana or hashish is twice as common in Ireland than

% All countries

100 83

Ireland

88

80

72

60

66

67

53 35

40

39 33 21

18

20 6

9

10

6

2

7 9

0 Any alc. last 12 months

Drunk last 12 months

Key results country by country

Ever smoked

Smoked last 30 days

Cannabis lifetime

Any drug but cannabis

Inhalants lifetime

Tranq./ sedatives

Alcohol +pills

209

Isle of Man for all ESPAD countries (39 versus 21%). The use of illicit drugs other than cannabis is also above average (10 versus 6%). Use if inhalants is about twice the average (19 compared to 10%). However, the use of tranquillisers or sedatives without a doctor’s prescription is about the same as the average (5 and 6% respectively). A higher proportion than average reported use of alcohol in combination with pills (10 versus 7%).

The proportion of students who had been drinking any alcohol during the last 12 months is higher than average (94 compared to 83%) and the proportion that had been drunk during the same period is substantially higher than the average (71 versus 53%). The lifetime smoking prevalence is a little lower than the average (60 compared to 66%) and the same is true for the 30 days prevalence (30 versus 35%). The use of marijuana or hashish is twice as common in Isle of Man than the average % 100

All countries

94

Isle of Man

83 80

71

66 60

60

53 35

40

39 30 21

19

20

10

6

10

6

5

7

10

0 Any alc. last 12 months

Drunk last 12 months

Ever smoked

Smoked last 30 days

Cannabis lifetime

Any drug but cannabis

Inhalants lifetime

Tranq./ sedatives

Alcohol +pills

Italy hashish is higher than average (27 versus 21%), while the use of illicit drugs other than cannabis is broadly the same (8 versus 6%). The use of inhalants is lower than average (6 compared to 10%) and the use of tranquillisers or sedatives without a doctor’s prescription is the same as average (6%). Use of alcohol in combination with pills is less common in Italy than in many other ESPAD countries (3% in comparison with 7% as the average).

Consumption of any alcohol during the last 12 months is as common among Italian students as the average of all ESPAD countries (82 versus 83%). However there are fewer Italian students who had been drunk during the same period (37 versus 53%). Lifetime smoking is as common as the average (64 versus 66%), and the same is true regarding the proportion of students who have been smoking during the last 30 days (38 versus 35%). The proportion of students who have used marijuana or

% All countries

100 83

Italy

82

80 66 60

64

53 37

40

35

38 27 21

20 6

8

10

6

6 6

7

3

0 Any alc. last 12 months

210

Drunk last 12 months

Ever smoked

Smoked last 30 days

Cannabis lifetime

Any drug but cannabis

Inhalants lifetime

Tranq./ sedatives

Alcohol +pills

Key results country by country

Latvia ish is smaller than the ESPAD average (16 compared to 21%) while the lifetime prevalence of any drug but cannabis is about the same (5 versus 6%). Use of inhalants is less common than the average for all countries (7 compared to 10%), and this holds true also for use of tranquillisers or sedatives without a doctor’s prescription (3 and 6% respectively). Use of alcohol in combination with pills is about as common in Latvia as the average for all ESPAD countries (6 versus 7%).

The proportion of Latvian students who had been drinking any alcohol during the last 12 months is somewhat higher than average for all countries (87 versus 83%). The tendency is the same regarding the proportion of students who had been drunk during the same period (57 compared to 53%). The lifetime prevalence of smoking is higher in Latvia than average (78 versus 66%) and so is the 30 days prevalence (40 and 35% respectively). The proportion of students who have used marijuana or hash%

All countries

100 83

Latvia

87 78

80 66 60

53

57

35

40

40 21

20

16 6

5

10

7

6

7 6

3

0 Any alc. last 12 months

Drunk last 12 months

Ever smoked

Smoked last 30 days

Cannabis lifetime

Any drug but cannabis

Inhalants lifetime

Tranq./ sedatives

Alcohol +pills

Lithuania proportion that reported use of any other illicit drug than cannabis is about the same (7 and 6% respectively). The use of inhalants in Lithuania is less common than the average for all countries (5 and 10% respectively). The use of tranquillisers or sedatives without a doctor’s prescription is about twice the average (14 versus 6%). However, the proportion of students who have used alcohol together with pills is the same as the average for all countries (7%).

A vast majority of the students in Lithuania had been drinking alcohol during the last 12 months (94 compared to 83% on average). The proportion of students who had been drunk during the same period is also higher than average (66 versus 53%). The lifetime prevalence of smoking is higher than the average for all ESPAD countries (80 compared to 66%) and the same is true for the 30 days prevalence (41 compared to 35%). The proportion of students who have used marijuana or hashish is lower than the average (13 versus 21%), while the

% 100

All countries

94 83

80

80 66 60

Lithuania

66

53 41

40

35 21

20

13 6

7

10

14 5

7

6

7

Tranq./ sedatives

Alcohol +pills

0 Any alc. last 12 months

Drunk last 12 months

Key results country by country

Ever smoked

Smoked last 30 days

Cannabis lifetime

Any drug but cannabis

Inhalants lifetime

211

Malta half the average for all countries (10 versus 21%), as is the proportion reporting use of illicit drugs other than cannabis (4 compared to 6%). Use of inhalants, however, is reported by 16% of the students in Malta compared to only 10% as the average. Tranquillisers and sedatives without a doctor’s prescription is less common in Malta than the ESPAD average (3 compared to 6%) while the tendency is the opposite for alcohol together with pills (9 and 7% respectively).

A vast majority of the students in Malta had been drinking alcohol during the last 12 months (90 compared to the average of 83%). In contrast, the proportions reporting drunkenness during the same period is less than average (38 versus 53%). This holds true also regarding lifetime and the 30 days prevalence of smoking cigarettes. The lifetime figure is 48% (66% on average) and the 30 days prevalence 27% (35% on average). The proportion of students who have used marijuana or hashish is %

All countries

100

Malta

90 83

80 66 60

53

48 38

40

35 27

21

20

16 10

10

6

6

4

3

9

7

0 Any alc. last 12 months

Drunk last 12 months

Ever smoked

Smoked last 30 days

Cannabis lifetime

Any drug but cannabis

Inhalants lifetime

Tranq./ sedatives

Alcohol +pills

The Netherlands in the Netherlands than the average for other countries (28 compared to 21%). On the other hand was use of any other drug than cannabis similar to that of other ESPAD countries (6%). Less students had been using inhalants in the Netherlands (6%) compared to the average (10%). Use of tranquillisers or sedatives was reported by 8% and use of pills in combination with alcohol by 4%. The average among other ESPAD countries was 6 and 7% respectively.

The Dutch students reported about the same prevalence of alcohol use during the last 12 months (85%) as the average for all ESPAD countries (83%). However, the proportion that had been drunk during the same period was lower (46 compared to 53%). Less students in the Netherlands had ever smoked (57%) compared to the average (66%). Also the proportion that had smoked during the last 30 days was a little lower than the average (31 and 35% respectively). Lifetime use of cannabis was more frequent

% All countries

100 83

Netherlands

85

80 66 60

57

53 46

35

40

31

28 21

20 6

6

10

6

6

8

7

4

0 Any alc. last 12 months

212

Drunk last 12 months

Ever smoked

Smoked last 30 days

Cannabis lifetime

Any drug but cannabis

Inhalants lifetime

Tranq./ sedatives

Alcohol +pills

Key results country by country

Norway marijuana or hashish is much lower than the average (9 compared to 21%) and the tendency is the same for the use of any illicit drug but cannabis (3 and 6% respectively). Use of inhalants also goes in the same direction (5 versus 10%), as well as the use of tranquillisers or sedatives without a doctor’s prescription (3 compared to 6%). The use of alcohol in combination with pills is also less common in Norway (5%) than the average of all ESPAD countries (7%).

The proportion of students in Norway, who had been drinking any alcohol during the last 12 months, is somewhat lower than the average for all ESPAD countries (76 versus 83%), while the proportion reporting drunkenness experience during the same period is about the same (54 compared to 53%). Smoking among the Norwegian students is a little less common than the average for all countries (62 compared to 66%) and the tendency is the same about smoking during the last 30 days (28 versus 35%). The proportion of students who have used %

All countries

100

Norway

83 76

80

66 60

53

62

54 35

40

28 21 20

9

10

6

3

5

6

3

7

5

0 Any alc. last 12 months

Drunk last 12 months

Ever smoked

Smoked last 30 days

Cannabis lifetime

Any drug but cannabis

Inhalants lifetime

Tranq./ sedatives

Alcohol +pills

Poland to 21%) and the proportion reporting use of illicit drugs other than cannabis is about the same as the average (6 and 7% respectively). Use of inhalants is also as common in Poland as the average of all countries (9 and 10% respectively). The use of tranquillisers or sedatives without a doctor’s prescription, however, is substantially higher than in many other countries (17 compared to 6% on average). The use of alcohol together with pills is close to the average for all countries (9 versus 7%).

The consumption of alcohol during the 12 previous months among Polish students is about equal to the average of all ESPAD countries (85 compared to 83%) and the proportion reporting drunkenness during the same period is rather close to average (48 versus 53%). The lifetime smoking figure is about average (67%), while the 30 days prevalence figure is slightly lower (31 compared to 35%). The proportion of students who have ever used marijuana or hashish is close to average (18 compared

% All countries

100 83

Poland

85

80 66 60

67

53 48 35

40

31 21

20

18

17 6

7

10

9

6

7

9

0 Any alc. last 12 months

Drunk last 12 months

Key results country by country

Ever smoked

Smoked last 30 days

Cannabis lifetime

Any drug but cannabis

Inhalants lifetime

Tranq./ sedatives

Alcohol +pills

213

Portugal use of marijuana or hashish is smaller than the average for all ESPAD countries (15 compared to 21%), while the use of any other illicit drug than cannabis is about average (7 versus 6%). Use of inhalants is slightly lower than the ESPAD average (8 and 10% respectively) and the same is true for the use of tranquillisers or sedatives without a doctor’s prescription (4 versus 6%). Alcohol together with pills is reported by fewer students in Portugal (3%) than the average (7%).

The proportion of Portuguese students who had consumed alcohol during the last 12 months is slightly lower than the average (78 compared to 83%). However, the proportion of students who report having been drunk during the same period is substantially lower than average (32 versus 53%). Also the lifetime and 30 days prevalence of smoking cigarettes are lower than the averages. The lifetime figure is 62% (66% on average) and the 30 days figure 28% (35% on average). The lifetime %

All countries

100

Portugal

83 80

74 66

60

62

53 35

40 28

28 21

20

15

10

7

6

8

6

5

7

4

0 Any alc. last 12 months

Drunk last 12 months

Ever smoked

Smoked last 30 days

Cannabis lifetime

Any drug but cannabis

Inhalants lifetime

Tranq./ sedatives

Alcohol +pills

Romania much below the average (21%). The proportion of students who reported use of any illicit drug other than cannabis is also lower than the average (2 compared to 6%). Very few students in Romania had used inhalants (2 compared to 10%), while the use of tranquillisers or sedatives without a doctor’s prescription was about the same as the average (5 and 6% respectively). The proportion of students who had used alcohol in combination with pills was 3%, which is half the average (7%).

The proportion of students in Romania who had consumed any alcohol during the last 12 months is close to the average for all ESPAD countries (80 versus 83%), while the proportion reporting drunkenness during the same period is substantially lower (36 compared to 53%). The lifetime smoking figure (62%) is close to the ESPAD average (66%), while the 30 days prevalence figure (29%) is lower than the average (35%). Very few students (3%) reported use of marijuana or hashish, which is

% All countries

100 83 80

Romania

79 66

60

64

53 36

40

35 28 21

20 3

6

10 3

2

6

6

7

3

0 Any alc. last 12 months

214

Drunk last 12 months

Ever smoked

Smoked last 30 days

Cannabis lifetime

Any drug but cannabis

Inhalants lifetime

Tranq./ sedatives

Alcohol +pills

Key results country by country

Russia (Moscow) 21%), as is the proportion that reported use of any other illicit drug than cannabis (4 versus 6%). Use of inhalants was reported by 7%, which is slightly lower than the average for all countries (10%). The tendency is the same for use of tranquillisers or sedatives without a doctor’s prescription (3 compared to 6%), while the use of alcohol together with pills is about the same as the ESPAD average (6 versus 7%).

In Russia 86% had been drinking any alcoholic beverage during the last 12 months and 53% had been drunk during the same period, which is very close to and equal to the averages for all ESPAD countries (83 and 53% respectively). The lifetime prevalence of smoking cigarettes is above average (74 versus 66%) and the same is true regarding the 30 days prevalence (44 versus 35%). The proportion of students who had used marijuana or hashish is about the same as the average (22 compared to %

All countries

100 83

Russia (Moscow)

86

80

74 66

60

53

53 44

40

35 21

20

22 6

10 4

7

6

7

3

6

0 Any alc. last 12 months

Drunk last 12 months

Ever smoked

Smoked last 30 days

Cannabis lifetime

Any drug but cannabis

Inhalants lifetime

Tranq./ sedatives

Alcohol +pills

The Slovak Republic the Slovakian students had used marijuana or hashish (27%) than the average for all countries (21%), while the proportion reporting use of illicit drugs other than cannabis is equal (6%). Inhalants are used in the Slovak Republic to the same extent as the average (9 versus 10%) and about the same is true for tranquillisers and sedatives without a doctor’s prescription (4 compared to 6%). However, many more had used alcohol together pills than the ESPAD average (15 and 7% respectively).

A vast majority of the students in the Slovak Republic had been drinking alcohol during the last 12 months (90%), which is higher than the average for all ESPAD countries (83%). The tendency was the same about the proportion reporting drunkenness during the last 30 days (57 compared to 53%). Also the lifetime prevalence of smoking cigarettes was a bit higher among students in the Slovak Republic (74 versus 66%). However, the 30 days prevalence figure was about the same as the ESPAD average (37 and 35% respectively). A higher proportion of

% All countries

100

Slovak Republic

90 83

80

74 66

60

53

57

35

40

37 27 21

20 6

6

10 9

15 6

4

7

0 Any alc. last 12 months

Drunk last 12 months

Key results country by country

Ever smoked

Smoked last 30 days

Cannabis lifetime

Any drug but cannabis

Inhalants lifetime

Tranq./ sedatives

Alcohol +pills

215

Slovenia The proportions of Slovenian students who had been drinking any alcohol during the last 12 months is the same as the ESPAD average (83%) and the number that had been drunk during the previous 12 months is very close to the average (56 and 53% respectively). The lifetime prevalence of smoking cigarettes is very equal (67 versus 66%), as is the 30 days prevalence (36 compared to 35%). The proportion of students who have used mari-

juana or hashish is higher than the average (28 compared to 21%), while the use of other illicit drugs is about equal (5 compared to 6%). The use of inhalants is higher (15%) than average (10%) and the use of tranquillisers or sedatives without a doctor’s prescription as well as alcohol in combination with pills are both very close to the averages of all countries (5 and 6% respectively).

% All countries

100 83

Slovenia

83

80 66 60

53

67

56 35

40

36 28 21

20 6

5

10

15 6

5

7

6

0 Any alc. last 12 months

Drunk last 12 months

Ever smoked

Smoked last 30 days

Cannabis lifetime

Any drug but cannabis

Inhalants lifetime

Tranq./ sedatives

Alcohol +pills

Sweden marijuana or hashish is reported by 7%, which is one third of the average of all countries (21%) and the proportion reporting use of illicit drugs other than cannabis is about half (3 versus 6%). The proportion of students who had used inhalants is close to average (8 compared to 10%). The proportion reporting use of tranquillisers or sedatives without a doctor’s prescription is equal to the ESPAD average (6%) and the proportion is also about the same for alcohol together with pills (8% in Sweden and 7% as the average).

The proportion of Swedish students who had been drinking any alcohol during the last 12 months is a little lower than the average of all ESPAD countries (77 versus 83%). However, the proportion reporting drunkenness during the same period is rather equal to the average (55 compared to 53%). The lifetime prevalence of smoking cigarettes is a little lower than average (60 versus 66%), while the difference is more pronounced when it comes to the proportion of students who had smoked during the last 30 days (23 compared to 35%). Use of

% All countries

100

Sweden

83 80

77 66

60

53

55

60

35

40

23

21

20 7

6

10 3

8

6

6

7

8

0 Any alc. last 12 months

216

Drunk last 12 months

Ever smoked

Smoked last 30 days

Cannabis lifetime

Any drug but cannabis

Inhalants lifetime

Tranq./ sedatives

Alcohol +pills

Key results country by country

Switzerland in Switzerland than the average for other countries (40 compared to 21%). On the other hand use of any other illicit drug than cannabis was equal to the average of the ESPAD countries (6%). Somewhat less students had been using inhalants in Switzerland (7%) compared to the average (10%). Use of tranquillisers or sedatives was reported by 6%, which is the same as the average. The use of pills in combination with alcohol was a little less common among Swiss students (4 and 7% respectively).

The Swiss students reported slightly higher prevalence of alcohol use during the last 12 months (88%) than the average for all ESPAD countries (83%), while the tendency was the opposite when it comes to the proportion that had been drunk during the same period (49 compared to 53%). The proportion of lifetime smokers (64%) was about the same as the average (66%). Also the proportion that had smoked during the last 30 days was very close to the average (34 and 35% respectively). Lifetime use of cannabis was much more frequent %

All countries

100 83

Switzerland

88

80 66 60

53

64

49

40

35

40 34 21

20

10

6

6

7

7

6

6

4

0 Any alc. last 12 months

Drunk last 12 months

Ever smoked

Smoked last 30 days

Cannabis lifetime

Any drug but cannabis

Inhalants lifetime

Tranq./ sedatives

Alcohol +pills

Turkey (six cities) smoking during the last 30 days were 18 and 35% respectively. Very few (4%) had used cannabis, which is much lower than the ESPAD average (21%). Any other illicit drug but cannabis was reported by 3% of the students in Turkey and by 6% as the average. Inhalants had been used by 4 versus 10%, tranquillisers and sedatives without a doctor’s prescription by 3 versus 6% and alcohol together with pills by 2 versus 7%.

Turkey is the only country in which the students show lower figures than the ESPAD average for all the nine variables summarised in this chapter. Much fewer had been drinking alcohol during the last 12 months (35 and 83% respectively) and the difference is also substantial when it comes to drunkenness during the same period (16 compared to 53%). Lifetime smoking of cigarettes was reported by 50% in Turkey and among 66% in the ESPAD countries. The corresponding figures for

% All countries

100

Turkey (6 cities)

83 80 66 60 40

53

50

35

35 18

16

20

21 4

6

10 3

4

6

3

7

2

0 Any alc. last 12 months

Drunk last 12 months

Key results country by country

Ever smoked

Smoked last 30 days

Cannabis lifetime

Any drug but cannabis

Inhalants lifetime

Tranq./ sedatives

Alcohol +pills

217

Ukraine The proportion of Ukrainian students who had been drinking any alcohol during the last 12 months is about equal to the average of all ESPAD countries (84 versus 83%), while the proportion reporting drunkenness is above (66 versus 53%). Lifetime and 30 days prevalence of smoking cigarettes are both slightly higher than the average (70 versus 66% for lifetime smoking and 39 versus 35% for the last 30 days prevalence). The proportion of students

who had used marijuana or hashish is the same as the average (21%), while the proportion reporting use of illicit drugs other than cannabis is lower (2 compared to 6%). The figure for use of inhalants is also lower than the average (6 and 10% respectively), and the same is true for tranquillisers or sedatives without a doctor’s prescription (2 versus 6%) and alcohol together with pills (4 compared to 7%).

% All countries

100 83

Ukraine

84

80 66 60

66

70

53

40

35

39 21

21

20 6

10

6

2

6

7 2

4

0 Any alc. last 12 months

Drunk last 12 months

Ever smoked

Smoked last 30 days

Cannabis lifetime

Any drug but cannabis

Inhalants lifetime

Tranq./ sedatives

Alcohol +pills

The United Kingdom tially larger proportions than the average (38 and 21% respectively), and so is the proportion reporting use of other illicit drugs than cannabis (9 versus 6%). Lifetime use of inhalants is slightly above the average (12 compared to 10%), while the use of tranquillisers or sedatives without a doctor’s prescription is less than half the average (2 versus 6%). Using alcohol in combination with pills is as common in the United Kingdom as the average of all countries (7%).

A vast majority of the students in the United Kingdom had been drinking alcohol during the last 12 months (91%), which is above the average of all ESPAD countries (83%). Also the proportion reporting drunkenness during the same period is higher than the average (68 versus 53%). Lifetime prevalence of smoking cigarettes, however, is lower than average (58 compared to 66%) and this holds true also for the 30 days prevalence (29 versus 35%). Use of marijuana or hashish is reported by substan-

% All countries

100

United Kingdom

91 83

80 68 60

53

66 58 38

35

40

29 21

20

6

9

10

12 6

2

7

7

0 Any alc. last 12 months

218

Drunk last 12 months

Ever smoked

Smoked last 30 days

Cannabis lifetime

Any drug but cannabis

Inhalants lifetime

Tranq./ sedatives

Alcohol +pills

Key results country by country

References

Adalbjarnardottir S. and Blondal K. (1996). Tobaksreykingar og hassneysla Reykviskra unglinga og vidhorf theirra til slikrar neyslu. [Tobacco smoking and cannabis consumpion among Reykjavik adolescents and their attitudes toward such use]. Uppeldi og Menntun, 5, 43–62. Adlaf EM and Ivis FJ. (1996). Structure and relations: The influence of familial factors on adolescent substance use and delinquency. Journal of Child and Adolescent Substance Use, 5, 1–19. Amey CH and Albrecht SL. (1998). Race and ethnic differences in adolescent drug use: The impact of family structure and the quantity and quality of parental interaction. Journal of Drug Issues, 28, 283–98. Arellano C, Chaves EL and Deffenbacher JL. (1998). Alcohol use and academic status among MexicanAmerican and White non-Hispanic adolescents. Adolescence, 33, 751–60. Barnes GM and Farrell MP. (1992). Parental support and control as predictors of adolescent drinking, delinquency and related problem behaviors. Journal of Marriage and the Family, 54, 763–76. Beck KH, Shattuck T, Haynie D, Crump AD and Simons-Morton B. (1999). Associations between awareness, monitoring, enforcement and adolescent involvement with alcohol. Health Education Research, 14, 765–75. Bjarnason T. (1995). Administration mode bias in a school survey on alcohol, tobacco and illicit drug use. Addiction, 90, 555–559. Bjarnason T. (2000). Adolescent Substance Use: A Study in Durkheimian Sociology. Doctoral dissertation, University of Notre Dame. Bjarnason T. and Adalbjarnardottir S. (2000). Anonymity and confidentiality in school surveys on tobacco, alcohol and cannabis use. Journal of Drug Issues, 30, 335–344.

References

Bjarnason T, Andersson B, Choquet M, Elekes Z, Morgan M, Rapinett G. (2003). Alcohol Culture, Family Structure and Adolescent Alcohol Use: Multi-Level Modeling of Frequency of Heavy Drinking Among 15–16 Year Old Students in Eleven European Countries. Journal of Studies on Alcohol, 64, 200–208. Bjarnason T, Davidaviciene AG, Miller P, Nociar A, Pavlakis A, Stergar E. (2003). Family Structure and Adolescent Cigarette Smoking in Eleven European Countries. Addiction, 98, 815–824. Bjarnason T. and Morgan M. (2002). Guidelines for Sampling Procedures in School Surveys on Alcohol and Other Drugs (stencil) The ESPAD project, The Swedish Council for Information on Alcohol and Other Drugs, Stockholm, Sweden. Campanelli P, Dielman T and Shope J. (1987). Validity of adolescents’ self-reports of alcohol use and misuse using a bogus pipeline procedure. Adolescence, 22, 7–22. Costa FM, Jessor R and Turbin MS. (1999). Transition into adolescent problem drinking: The role of psychosocial risk and protective factors. Journal of Studies on Alcohol, 60, 480–90. Cronk CE and Sarvela PD. (1997). Alcohol, tobacco and other drug use among rural/small town and urban youth: A secondary analysis of the Monitoring the Future data set. American Journal of Public Health, 87, 760–4. Currie C, Hurrelmann K, Settertobulte W, Smith R and Todd J (eds). (2000). Health and Health Behaviour among Young People. World Health Organization, Regional Office for Europe, Copenhagen, Denmark. Currie C, Roberts C, Morgan A, Smith R, Settertobulte W, Samdal O and Barnekow Rasmussen W. (2004). Young people’s health in context. World Health Organization, Regional Office for Europe, Copenhagen, Denmark.

219

Dorsselaers S. and Monshower K. (2004). Written communication. Ellickson P, Bui K, Bell R and McGuigan KA. (1998). Does early drug use increase the risk of dropping out of high school? Journal of Drug Issues, 28, 357–80. Ennett ST, Flewelling RL, Lindroth RC and Norton EC. (1997). School and neighborhood characteristics associated with school rates of alcohol, cigarette, and marijuana use. Journal of Health and Social Behavior, 38, 55–71. Foxcroft DR and Lowe G. (1995). Adolescent drinking, smoking, and other substance use involvement: links with perceived family life. Journal of Adolescence, 18, 159–77. Glendinning A, Shucksmith J and Hendry L. (1997). Family life and smoking in adolescence. Social Science and Medicine, 44, 93–101. Grube J, Morgan M and Kearney K. (1989). Using selfgenerated identification codes to match questionnaires in panel studies of adolescent substance use. Addictive Behaviors, 14, 159–171. Hagquist C. (1997). Variations in adolescents’ smoking and alcohol behaviour between Swedish schools: An ecological analysis. Drugs: Education, Prevention and Policy, 4, 139–50. Hakkarainen P. and Metso L. (2003). Narkotikamissbrukets nya generation. Nordisk alkoholoch narkotikatidskrift, 20:333–347. Harrison L. (1997). The Validity of Self-Reported Drug Use in Survey Research: An Overview and Critique of Research Methods. In Harrison, L and Hughes, A (eds.): The Validity of Self-Reported Drug Use: Improving the Accuracy of Survey Estimates. National Institute of Drug Abuse, NIDA Research Monograph 167, Rockville, USA. Hibell B (ed), Adlaf E, Andersson B, Bjarnason T, Delapenha C, Hasbun J, Johnston L and Sathianathan R. (2003). Conducting School Surveys on Drug Abuse. Toolkit Module 3. United Nations Office on Drugs and Crime, Vienna, Austria.

220

Hibell B and Andersson B. (2002a). Project Plan (stencil). The ESPAD project, The Swedish Council for Information on Alcohol and Other Drugs, Stockholm, Sweden. Hibell B and Andersson B. (2002b). Regional Seminars and National Project Plans (stencil). The ESPAD project, The Swedish Council for Information on Alcohol and Other Drugs, Stockholm, Sweden. Hibell B and Andersson B. (2002c). Outline for Country Report 1, Data Collection (stencil). The ESPAD project, The Swedish Council for Information on Alcohol and Other Drugs, Stockholm, Sweden. Hibell B and Andersson B. (2002d). Outline for Country Report 2, Results (stencil). The ESPAD project, The Swedish Council for Information on Alcohol and Other Drugs, Stockholm, Sweden. Hibell B, Anderson B, Ahlström S, Balakieva O, Bjarnason T, Kokkevi A and Morgan M. (2000). The 1999 ESPAD Report. Alcohol and Other Drug Use Among Students in 30 European Countries. The Swedish Council for Information on Alcohol and Other Drugs, Stockholm, Sweden. Hibell B, Andersson B, Balakireva O, Davidaviciene A, Muscat R, Nociar A, Sabroe S and Veresies K. (2000). Do they tell the truth? A methodological study in seven countries about the validity in school surveys. Manuscript. Hibell B, Anderson B, Bjarnason T, Kokkevi A, Morgan M and Narusk A. (1997). The 1995 ESPAD Report. Alcohol and Other Drug Use Among Students in 26 European Countries. The Swedish Council for Information on Alcohol and Other Drugs, Stockholm, Sweden. Hundleby JD and Mercer WG. (1987). Family and friends as social environments and their relationships to young adolescents’ use of alcohol, tobacco and marijuana. Journal of Marriage and the Family, 49, 151–64. Hvitfeldt T, Andersson B and Hibell B. (2004). Skolelevers drogvanor 2004. CAN-rapport 77, Centralförbundet för alkohol- och narkotikaupplysning, Stockholm, Sweden.

References

Irgens-Jensen O. (1991). Changes in the use of drugs among Norwegian youth year by year from 1968 to 1989. British Journal of Addiction, 86, 1449–1458.

Morgan M and Grube JW. (1989). Adolescent cigarette smoking: A developmental analysis of influences. British Journal of Developmental Psychology, 7, 179–89.

Jackson C, Hendriksen L and Dickinson D. (1999). Alcohol specific socialization, parenting behaviors and alcohol use by children. Journal of Studies on Alcohol, 60, 362–7.

Mulhall PF, Stone D and Stone B. (1996). Home alone: Is it a risk factor for middle school youth and drug use? Journal of Drug Education, 26, 39–48.

Johnston L and O’Malley P. (1985). Issues of validity and population coverage in student surveys of drug use. In: Rouse, B, Kozel, N and Richards, L, (eds.): Self-Report Methods of Estimating Drug Use: Meeting Current Challenges to Validity. National Institute on Drug Abuse Research Monograph 57, Washington, USA.

O’Malley P, Bachman J and Johnston L. (1983). Reliability and consistency of self-reports of drug use. International Journal of Addiction, 18:805–824.

Johnston L, Driessen F. and Kokkevi A. (1994). Surveying Student Drug Misuse: A Six-Country Pilot Study. Council of Europe, Strasbourg, France. Johnston L, O’Malley P, Bachman J and Schalenberg J. (2004). Monitoring the Future, National Survey Results on Drug Use, 1975–2003, Volume I: Secondary School Students. National Institute of Health, Bethesda, Maryland, USA. Kokkevi A and Stefanis C. (1991). The epidemiology of licit and illicit substance use among high school students in Greece. American Journal of Public Health, 81, 48–52. Koshkina E and Vyshinsky K. (2004). Written communication. Krohn MD, James L. Massey JL, Skinner WF and Lauer RM. (1983). Social bonding theory and adolescent cigarette smoking: A longitudinal analysis. Journal of Health and Social Behavior, 24, 337–49. Measham F. (1996). The ’big bang’ approach to sessional drinking: Changing patterns of alcohol consumption amongst young people in northwest England. Addiction Research, 4, 283–99. Metso L. (2000). Written communication.

Parker H and Measham F. (1994). Pick’n mix: Changing patterns of illicit drug use amongst 1990’s adolescents. Drugs: Education, Prevention and Policy, 1, 5–13. Pedersen WP. (1990). Adolescents initiating cannabis use: Cultural opposition or poor mental health? Journal of Adolescence, 13, 327–39. Reifman A, Barnes GM, Dintchieff BA, Farrell MP and Uhteg L. (1998). Parental and peer influences on the onset of heavier drinking among adolescents. Journal of Studies on Alcohol, 59, 311–7. Shucksmith J, Glendinning A and Hendry L. (1997). Adolescent drinking behaviour and the role of family life: A Scottish perspective. Journal of Adolescence, 20, 85–101. Skager R and Fisher DG. (1989). Substance use among high school students in relation to school characteristics. Addictive Behaviors, 14, 129– 38. Skretting A. (2000). Ungdom og rusmidler. Rusmiddelsdirektoratet, Oslo, Norway. Skretting A. (2004). Written communication. Stormshak EA, Comeau CA and Shepard SA. (2004). The relative contribution of sibling deviance and peer deviance in the prediction of substance use across middle childhood. Journal of Abnormal Child Psychology, 32, 635–649.

Morgan M. (1997). EMCDDA Instrument Bank: Core Scales, Sources and Guidelines. Available from EMCDDA, Lisbon, Portugal.

References

221

Thomas G, Farrell MP and Barnes GM. (1996). The effects of single-mother and non-resident fathers on delinquency and substance abuse in black and white adolescents. Journal of Marriage and the Family, 58, 884–94. Thorlindsson T, Sigfusdottir ID, Bernburg JG and Halldorsson V. (1998). Vimuefnaneysla Ungs Folks: Umhverfi og Adstaedur [The Social Context of Drug Use among Icelandic Youth]. Reykjavik: Institute for Educational Research.

222

Thorlindsson T and Vilhjalmsson R. (1991). Factors related to cigarette smoking and alcohol use among adolescents. Adolescence, 26, 399–418. Tuinstra J, Groothoff JW, Vand den Heuvel WJA and Post D. (1998). Socio-economic differences in health risk behavior in adolescence: Do they exist? Social Science and Medicine, 47, 67–74.

References

Appendix I

Sampling and data collection in participating countries

This section includes an overview of each country’s sampling and data collection as well as the results of some measures of validity and reliability. The corresponding figures are to be found in tables A–G in the chapter “Methodological considerations” earlier in this report. The presentations are based on each country’s “Country report”, which included standardised descriptions of how the surveys were performed. However, despite the fixed structure, the reports differ somewhat in the level of details. In some of them, the sampling and data collection procedures are described in detail, while in others a briefer and more summarised information is provided. The reason for this might be that some investigators followed the common methodology and therefore thought that there was little to explain. The general

procedure and methodology are described in detail in the chapter “Study design and procedures” earlier in this report. Overall, the sampling and data collection followed the guidelines in the ESPAD project plan. The availability of official statistics and their level of detail differ, however, between countries. Another factor, that influences the methodology, is differences in available funds, which put limits to what is possible to achieve. The reliability and validity are commented in relation to certain measures which also are discussed in the chapter “Methodological considerations”, e.g. inconsistent answering, missing data rates, unwillingness to admit drug use and reported use of the fictitious drug “relevin”.

Austria Dr. Karl Bohrn (Institute for Social and Health Psychology, ISG) and Dr. Alfred Uhl (LudwigBoltzmann-Institut für Suchtforschung, LBI Sucht) in Vienna were responsible for the Austrian study. Austria took part in the ESPAD project for the first time in 2003.

year). 5.7% of the 1987 birth-cohort were not in school when the sample was drawn. 7.9% were in the 8th grade or less (due to repeating classes), 41.4% in 9th grade, 44.8% in 10th grade and 0,1% in 11th grade (Statistik Austria, 2004). Using a representative sample of 9th and 10th grade therefore covers 86.2% of the 1987 birth-cohort.

Population The target population consists of all students in Austrian schools born in 1987. School is compulsory in Austria for 9 school years, thus only around one third (those born between September and December 1987) were obliged to still be in school at the time of the survey. However, a large number of the students continue to secondary education, some of those are enrolled in the “dual system” (school plus vocational training. In vocational schools, students are not enrolled during the whole school year but only blocked for some weeks during the school

Appendix I

Sample and representativeness The survey population was stratified by grade and school-types into subpopulations. For each stratum a sample had to be drawn independently since the average number of students per class varied greatly between school-types. In order to make weighing obsolete for the analyses, the number of classes per strata was defined in a way to achieve the true population proportions in the sample. A complete list of all classes in 9th and 10th grade in Austrian schools and the number of students per class – or

223

even better the students born in 1987 in every class would have been the ideal basis from which to start. Unfortunately such lists were unavailable for the ongoing school year and the tables for the previous school years only provide information on the total student numbers per class but no information on the students belonging to the 1987 birth cohort. The study aimed at 2,400 students of the 87 birth-cohort and as such one would expect that slightly more than 50% of the students in grade 9 and 10 did not belong to the relevant birth cohort. Considering this the total sample size to aim at was laid down at 5,500 students. Since the original design of the study had planned for a smaller total sample of 4,000 students, it was necessary to redefine the survey administration procedures to make them more efficient in order to stay within the budget in spite of the larger sample size of 5,500 students. A way to reduce the administrative work was linking the 9th and 10th grade in school types where both grades were available. In these instances a class from 9th grade was randomly selected and in the same school one of the 10th grade classes was added through a random selection procedure. In the course of the study, in a few situations when the average observed sample size fell below the expected average sample size, an extra class of the same grade in the very same school was added immediately to compensate for the reduction without causing relevant additional administration work. In the instances when more than one class was available selection was done using a random procedure. Since falling below the expected average sample size could only happen when extremely small classes had been selected, adding another equivalent class in the same school was justifiable. To compensate for selected classes that did not participate a second random list of classes per strata was produced. Whenever a class dropped out the next class in the substitution list of the same strata was chosen, to compensate for the loss. 238 classes constituted the initial sample and 93 additional classes were contacted to compensate for failures to include some classes. Of the total sample of classes (original sample plus substitutes) approached (331). • 252 (76.1%) participated in the study • 19 (5.7%) were lost for technical reasons (e.g. a wrong school addresses or a class that was not available in the critical phase) • 42 (12.7%) refused due to understandable timeproblems (e.g. a similar survey just before ESPAD

224

approached them, many exams or excursions in the relevant time period, etc.) • 8 (2.4%) openly refused (e.g. general objection to school surveys on drugs or to the wording of some items) • 10 (3.0%) refused without stating why Thus 79 or 24% of the classes were lost.

Field procedure In the beginning, the schools with selected classes had been selected were contacted to ask for cooperation. If schools or classes rejected participation other schools or classes were selected and contacted in a consecutive order from the substitution list. After agreement to participate the questionnaires were sent by mail to the sampled schools. In each school a self-assigned teacher organised the survey, collected the questionnaires in a way that guaranteed anonymity to the students and returned them to the study team. The teacher stayed in the classroom while the questionnaires were completed and provided assistance if students did not understand questions if necessary and to prevent disturbances. A letter, which was sent to the schools together with the questionnaires, contained instructions to the students and to the teacher. No individual envelopes were used, but the batch of completed questionnaires was put into a large envelope, sealed in front of the students and sent back to the research institute. Most of the data collection happened in the two first weeks of April, the last questionnaires were returned at the end of May. The average age of the 1987 birth cohort was thus around 15,8 years.

Questionnaire and data processing The questionnaire was translated into German in collaboration with the German and the Swiss ESPAD teams. The three German versions in the 3 countries are now almost identical except for some minor country specific adaptations. The Austrian questionnaire was pre-tested with 15 students from different school types and grades. They filled in the questionnaire and were asked to add written comments concerning the wording and comprehensibility of the questions. The only difference between the German and the English version of the questionnaire was one extra question concerning the month of birth, and two additional questions 21 and 22 (21a and 21b). The wording of the questions concerning the alcoholic beverage consumption at the last drinking occasion (10–14) was changed. Since Austrian pupils are not familiar with drink

Appendix I

sizes in centilitres, glass sizes that are common in Austria were introduced to relate to the cultural context. To recalculate the amounts to quantities in the English ESPAD questionnaire the closed questions were changed to an open format. The data were entered in two steps. After the first 500 questionnaires, the data set was checked for mistakes and immediate feedback was given to the person entering the data. Random control of data entries was also done. Questionnaires with more than 170 missing values or a missing year of birth were discarded. The number of discarded questionnaires was 13 in grade 9 and 38 in grade 10 (about 1%). For each person entering data some randomly chosen data sets were compared systematically to the original questionnaires. The data quality was very high (less than 1%). However, the gender distribution in the data set is uneven, with 56% boys and 44% girls. This was due to an uneven sex distribution in the 10th grade, while it was almost perfect in the ninth grade. Due to this discrepancy the data should have been weighted. The average time to fill out the questionnaire was 41 minutes.

School and student co-operation A majority of the schools and classes were positive towards the study, and the contact could easily be established and maintained. However, as mentioned earlier, 91 classes had to be added due to various reasons. There were almost no total individual refusals to fill in the questionnaire (only 4 students = 0.07%). Out of the 252 classes participating in the survey only 10 reported problems in understanding the questionnaire. These comments mainly referred to certain substances such as tranquillisers, anabolic steroids and alcopops that the students were unfamiliar with. According to the classroom reports 76% of the classes reported no disturbances. The most common disturbances were loud comments, and in many cases questions related to the survey (see above). If problems were reported, they mostly concerned the fact that the questionnaire seemed a bit too long (with similar questions) for students with low reading skills or there were difficulties in comprehending some questions. These kinds of comments came from a minority (20 classes) and mainly from vocational and polytechnic school classes. In 95% of the classes, the organising teacher reported that a majority of the students were interested in the study, and that almost all of them seemed to work seriously. The response rate was 89%, i.e., the number

Appendix I

of students in the classes that participated was 6,187 and the number of students who were in school on the day that the survey was conducted and completed the questionnaire appropriately was 5,503.

Reliability and validity The inconsistency rates between equivalent questions in a single administration were rather low. The highest rates of inconsistency were observed for “having been drunk” the use of inhalants (about 5%), the use of amphetamines, the use of alcohol together with prescription drugs (4%), ever smoked (3%) and cannabis use (3%). For all other variables the inconsistency rate were around 1%. Missing data rates on lifetime questions were overall low; the highest rates were observed for any alcohol (4%) and for “having been drunk” (2%). The latter variable had an increasing proportion of missing data concerning the 12 months window (3%) and the 30 days window (10%). The corresponding rates on “any alcohol use” were 4% and 3% respectively. The rates of inconsistent answering between lifetime, 12 months and 30 days use, were not very high for any of the included variables, 3% in relation to alcohol, 1 % in relation to cannabis and less than 1% in relation to inhalant use. 7% of the students said that they would “definitely not” have admitted the use of cannabis, while the rate concerning heroin was 11%. Markedly more boys than girls claimed to be reluctant to admit such use, for cannabis it was 10% vs. 4%, and for heroin 16% vs. 5%. The proportion answering, “I have already said I have used it” was 20%, which was very close to self-reported lifetime prevalence (21%). Use of the fictitious drug “Relevin” was reported by 1% of the students, while 11% thought that they had heard of it.

Methodological considerations School in Austria is compulsory for nine school years only, but only a minority (1/3) of those born in 1987 were obliged to be in school since enrolment is related to the time of the year the students were born. However, it was estimated that about 94.3% of the students in this age group were actually enrolled in school. This implies that the Austrian survey is representative for the students born in 1987 still in school. The random sampling procedure per cohort and a sample-size per cohort representing the estimated proportion of the cohort in the population guarantees that the total sample is close to a representative

225

sample of all members of the 1987 cohort who are still in school. However, the sample is not perfectly related to class size, although an attempt to correct for the un-proportional distribution of small classes was made by the sampling of one extra class if the sampled class was smaller than average. In Austria the technique with random replacement of refusing or non-responding schools was adopted to avoid a loss of classes. However, a large number of classes did not participate in the end, which lead to a loss of 24%. The student willingness to cooperate was on the other hand good with only 4 students refusing to participate. The questionnaire was almost identical with the common ESPAD version, but two own questions were inserted into the main body of it (instead of putting them at the end). In addition the question on the last occasion of alcohol consumption was changed into an open format. It was assumed that it would be too difficult for the students to adapt to

response categories with unfamiliar glass sizes and alcohol content. This deviance in relation to the results of other ESPAD countries makes it necessary to put the Austrian figures for these variables under the bottom line in the tables, since they are not directly comparable with the results from other ESPAD countries. The methodological measures such as inconsistency rates between two questions in a single administration, missing data rates and inconsistencies between lifetime, 12 months and 30 days prevalence were overall rather low. Other details, such as a loss of classes when data was collected, and an uneven sex distribution that was not weighted for, calls for a certain awareness when analysing the data. However, apart from these facts the data quality of the Austrian survey seems to be satisfactory and the survey has been completed without any major problems.

Belgium Belgium has four language areas: The Dutch, French and German speaking areas as well as one bilingual. The latter is the capital city Brussels that includes Dutch and French speaking people. The Belgian ESPAD 03 study included two separate samples and data collections. Professor Caroline Andries and Dr. Patrick Lambrecht at the Department of Developmental and Life Span Psychology at the Vrije Universiteit Brussel were the principal researchers for the Dutch speaking areas whereas in the French speaking part of Belgium Professor Danielle Piette from the School of Public Health at the Université Libre de Bruxelles was the principal researcher. The Belgian study was co-ordinated by Professor Andries and Dr. Lambrecht. It was the first time that Belgium participated in the ESPAD project.

Population The population consisted of all students born in 1987 going to regular schools in the Dutch and French speaking areas. Since the German speaking part of Belgium only consists of 0.7% of the population, which accounts for 35 students in a national sample, it was excluded for pragmatic reasons. Of the students born in 1987 56% lived in the Flemish community and 44% in the French speaking community.

226

Of all young people born in 1987 99% were enrolled in school at the time of the data collection.

Sample and representativeness Two separate samples were drawn, one in the Flemish, in which grades 9 and 10 participated, and another among French speaking students in which grades 8, 9 and 10 were included. Both samples were stratified two step samples. Earlier school surveys have demonstrated that approximately one third of sampled schools would be expected to participate, a factor which was taken into consideration when drawing both samples. The first step in the Flemish (Dutch speaking) sample was a systematic sample of 184 schools ( of which four were never asked to participate) in four geographical areas proportional to school size. This was about three times as many schools than was calculated to be necessary to obtain the expected proportion of Flemish students (about 60 schools). Each of the schools that agreed to participate was asked to provide a list of the different programs that the school organised. These lists were used to randomly sample classes (clusters/programmes of 20– 30 students). In the 82 schools that agreed to take part in the survey 212 classes (clusters) were sampled.

Appendix I

In the French speaking sample the first step was a random sample of 100 schools stratified by geographical area. Since it was expected that many schools would refuse to participate, two “reserve samples” with another 100 schools each were sampled in the same way. In sampled schools that accepted to participate the headmaster sent a list of all grade 8, 9 and 10 classes. The second step was a random sample of classes from these lists. The Belgium sample is said to be self-weighted and representative for all 1987 born students in participating grades, which include 95% of all students born in 1987.

Field procedure Headmasters in sampled schools were contacted and asked to participate in the study. Headmasters that accepted to participate were asked to send a list of all classes in participating grades as well as appointing a “school co-ordinator”. In the French speaking community the questionnaires as well as all relevant material were sent to the school co-ordinators, who were responsible for giving the relevant material and information to the teacher(s) to enable them to conduct the data collection. Data collection in the Flemish schools was conducted by 15 trained research assistants. Before data collection, students were informed in line with the ESPAD protocol. The students participated in the survey under the same conditions as a written test. When the questionnaires were completed the French speaking students put their questionnaires in individual envelopes while the research assistants in the Dutch speaking schools collected the questionnaires and put them all in one large class envelope. The Flemish data were collected between March and May and the Walloon data in April and May, which gave an average age of 15.8 years. The average time to complete the questionnaire was 40 minutes in the Flemish schools and 50 minutes in the French speaking schools.

Questionnaire and data processing All core questions were included in the Belgian survey. In addition, to this core segment the Dutch questionnaire contained modules A (Integration) and C (Psycho-social measures) as well as three extra questions that amounted to an extra 35 variables. The French version of the questionnaire included module A (Integration) as well as all but one of the questions in module B (Mainstream). It also contained another 23 questions that amounted to 120 variables.

Appendix I

Since Belgium borders the Netherlands the category “coffee shop” was added to the question on where the students think they can buy cannabis (Q33). This is further commented on in the result section of this report. The Flemish questionnaire was pilot tested on 38 students in four classes. The test resulted in some minor changes in the introduction as well as in the instructions of the questionnaire. The French speaking questionnaire was pilot tested on 32 students in two classes. The data entry was checked. In the Flemish part this was done by re-entering every 20th questionnaire, which showed that less than 0.1% of mistakes were made during the data entry process. In Walloonia, the quality check was done by a research assistant that regularly observed the data entry. Data were not weighted.

School and student co-operation Prior to any ESPAD data collection it was already apparent that there were complaints from secondary schools in relation to number of requests to participate in such surveys. Hence, the researchers expected a large non-response and to reach the ESPAD goal of at least 2,400 participating students there was “heavy oversampling” of the number of schools. In the Dutch speaking part 82 out of 180 sampled schools agreed to participate in the survey. In the French speaking part the corresponding number was 59 from a sample of 100 schools (and two “reserve groups” each of which contained 100 schools). Of the 141 schools that agreed to take part in the survey only 131 actually did so. In these 141 schools a total of 442 classes were supposed to participate. At the end of the field procedure data were available from 390 classes. In addition to this it should be acknowledged that in 7 Dutch and 10 French speaking schools a selected class that did not participate was replaced by another “similar” class. The major reason why schools did not take part in the study had nothing to do with the fact that it was a survey about alcohol and drugs but rather that Belgian schools are asked to participate in too many surveys and as a result do not accede to all requests. In the Dutch speaking community all schools were asked to fill in a form that contained information about the number of students in the grades included in the survey. The analysis of these forms

227

does not indicate any major differences between participating and non-participating schools. Seven Dutch speaking students refused to participate. (The corresponding figure is not available from the French speaking community.) In the scrutinising process 13 questionnaires (0.5%) were excluded in Flanders (including questionnaires from participating students not born in 1987). In the French speaking area 22 questionnaires from participating students born in 1987 were excluded. The response rate, measured as the proportion of participating students in participating classes, was 93% in Flanders and 74% in Walloonia, which gave a country average of 81%. All students in sampled classes answered the questionnaire. However, only data from students born in 1987 were included in the ESPAD report. Information from the data collection leaders was only available from the Flemish area. About four out of 10 (41%) reported that there were no disturbances during completion of the questionnaires while 45% reported that this happened in a few cases by a few students. Of all survey leaders about one out of three (34%) answered “other comments” and 25% reported giggles or eye makings to classmates. A large majority of the survey leaders (92%) reported that “all”, “nearly all” or “a majority” of the students were interested in the survey (80% answered “all” or “nearly all”). About the same proportion answered that they found that the students worked seriously (93 and 78% respectively). The over-all assessment of student co-operation was judged to be “rather good” and that student comprehension was satisfactory.

Reliability and validity Reliability measured by inconsistency rates between two questions in a single administration was highest for the variables “been drunk” and “cannabis” (6% each). It was lower for cigarettes, inhalants and tranquillisers and sedatives (3–4%) and even lower for other illicit drugs and anabolic steroids (1% each). Missing data rates were low for different kinds of drugs (1–2%). It was also low for core and module questions (2–3%) but a bit higher for own questions (7%) as they were situated at the end of the questionnaire and some respondents ran out of time. For the questionnaire as a whole 3% of the questions were unanswered. The inconsistency rates between life time, 12 months and last 30 days prevalence rates were a

228

little higher for the two alcohol variables (2–4%) than for inhalants and cannabis (0–1%). For cannabis 5% of the students answered “definitely not” on the question “If you had used marihuana or hashish, do you think that you would have said so in this questionnaire?” The corresponding figure for heroin was a bit higher (8%). On this “willingness question” 22% answered that they had already said that they had used cannabis, which was lower than the prevalence figure (32%). Eight percent answered that they had heard about the dummy drug NTSC/BKR (which was used instead of relevin). However, only 0.3% said that they had used it.

Methodological considerations From earlier experiences it was well known to the Belgian ESPAD researchers that many schools are asked too often to participate in surveys. Hence, it was expected that many schools would refuse to participate in the ESPAD study. To “compensate” for this the sample in the Flemish community included as many as 184 schools, while the researchers in the French speaking community choose to have two reserve samples, each of which was as large as the original sample of 100 schools as the solution to this particular problem. Of the Flemish schools 82 agreed to participate (46%) and among the French speaking 59. Of these 141 schools data were finally collected for 131. In the 141 schools 442 classes were sampled to participate, of which 390 did so in the end. In an ESPAD context the proportion of non-participating schools was high. However, it should be noted that a comparison between participating and non-participating schools in the Dutch speaking area did not indicate any important differences. Unfortunately, this type of information was not available from the French speaking part. The low number of participating schools is “normal” for the Belgian situation. The major reason has to do with the autonomy of local school heads and with the fact that Belgian schools are overloaded with school surveys. It is not related to the content of the survey. Analysis from earlier school surveys indicate that it is unlikely that participating and refusing schools differ in a systematic way. In combination with what is mentioned above, this indicates that the large number of non-participating schools should not jeopardise the possibility for comparisons with ESPAD data from other countries. However, some uncertainty still remains.

Appendix I

The proportion of participating students in participating classes was 81% in the country as a whole. The response rate was higher in the Flemish schools (93%) than in the French speaking community (74%). This relatively low figure is among the lowest in the whole ESPAD study. Few students refused to participate and relatively few questionnaires were rejected. On the other hand, there were relatively more survey leaders in the Flemish schools that reported some kind of disturbances during the completion of the questionnaire than among survey leaders in other countries. A plausible explanation for this from the Flemish ESPAD researcher is that data in the Flemish areas were collected by research assistants. In the training they received, they were informed that they should note all disturbances, which made them very observant. It was also commented that research assistants, compared with teachers that are used to have “normal disturbances” in the classrooms, have a lower “tolerance level”. Hence it seems reasonable to assume that the disturbances during the data collection were not more serious in Flanders than

in most other parts of Europe. Such a conclusion is supported by the fact that a very large majority of the survey leaders reported that the students were interested and worked seriously. No information from the classroom reports is available from the French speaking schools. Even though there are sufficient reasons to believe that the situation is similar in this community as in the Dutch speaking areas, this cannot be taken for granted. The reliability and validity measures do not indicate any major problems. In summary, a large proportion of schools and classes refused to participate and that some information was not available from the French speaking schools, would suggest that the uncertainty might be higher in Belgium than in most other ESPAD countries. However, it seems reasonable to assume that the methodological complications are not sufficient to cause major problems with comparisons with other ESPAD countries. On the other hand, some caution is recommended.

Bulgaria Anina Chileva, psychologist at the National Centre of Public Health in Sofia co-ordinated the Bulgarian ESPAD survey. Bulgaria also participated in the 1999 ESPAD survey.

Population In Bulgaria children start going to school at 6 or 7, depending on the parents’ decision. Thus, students born in 1987 are to be found in 9th or 10th grade as well as in 1st and 2 nd grade in secondary technical and vocational schools. There was s no information available with respect to the proportion of students born in 1987 found in different grades. School attendance is compulsory in Bulgaria until grade 8 of secondary education. It was estimated that approximately 72% of the 1987 birth cohort were in school in Bulgaria in May 2003.

Sample and representativeness Data from the Ministry of Education and The National Institute of Education revealed that students born in 1987 were taught in 1041 schools, of which 17 were high schools (gymnasiums), 94 specialised language high schools (specialised gymnasiums),

Appendix I

463 were secondary general education schools, 334 secondary technical schools, 118 secondary vocational schools, and 16 secondary sport schools. Reliable information on class size or lists was not available. Thus, a two stage random sample of schools and classes was drawn. The sample of schools was drawn with a probability related to size, but classes were drawn with equal probability using the SPSS random number generator. To generate a sufficient sample with students born in 1987 a total number of 278 classes including 6,547 students was drawn. The net sample consisted of 2,739 students born in 1987.

Field procedure A recommendation letter from the Ministry of Education served both as permission for the conduct of the survey in school and also ensured the support of the school administration. It was decided that people not affiliated to the school, in order to better guard students’ anonymity and thus facilitate the collection of quality data, should conduct the survey. One of the best-operating networks in the coun-

229

try is the Bulgarian Public Opinion Centre, with specially trained supervisors in all 28 regional centres of Bulgaria. Each has a local network of research assistants with a vast experience. The supervisors were provided with all necessary information and material. In addition they were supported via telephone link throughout the data collection period. The supervisors organised a half-day training workshop for the research assistants to acquaint them with the instructions, and to provide them with support letters, questionnaires and envelopes. A school staff member who also assisted in the completion of the classroom report following which he/she left the classroom introduced the research assistants to the class. The class answered the questionnaire under the same conditions as required for a written test. The study was conducted during the period May 15–26, which gave an average age of 15.9 years.

Questionnaire and data processing All core questions (except cider/alcopops and GHB) and the modules A–D were included in the Bulgarian version of the questionnaire. The questions that were omitted involved those that were based on substances not available in Bulgaria. No country specific question was added. The new parts of the ESPAD questionnaire were translated into Bulgarian by two independent translators and both versions were used for the Bulgarian edition. Later on, another specialist did back translation into English, the two English versions were compared and the final version was printed. There was no time for pre-testing of the questionnaire, but in the main it was the same as that used in the 1999 study. Data verification was augmented by direct comparison with source documents and by logical crosschecking. The SPSS DE was used for data input and SPSS v. 11.5 for analysis.

School and student co-operation Co-operation with school staff as well as with the students was very good. Only one school director from a private school refused to participate. This school was not replaced. No class refused to participate. 11 students from 5 classes refused to participate at the beginning of the survey and left their questionnaires blank. 22 questionnaires were excluded due to inconsistent answering. The data collection leaders reported disturbances during completion of the survey from about half of the students in 1.8% of the classes, and from more

230

than half of the students in 1.1% of them. Most common disturbances were “giggles or eye makings to the classmates” (30%). Loud comments were observed in 14% and other comments in 9% of the classes. Loud comments were mostly connected with unknown illicit drugs and with some jokes about alcohol and drug use. Other comments reported were connected with the meaning of different questions, and with some questions on the process of filling in the questionnaire. However, in 56% of the cases there were no disturbances during completion. Moreover, the data collection leaders reported that all of the students were interested in the survey in 59% of the classes and that the students worked seriously in 66% of them. As in the 1999 study, two common problems were reported. The first related to the fact that some of the students had difficulties in understanding some of the questions. These difficulties appeared in classes with students with lower ability, and in classes with students from minority groups who had some language problems. The second problem reported in a few cases was that the questionnaire was too long and thus some students lost interest by the end of the session. Despite these problems the main impression was that student comprehension was good. In most cases students were interested to know of the outcome at the end of the research process. The response rate was 85% and the average time to complete the questionnaire was 51 minutes.

Reliability and validity The inconsistency rate between two questions in a single administration was highest in relation to drunkenness (12%), but also somewhat high for smoking cigarettes (8%) and cannabis use (7%). For inhalants use it was lower (3%) and for all other illegal substances or behaviours it was 2% or less. Inconsistency rates were generally somewhat higher amongst the boys than amongst the girls. The missing data rates were overall rather low. The highest rates were observed in relation to alcohol. However, the percentage for lifetime use of alcohol was somewhat higher (6%) than for 12 months and 30 days use (5% both). For the variable “been drunk” the reverse pattern was observed – the missing data rates for lifetime were 4%, while for 12 months and 30 days it was 5 and 6% respectively. For smoking cigarettes it was 2% for lifetime and 1% for 30 days use. For both cannabis and inhalants the missing data rates were lowest in

Appendix I

relation to lifetime use (1% both), than for 12 months or 30 days prevalence rates (4% for each). For all other variables the missing data rates were 2% or less. The average number of unanswered core questions was 18 (6%) but lower for module questions (3%). The total figure was 5%. No gender differences were observed. The inconsistency rates between lifetime, 12 months and 30 days prevalence were highest for alcohol (10%) and for drunkenness (9%) for all respondents. The rates for cannabis use were less than 1.5% and for the use of inhalants less than 1%. The gender pattern shows that the girls gave somewhat more inconsistent answers to questions regarding drunkenness and cannabis use than the boys. About 8% of the students answered that they would definitely not admit use of cannabis and a more or less same percentage (9%) claimed that they would not admit heroin use. There was a clear gender difference both in relation to cannabis (10 vs 5%) and heroin (13 vs 5%). The proportion of respondents who answered that they already said that they had used cannabis was only slightly higher than the lifetime prevalence figure (23 vs 21%). The proportion reporting use of the dummy drug relevin was very low (1%). However, 10% of the students claimed that they had heard of such a drug.

Methodological considerations The survey in Bulgaria seems to have functioned very well without any major difficulties. The sample was carefully drawn from all types of schools where students born in 1987 were taught. However, the sample only included students still enrolled in some form of schooling (72%), which implies that the results cannot be generalised to the

whole birth cohort. The sampling of schools was done randomly with a probability proportional to school size while the selection of classes was simply random, i.e. each class had the same probability of being selected regardless of size. It should be noted that the sample resulted in a better representation of the age cohort in question than in 1999, as the sample this time covered all grades (4) where students born in 1987 were taught. The cooperation with the schools was good with then result that only one school refused to participate but no class did so. Students’ cooperation was also good and the majority of the students expressed a positive attitude. Only a small number of questionnaires were excluded as a consequence of invalid data. The reliability and validity measures are indicative of a rather good quality data set. The inconsistency rate between two questions in a single administration was, however a bit high for questions on drunkenness (12%). A suggested explanation that emerged in the country report from Bulgaria was that there is a difference in the Bulgarian language between “being drunk” and “getting drunk”. The former refers to a more unconscious state than the latter, and this in part may provide a reason for the high rate. Other methodological measures suggest a relatively good quality data set. The missing data rate was rather low. It is difficult to explain, however, why the rate of missing data was higher for the lifetime use of alcohol, than on the 12 months and 30 days prevalence on the same variable. Usually it is the opposite, which was the case for the other variables analysed. A relatively high (5%) number of unanswered questions should be noted. The overall impression, however, still remains; the Bulgarian study was well designed and that data provided most probably are both reliable and valid.

Croatia Dr. Marina Kuzman, Social Medicine Department at the National Institute of Public Health, Zagreb was responsible for the Croatian survey. Croatia also participated in the 1995 and 1999 ESPAD studies.

Population The population consists of all students born in 1987 and enrolled in the first and second grades of sec-

Appendix I

ondary education in Croatia. According to the Ministry of Education approximately 95% of the age cohort born in 1987 were in school in March 2003. The population was split between two grades, with approximately division of 70% in the first and 30% in the second grade. Croatia is divided in 21 counties. In each there are schools of every type, except for the absence of secondary schools on small islands and in sparsely populated areas.

231

Sample and representativeness The survey was conducted in the whole country. There are three types of secondary education in Croatia: Gymnasiums, Vocational 4-year and Industrial/Craft 3-year. Both grades (1 and 2) in each type of education were included in the sampling frame. The administrative division of geographical areas was disregarded in relation to sampling. For both grades three lists of classes were made and according to the average number of students in each class, using a random sampling method, the number of classes (238) sufficient to cover up to approximately 3,200 students was selected. The sample was a simple random sample of classes where each class had the same probability to be chosen. According to the number of responding students born in 1987, it was assumed that 97% of the cohort was covered. The male/female ratio was the same as the gender ratio of the whole generation.

Field procedure After that the sample was drawn all schools were contacted by telephone to inform about the survey and to ask them to participate. All of them agreed. The questionnaires were packed in paper boxes together with a letter of approval from the Ministry of Education and other informational material and were sent to the schools. The boxes were pre-coded as well as classroom reports, but not the questionnaires. School counsellors or class-masters collected the data. After filling-in the questionnaires, students were instructed to put them into envelopes and to seal them and hand them to the school counsellor. He/she completed the classroom report and put everything together in the same paper-box and returned it to the research institute. Data was collected in the period 1–15 April 2003, which gives an average age of 15.8 years. The response rate was 90%.

Questionnaire and data processing The previous ESPAD questionnaire was used as a base; a professional interpreter translated only changes or amendments. A person at the research institute did back-translation. Cider was excluded as it was considered inappropriate for the Croatian students. As in earlier studies, questions were added on parental behaviour regarding smoking, drinking etc. The questionnaire was not pre-tested. However, the questions on alcohol consumption on last drinking occasion caused difficulties among the students. It seemed to provoke over-estimation

232

even among students who otherwise take the study seriously. The packages from two classes in one school arrived very late and as it was uncertain if they had collected the data during the recommended period they were excluded from the data set. During the coding process year of birth and gender were checked. At this stage two to three questionnaires from each class were randomly selected and checked whether they were properly filled in. During data processing 16 questionnaires were excluded, as they were almost empty or obviously poorly filled. Data was not weighted. The Access software was used for data entry.

School and student co-operation All schools and classes expressed willingness to participate in the study. None of the students refused to participate. According to the classroom reports the student co-operation was very good. Only 5% of the classes reported any disturbances. In cases of disturbance all kinds were reported such as loud comments, giggles or eye makings or other kind of comments. Half of the survey leaders reported that all students were interested and worked seriously and another half reported that nearly all did so. The average time to complete the survey was 45 minutes.

Reliability and validity The inconsistency rate between two questions in a single administration was highest for “been drunk” (7%), use of inhalants (4%), cigarette smoking, use of cannabis, tranquillisers or alcohol together with pills (2% each). For other substances the corresponding figures were around 1% or less. The proportion of unanswered questions was generally very low, 1% or less, but the missing data rates increase in relation to prevalence period, i.e. the 30 days prevalence questions have a slightly higher rate of missing data than the lifetime prevalence. The average number of unanswered questions was 1% for core questions, but somewhat higher for module questions (2%) and own questions (4%). The mean value for all questions was 1%. The rates of inconsistency between answers to lifetime, 12 months and 30 days prevalence questions were generally low on any of the variables concerned. The highest value was observed in relation to alcohol (about 3%), but for cannabis use it was 1% and for use of inhalants less than 0.5%. The proportion that answered that they would not admit cannabis use was 12%. The proportion

Appendix I

among boys was significantly higher (19%) than among girls (6%). The same is true in relation to heroin (15%), where the proportion among boys was 24% and among girls 7%. The proportion that answered on this question that they already said they had used cannabis was only somewhat lower (19%) than the actual prevalence figure (22%).

Methodological considerations The sample for the Croatian study was well designed and included both grade 1 and 2 in all three types of secondary education. This was an important improvement compared to the earlier studies, which were restricted to only one grade. The two grades were estimated to cover 97% of the age cohort. Since it was only possible to draw the classes as a simple random sample where each class has the same probability to be drawn, small classes might be over-represented in the sample. The co-operation with the schools was very good. The two (out of 238) non-participating classes were

those, which were sent to the research institute very late in the process and therefore excluded from the analysis. The proportion of unanswered questions was low. No present student refused to participate and the number of out-sorted questionnaires because of bad data was low. Overall, as the classroom reports indicate, the study seems to have functioned very well in Croatia. As for the methodological measures such as inconsistency rates and unanswered questions the quality of the study should be considered very good. A somewhat high percentage of students not willing to admit cannabis or heroin use is more obvious among boys than among girls. However, the proportion that on this “honesty” question answered that they already had said so in the questionnaire was not too different from the proportion that reported use in other parts of the questionnaire. Generally, the survey seems to have functioned well and can be assumed providing reliable and valid data.

Cyprus Dr Kyriacos Veresies at the organisation KENTHEA and Dr Andreas Pavlakis at the Ministry of Education, Cyprus were responsible for the Cyprian study. Cyprus also participated in the 1995 and the 1999 studies.

Population The target population consists of all students who turned 16 in the calendar year of 2003 and who were registered in public schools (lyceums, vocational schools, hotel schools), at the time when the survey was conducted. The students born in 1987 were to be found in two grades 1st and 2nd in lyceum since they enter the school system in relation to the month of birth. In grade 1 about 90% was born in 1987 and in grade 2 about 67%, some of which were in the lower grade were repeaters. Of the total birth cohort 74% were to be found in secondary in public schools. Of the students in those two grades 81% were born in 1987.

Sample and representativeness The sample of classes was drawn from grades 1–2 in secondary public schools. The sample was planned to include 108 classes from lyceum, 24 from technical schools and 3 from hotel schools. However, due to

Appendix I

a smaller class size than average in technical schools, these were over-sampled (34 in total) to compensate for this. Generally, the average class size is 25 students. However, the class size in hotel schools is smaller (16). The sample was drawn as a simple random sample regardless of class size. As the average class size in lyceum is even and the technical schools were over sampled the sample is representative for these grades. However, the hotel schools are underrepresented in number of students.

Field procedure In the autumn 2002 an official letter was sent to the Ministry of Education asking for permission for the administration of the questionnaires in schools. The Ministry subsequently communicated the approval of the study implementation to school directors regarding the study and their expected role in it. Research assistants contacted the school directors by telephone prior to their visit to the schools in order to arrange appointments for the administration of the questionnaires. During the administration period, and mainly the first two weeks, several regular meetings were held involving research assistants and the core research team to

233

discuss issues related to the implementation of the study. The survey was scheduled for two lecture periods, i.e. 90 minutes. A research assistant supervised the data collection. No teacher was present in the classroom during administration. The filled-in questionnaires were placed in special folders in a way that safeguarded the anonymity of the respondents. The data collection took place during March–April 2003, which gives an average age of 15,8 years.

Questionnaire and data processing The questionnaire contained almost all ESPAD questions, except the question on cider and the question on 12 month and 30 days prevalence of illicit drug use. In addition the module C, (psychosocial) was included. All questionnaires were checked prior to data entry. All invalid questionnaires were discarded based on a number of criteria such as: no date of birth specified, too many inconsistencies, strong indications of open or covered refusals in disclosing personal information (e.g. too many unanswered questions. Exaggerated replies, systematic selection of specific replies, written comments on the questionnaire, etc.), other reasons (e.g. students failure to understand large sections of the questionnaire, incomplete questionnaires, etc.)

School and student co-operation The cooperation with the schools was very good. No major disturbances were reported and the students seemed interested and co-operative. Based on the classroom reports 92% of the classes reported no disturbances. In classes where disturbances were reported those were equally divided in giggles and comments or other kind of comments. An absolute majority of the students were reported to be interested and working seriously. The average time to complete the survey was 57 minutes.

Reliability and validity The inconsistency rates between two questions in a single administration were overall rather low. The highest was observed in relation to “been drunk”

234

(5%) and smoking (4%). For cannabis it was 2%, inhalants 7% and tranquillisers or sedatives 4%. Other drugs around 1%. The rates of inconsistent answering were somewhat high for any alcohol (10%) but lower for been drunk (4%). For inhalants it was 2% and for cannabis 1%. The missing data rates were overall low. For lifetime use of alcohol it was 2%, for 12 months it was unchanged and for 30 days prevalence only slightly higher (3%). For smoking it was 0.1%, for lifetime prevalence of being drunk 1% and for 30 days prevalence only somewhat higher (2%). For other illicit drugs it was overall very low (0.2%). The average number of unanswered questions was not calculated. The willingness to admit drug use was relatively good. About 6% would definitely not admit neither cannabis nor heroin use. The number of students that answered that they already had said they used cannabis was only slightly higher (6%) than the proportion, which in the question reported such use (4%). Only 0.3% reported use of the fictitious drug “relevin”, while 10% thought that they had heard of it.

Methodological considerations Overall the Cyprian survey seems to have been well functioning. No problems were reported and the results on the methodological measures do not flag for any important reliability problem. It has, however, been very difficult to establish the representativity of the sample. Since the average class size of the main part of the sample (lyceums) is very even and the technical schools were over-sampled to correct for smaller class size the main part of the sample seems to be satisfactory representative. The hotel schools, however, are under represented and would have needed to be weighted. In relation to the total sample they would however still have limited influence on the results. Other methodological measures such as inconsistency rates and missing data rates point at a quite good data quality. There are no reasons not to consider the Cyprian study valid and reliable.

Appendix I

The Czech Republic Dr. Ladislav Csémy at the Prague Psychiatric Centre was responsible for the survey in the Czech Republic. The Czech Republic also participated in the 1995 and 1999 ESPAD surveys.

Population The target population consists of students in the level of secondary education born in 1987. Approximately 95% of the pupils in elementary education continue to studies in secondary education. However, elementary education in the Czech Republic starts with those who achieved the age of six before the 1st of September each year. This means that of the 1987 born cohort, only 65–70% have entered secondary school. As a complementary study within the project pupils born in 1985 were also surveyed. Most of the students born in 1987 were to be found in gymnasium – grammar schools (students who are expected to continue their studies at university), secondary with leaving exams (students are prepared for employment, but may also enter at university), and vocational schools (qualified skilled workers). Available information related to the school year 2001/2002 gave the number of 346 gymnasiums, 813 secondary schools with leaving exams and 570 vocational schools. The total numbers of students were 21,415.

Sample and representativeness The sample is a multistage random stratified sample, including selection of schools by region (14 regions) and school type (3 types). The required number of classes from respective type of school in a given region was set up according to information about the distribution of students born in 1987 in the regions and the proportion of students in the three types of education in each region. The schools were chosen randomly from the list by using the SPSS program for random selection of cases. To enhance the probability for larger schools to be drawn, each school with 3 or more classes was represented twice in the list (the majority of secondary schools have 2–4 classes in each grade). A total number of schools (and classes) in the sample was 180, resulting 5 048 students.

Field procedure As in previous ESPAD surveys in the Czech Republic, a professional company specialised in survey research for the health care sector (INRES-

Appendix I

SONES) undertook the data collection. The headmaster in each school received two informational letters asking them for co-operation, one of which was signed by the director of the National Drug Commission, and the second was a supporting letter from the Ministry of Education. Of the existing network of interviewers at the data collection company 104 persons participated in the data collection. The teachers were allowed to be present, but the data collection procedure was fully in the hands of the research assistants. Data was collected during April 3 through 16 except in 2 schools, which were allowed to do their data collection in May/June. If those classes are disregarded the mean age of the Czech participants was 15.7 years.

Questionnaire and data processing The questionnaire consisted of all ESPAD core questions and the larger part of the psychosocial module (25 variables) as well as own questions (36 variables). The questionnaire was not piloted, mainly due to limited time and economical resources. However, only a minor part of the questionnaire was new compared to the 1999 survey, and these parts were translated under supervision of a professional interpreter. Of the total number of 3,195 questionnaires answered by students in the target age group 23 were discarded because of apparently invalid data or because of a large number of missing values. A validation of the data input by double-checking 303 questionnaires revealed a very small error frequency (0.15%).

School and student co-operation The data collection was carefully prepared and was functioning without any problem. None of the selected schools refused to co-operate, to a large extent probably because of personal interventions and phone calls. No present student refused to participate. In two thirds of the classrooms no disturbance was reported and in another third only few students were reported to have disturbed the data collection in class. Moreover, according to the data collection leaders a vast majority of schools participated in the study with interest and 92% of the classroom reports indicated that “all” or “nearly all” students were interested in the survey. In the classes where disturbances were noted it was mainly a matter of

235

giggles or eye makings to the classmates. It was stated that the majority (88%) also worked seriously with the survey. However, many students thought the questionnaire was too long. The average time to complete the questionnaire was 47 minutes.

Reliability and validity The inconsistency rate between two questions in a single administration is highest for non-prescribed tranquillisers or sedatives (5%) and the use of inhalants, cannabis or “been drunk” (3%). For smoking, or the use of LSD and alcohol together with pills the inconsistency rate was 2%. The corresponding values for other drug use was lower (less than 1%). Missing data rates on drug related questions were low, the highest were found in relation to any alcohol use and “been drunk” (2%). The average number of unanswered questions was 2% (core questions 1%, module and own questions 4%). The rate of inconsistent answering on lifetime, 12 months and 30 days questions respectively was low both for alcohol (any alcohol 2%, been drunk 1%) and other drug use (cannabis and inhalants less than 1%). The proportion that answered to the “honesty question” that they “definitely not” should have admitted cannabis use was rather low (3%), but higher for boys (5%) than for girls (2%). The corresponding value for heroin use was higher (7%) but with the same gender difference (10 vs. 4%). The proportion that on this question answered, “I already have said I have used it” was 37% for cannabis and 5% for heroin. These numbers can be compared with the lifetime prevalence for these drugs, which was 44% and 1% respectively. A very small percent of the students reported use of the dummy drug “relevin” (0.2%), while 9% claimed that they had heard of it.

Methodological considerations Compared to the two previous data collections in 1995 and 1999 the geographical coverage of the sample was improved in this survey, since all regions were represented in the sampling frame. The drawing of the sample could, however, have been better related to the total number of classes. As it was, those schools with a number of classes greater than 2 were represented twice in the drawing list (most schools had 2–4 classes in the actual grade). This means that all classes didn’t have the same chance to be chosen, but the probability is not directly related neither to school nor class size. If a distinction between schools with different number of classes should be made, it would have been preferable to list all classes in the sampling frame and to let the random technique work fully. The fact that class sizes were not known means that each class had the same weight regardless of size and small classes are over-represented in the sample. Another fact, that makes the Czech sample somewhat less representative is that only 65–70% of the actual age cohort have entered secondary school. Despite these problems the Czech sample probably reflect approximately the student cohort under study. Apart from this, the data collection seems to have been successful and no class or individual student refused to participate. The reliability and validity measures did not indicate any problems; all these values were generally low. The deviant outcome when comparing lifetime prevalence of cannabis and heroin, with the honesty question is difficult to interpret. A possible explanation could be that some students misinterpreted the honesty question. The overall impression is, however, that the survey resulted in reliable and valid data.

Denmark In Denmark, Dr Svend Sabroe, Department of Epidemiology and Social Medicine, Aarhus University and Dr Kirsten Fonager, Department of Social Medicine, Aalborg Hospital were responsible for the ESPAD study. Denmark also participated in the 1995 and 1999 ESPAD studies.

Population The target population consists of all students in Denmark born in 1987. More than 98% of all chil-

236

dren born in 1987 were still in school at the time of the data collection.

Sample and representativeness Of all students born in 1987 about 85% were found in grade 9, about 10% in grade 8 and the last 5% in grade 10. Like in the 1995 and 1999 surveys data collection was limited to students in grade 9. They were found in public schools as well as private and boarding schools.

Appendix I

The sampling frame consisted of six strata. Four of them were public schools where the stratification variables were size of the municipality and school size. The fifth stratum was private schools while boarding schools was the sixth. In the four strata of public schools, classes were sampled in proportion to the number of classes. In the last two strata sampling was made at school level since these schools are often not organised into classes In these schools as well all grade 9 students were sampled. In the first four strata 10% of the classes were included in the sample, all together 214 classes. Stratum 5 contained 41 schools and stratum 6 33 schools. Within each stratum each class (strata 1-4) and school (strata 5-6) had the same sample probability. In practice this meant that students in small classes and schools were over-represented in the samples. It is stated in the country report that there are usually no large differences in the sizes of the classes in Denmark. It is also worthy of note that the ESPAD 99 study did not indicate any important differences in alcohol and drug habits between students in small or large schools in the two samples. The sample was done in the same way as in 1995 and 1999 and is considered representative for all grade 9 Danish students born in 1987.

Field procedure The selected schools were contacted in January 2003 by a letter to the principal. It contained an inquiry form as to whether the school wanted to participate or not. It also contained a request for information on the name of the class teacher in the sampled classes. Two weeks before the data collection all relevant material was sent to the teacher. The students answered the questionnaire under the same conditions as a written test. The average time used was 37 minuets. After completion, the questionnaires were put in individual envelopes. Data were collected under the supervision of the class teacher and was performed between March 6 and May 2, 2003, which gives an average age of 15.8 years. All students in grade 9 participated. However, the ESPAD report only includes data from students born in 1987.

Questionnaire and data processing All core questions were asked except two (play on slotmachines and the consumption of cider). The questionnaire also contained the Integration module and two questions from the Mainstream module

Appendix I

as well as 8 new questions. The new questions were translated and back translated. No pre-test was done. Questionnaires with many strange comments or extremely many outliers were flagged and checked manually by the research team. In the national report it is evident that students in private schools were underrepresented. However, national data were not weighted.

School and student co-operation Of the 74 sampled private and boarding schools 39 participated in the study. In the sample of 214 classes in public schools 140 took part in the survey. Non-participating schools or classes were not replaced. The research team made a phone call to all schools that did not return the letter, which resulted in another 18 schools accepting the invitation to participate. The most common explanations for non-participation were that the schools did not have the time and that they had received many inquiries to participate in lifestyle surveys. In the national report it is stated that there are “no indications that non-participating schools should be associated with a different level of alcohol consumption or drug use...”. The assumption is manly based on the information in the paragraph above and on the fact that no school mentioned alcohol or drug consumption as a reason to refuse. One other aspect mentioned is that the schools had not seen the questionnaire in advance so they did not know that all the detail of the content on alcohol and other drugs. No present student refused to participate. The response rate was 90%. Very few questionnaires (0.3%) were eliminated during the scrutinising process. Most teachers (84%) did not notice any disturbances during the data collection while 13% reported that this happened only among a few students. The most common reported disturbance was “other kinds of comments” (9% of all classes) followed by loud comments (8%) and giggles or eye makings (7%). In nearly all participating classes (99%) the sur vey leaders reported that “all”, “nearly all” or “a majority” of the students were interested in the study (95% answered “all” or “nearly all”). The corresponding figures on the question whether the students worked seriously were 100 and 99% respectively.

237

Reliability and validity The inconsistency rate between two questions in a single administration was highest for cigarettes and inhalants (3%) and lowest for all other substances (0–2%). Missing data rates on some drug related questions was highest for the variables been drunk and alcohol consumption (3%) and 0–2% for other drugs. Looking at the questionnaires as a whole, 1% of the questions were left unanswered. The rates of inconsistent answers to questions about use in lifetime, last 12 months and last 30 days were low (0–1%) for all 4 drug related variables. For cannabis 3% of the students answered “definitely not” on the question “If you had used marihuana or hashish, do you think you would have said so in the questionnaire?”. The corresponding figure for heroin was about the same (5%). On the “willingness question” 20% answered that they had already said that they had used cannabis, which is close to the reported proportion (23%). Six per cent answered that they had heard about the dummy drug relevin. However, only 0.1% said that they had used it.

Methodological considerations No student refused to participate, the number of eliminated questionnaires was very low, nearly all survey leaders answered that the students were interested in the study and worked seriously. Nearly all comments from the teachers were positive. Hence, available information indicates that student co-operation was good. None of the reliability and validity measures indicate any major problems in the Danish ESPAD study. The sample probably included an overrepresentation of students from small classes (strata 1–4) and schools (strata 5–6). However, this does not appear to be a major problem since there are no big

differences in size between small and large classes and since the 1999 Danish ESPAD survey did not report any significant differences in alcohol and drug habits between students in small and large schools in the two samples. Hence, the sampling of classes in public schools and of schools in the two strata of private and boarding schools seems to have functioned without any problems of note. The high non-response rate is a concern. 74 out of 214 classes in public schools (35%), 21 out of 41 private schools and 14 of the 33 boarding schools did not participate in the data collection exercise. Taken together this implies that 38% of the sampled units refused to take part in the study. Even if these figures are high it should be appreciated however that they are lower than those reported in the 1999 Danish ESPAD study. Schools that did not respond to the first contact were contacted by telephone. The main reason for them not to participate was that they did not have the time and that they had taken part in many other surveys. A comparison between participating and non-participating schools did not indicate any systematic differences. Once again, taken together this indicates that the relatively large number of non-participating schools and classes probably did not cause any major problems as far as representativeness is concerned. However, some uncertainty still remains. Of all 1987 born students about 85% were to be found in grade 9, while the others were mainly in grade 8 (about 10%). Hence, the sample is representative only for 1987 born students in grade 9 (with some uncertainty related to the relatively large number of schools and classes that did not participate). It seems reasonable to assume that the Danish data are comparable with the results from other countries. However, the relatively large number of classes and schools that refused to participate must be borne in mind.

Estonia Airi-Alina Allaste, from the Institute of International Studies, Tallinn Pedagogical Institute was responsible for the conduct of the Estonian study. Estonia also participated in the ESPAD studies in 1995 and 1999.

Population The population consisted of all students born in 1987 in grades 8, 9 and 10 in basic and secondary

238

schools. Since there were rather few students in evening and vocational schools they were excluded from the survey. Compared to the sample in the 1999 ESPAD study, grade 8 was added to the target population in 2003. It has been calculated that approximately 80% of all those born in 1987 were at school at the time of the data collection.

Appendix I

Sample and representativeness A list containing the number of students in each Estonian school and class was made available. In the first step a systematic sample was done to identify 100 schools. Since this did not provide enough 1987 born students in the final sample another 20 schools were sampled. One of them was already sampled which resulted in a total of 119 schools. In all schools a random sample of one grade 8 class, one grade 9 class and one grade 10 class was drawn. All schools did not have classes in all grades and thus the final number of sampled classes was 324. It has been calculated that about 80% of all students born in 1987 were to be found in the three participating grades (8–10). The sample is selfweighted and the results are judged to be representative for 1987 born students in Estonia.

Field procedure The heads of the sampled schools got a letter, which explained the study. The letter was accompanied by supportive letters from the Ministry of Social Affairs as well as from the Ministry of Education. The material was brought to the schools by research assistants, which was not the case in 1995 and 1999 when it was sent by mail. The reason for this change was that there had been a number of “unprofessional surveys” sent to Estonian schools over the last couple of years and it was judged necessary to deliver by hand in order to negate the possibility of any mishaps. Data were collected by research assistants. However, in most cases a teacher was also present and he/she was responsible for answering part of the classroom report, that dealt with the number of present and absent students. After the instructions were given, the questionnaires were answered under the same condition as a written test. When they were finished the students put their questionnaires in individual envelopes. In a majority of the schools students born in 1987 in selected classes were asked to go to a special room to answer the questionnaire. In some schools data were collected in the classrooms, after students not born in 1987 were asked to leave the room. The study was conducted in March, which gave an average age of 15.7 years. The average time to answer the questionnaire was 35 minutes.

Questionnaire and data processing All core questions were asked together with four out of six questions in the Integration module and

Appendix I

all questions contained in the module referred to as Psycho-social measures. The questionnaire also included the same country specific questions as in 1999 as well as a new question about involvement in subcultures. Since Estonia also participated in earlier ESPAD data collections the core questions had already been translated for the previous studies. The Estonian as well as the Russian version of the questionnaire were tested, after which some minor changes were made in both versions. For some reason 20 students not born in 1987 answered the questionnaire. These questionnaires were excluded together with 2 others (0.1%) that did not satisfy the inclusion criteria. Data were not weighted.

School and student co-operation Ten schools refused to participate or were impossible to contact. Data were also missing from 66 classes. Some of the schools that did not participate were small schools with only a few students born in 1987. This was also the case with many of the classes that were reported as missing. In some of them there were no students born in1987. In most others only very few students of the target population were to be found. Of the 2,863 1987 born students that were calculated to be found in the sampled schools and classes 2,463 were found in participating schools and classes. This would appear to confirm that most of the non-participating schools and classes included no or only a few students born in 1987. No present student refused to participate. The response rate was 86% which was a bit lower than 1999. The main reason put forward to account for this was that the data collection in ESPAD 03 was done during a flu-period. About half of the survey leaders (51%) did not notice any disturbances during the data collection, while 39% reported that this happened with a few students. The most common reported disturbance was giggles or eye makings, which was answered by 41% of the research assistants. In a large majority of the classes (89%) the data collection leaders reported that “all”, “nearly all” or “a majority” of the students were interested in the study (72% answered “all” or “nearly all”). The proportions answering that the students worked seriously were even higher (96% and 83% respectively). It is stressed in the Estonian report that the data collection went well without any important disturbances.

239

Reliability and validity Reliability measured by inconsistency rates between two questions in a single administration was highest for cigarettes, “been drunk” and cannabis (4–5%). For other substances the corresponding figures were 1–3%. The proportion of unanswered drug questions was highest for the variables alcohol consumption and been drunk (3% each). For other substances it varied between 1 and 2%. The number of unanswered core questions is high (25%), which also gave rise to a high figure for the questionnaire as a whole (21%). The reason for such is attributable to mistakes in the lay out and coding of Q37 and some other multiple questions. For cannabis 8% of the students answered “definitely not “ on the question “If you had used marihuana or hashish, do you think that you would have said so in this questionnaire?”. The corresponding figure for heroin was 9%. On this “willingness question” 18% answered that they had already answered that they had used cannabis, which is a slightly lower than the lifetime prevalence figure (23%). Nine per cent answered that they had heard of the dummy drug relevin. However, only 0.2% said that they had used it.

Methodological considerations The stratified sample seems to have functioned without any problems, which is indicative of the fact that the sample was representative for Estonian students born in 1987. Contrary to the 1999 survey, students in grade 8 participated in the 2003 data collection. However, this factor per se has had a minor influence on the results and thus the possibilities to compare the two surveys has not been compromised. The number of schools that did not participate was rather low (10), while the number of classes that did not take part in the study was higher (66). However, in most of these missing schools and classes no or only a few of the 1987 born students were to be found. Hence, the number of students from the target population that were missing due to non-participation of schools and classes was rather low. Data were collected by research assistants which is a change compared to the 1999 survey when teachers were responsible for the data collection. The reason for this change was to counter the possibility of teacher withdrawal as they might

240

have already participated in other school surveys (some of them of a rather low quality). The use of research assistants was judged to increase the number of participating classes. This possible change in the teachers willingness to work as survey leaders occurred following the 1999 survey and the Estonian ESPAD researcher however has assumed that the quality of the data collection in 2003 is more or less on par as that in 1999. Rather many survey leaders reported some kind of disturbances during the data collection. However, since they were research assistants, and not teachers that are used to a “normal level of disturbances” in a classroom, they were probably more observant or more sensitive to specific type of disturbances than teachers. Hence, it seems reasonable to assume that the disturbances during the data collection in Estonia were not more serious than those found in most other ESPAD countries. Such a conclusion is supported by the fact that a large majority of the survey leaders reported that the students were interested and worked seriously. In most schools students born in 1987 in sampled classes were asked to go to the same room to participate in the study. In some other schools the data collection took place in the classrooms of the sampled classes, but only with the participation of those born in 1987. This is not in line with the ESPAD guide lines, but it would appear from both the classroom reports and validity checks that this factor has not influenced the outcome to any significant degree. Very few students refused to answer questions about their alcohol and drug habits. On the other hand, the proportion of unanswered questions in the questionnaire as a whole is high due to a technical mistake with some of the questions with multiple answering categories. This problem was limited to these multiple questions and did not influence the quality of the answers to the questions about alcohol and drug use. No present student refused to participate, the response rate was relatively high and the number of eliminated questionnaires was low. All of this is indicative that student co-operation was satisfactory. None of the reliability and validity measures suggest any major problems in the Estonian study. As a whole, data seem to be representative and comparable with the results from other ESPAD countries.

Appendix I

The Faroe Islands Dr Pál Weihe, Department of Occupational and Public Health, Faroe Hospital System, was responsible for the study in Faroe Islands. The country also participated in the 1995 and 1999 ESPAD studies.

Population The target population consisted of all students in the Faroe Islands born in 1987. The total number of students was 743, which is 95% of all persons born in the country in 1987.

Sample and representativeness No sample was drawn since the total target population was so small. Students born in 1987 were in the main found in grade 9 (92%). All together there were 39 grade 9 classes in 19 schools. The study is representative for all students in the Faroe Islands born in 1987.

Field procedure Staff members from the research institute made an appointment with the principal of every single school bout the day and the time of data collection. In accordance with the routines of earlier studies the material was distributed to each school. Staff from the Department of Occupational and Public Health were responsible for the data collection and the students filled out the questionnaires under the same conditions as a written test. After completion each student put his/her questionnaire in a sealed box. Data collection took place on March 10–21, 2003, which gave an average age of 15.7 years. The average time to complete the questionnaire was 55 minutes.

Questionnaire and data processing Skilled staff from the department translated and back-translated the questionnaire. All core questions were included in the Faroese version of the questionnaire. It also contained the questions on all 4 ESPAD modules as well as the optional questions, all together 94 variables. In addition 11 questions were asked about national identity and 7 about leisure time activities. Most questions had been used in earlier studies, and as such the pre-test was limited to 2 15 years old volunteers. A scanner was used to enter the data into the computer with appropriate software that signalled any errors. Data were not weighted.

Appendix I

School and student co-operation One small school with 2 students did not participate for technical reasons. No present student refused to answer the questionnaire. The response rate was 86%. No questionnaire was excluded. In rather few classes (19%) some kind of disturbance was noted during the data collection. However, in nearly all cases this was only reported for a few students. In the national report it is stated that the overall assessment of the student co-operation was judged to be excellent. All schools reported that “all” or “nearly all” students were interested in the study and the figures were more or less the same on the question whether the students worked seriously.

Reliability and validity The inconsistency rates between two questions in a single administration, which is used as a measure of reliability, was a little higher for cigarettes (7%) compared to other substances (0–3%). The proportion of unanswered questions on different substances varies between 1 and 5%. Looking at the questionnaire as a whole 5% of the questions were not answered. The rates of inconsistent answers to questions of use in lifetime, last 12 months and last 30 days were low (0–2%) for the four variables alcohol consumption, been drunk, cannabis and inhalants. For both cannabis and heroin about 3% of the students answered that they would not have admitted use of these drugs. On the same question, 11% of the students answered that they have already said they have used cannabis, which is a little higher than the reported value (9%). Of all students, 5% reported that they had heard about the dummy drug relevin. However, only 0.3% answered that they had used it.

Methodological considerations Since the country is so small all students were included in the study. Only one school with 2 students did not participate for technical reasons. The response rate is acceptable and no important disturbances were reported from the data collection. No student refused to participate, no questionnaire was eliminated and the proportion of schools with reported disturbances was not high. All these indicators suggest that the school and student cooperation was good.

241

In the 1999 ESPAD study the non-response rate was 22% compared to 14 in 2003. The proportion of unanswered questions was very high in 1999 (27%) and is now down to 5%. Hence, the quality of the data collection has improved since last time.

None of the reliability and validity measures indicate any methodological problems in the Faroe study. As a whole, data seem to be representative for students born in 1987 and comparable with other ESPAD data.

Finland Professor Salme Ahlström and Leena Metso at the National Research and Development Centre for Welfare and Health (STAKES) were the principal co-ordinators for the ESPAD study in Finland. Finland also participated in the ESPAD studies in 1995 and 1999.

Population The target population was all students in Finland born in 1987. Of all the persons born in this year nearly 100% were at school at the time of the data collection.

Sample and representativeness The study was conducted with students in grade 9. In this grade, approximately 95% of all students born in 1987 were to be found. Finland was divided into five regions according to EU area-divisions. These five regions were further divided into urban and rural areas. Besides these 10 strata, the Helsinki metropolitan area was assigned a stratum to itself. A systematic random sample was drawn and in each stratum the probability of a school to be sampled was proportionate to the size of the school. All together 200 schools were included in the sample. Each school was also assigned a substitute school, which was the next school in the file. In each of these schools one class was randomly chosen. The sample was selfweighted and representative for Finnish students born in 1987.

Field procedure All principals in selected schools received a letter with information on the objectives of the study. They were asked to name the teacher from the sampled class. In the middle of March material was sent to the contact teachers. Since some principals did not answer before a set deadline material was also sent to 16 schools from the extra sample (to replace possible non-participating schools). (Data from only seven of these schools were included in

242

the final data set.) After an introduction the students answered the questionnaires under the same conditions as a written test. Every student put his/her questionnaire in an individual envelope. Together with the classroom report the teachers returned the envelopes to the research institute. In a large majority of the schools data collection occurred during the last week of March. A small number of schools collected data during the following weeks. Based on a calculation of the large majority that conducted the survey in late March the average age was 15.7 years. The average time to complete the questionnaire was 31 minutes. All students in sampled classes took part in the study. However, the questionnaires from the few who were not born in 1987 were excluded afterwards.

Questionnaire and data processing Nearly all ESPAD core questions were included. Since alcopops is not well known in Finland it was replaced by “long drinks”, which are quite popular. The questionnaire also included questions from the Integration module as well as three own questions. The new questions, i.e. the ones not used in earlier ESPAD studies, were translated by the research team. No pilot study was conducted to test the limited new questions. In the scrutinising process data from 23 students (0.6%) were excluded due to unreliable and inconsistent answers. Data were not weighted.

School and student co-operation Of the 200 sampled schools and classes seven did not participate. They were replaced by substitute schools/classes. No present student refused to participate in the study. The response rate was 92%. According to the Finnish country report student co-operation was very good.

Appendix I

Most teachers (76%) did not notice any disturbances during the data collection. When this occurred it almost always included a few students (reported from 22% of the classes). The most commonly reported disturbances were “loud comments” and “other comments”, each of which was reported by 13% of the survey leaders. In nearly all participating classes (96%) the sur vey leader reported that “all”, “nearly all” or “a majority” of the students were interested in the study (84% answered “all” or “nearly all”). The corresponding figures on the question whether the students worked seriously were 99 and 94% respectively.

Reliability and validity Inconsistency rates between two questions in a single administration, which are used as measures of reliability, were highest for cigarettes and inhalants (3–4%). The figures for all other substances varied between 0–2%. The proportion of unanswered questions about different drugs varied between 0 and 2%. Looking at the questionnaire as a whole, 1% of the questions were not answered. The inconsistency rates between lifetime, last 12 months and last 30 days were low (0–1%) for all four variables (alcohol consumption, been drunk, cannabis and inhalants). For cannabis 2% of the students answered “definitely not” on the question “If you had used mari-

huana or hashish, do you think you would have said so in the questionnaire?”. The corresponding figure for heroin was 4%. On this “willingness question” 10% answered that they had already said that they had used cannabis, which is about the same as the reported prevalence figure (11%). Eight per cent answered that they had heard about the dummy drug relevin. However, only 0.4% said that they had used it.

Methodological considerations The stratified sample was configured without any difficulties and is representative for all students born in 1987. Only seven schools/classes refused to participate. Since no important problems were reported in the contacts with the schools, school co-operation seemed to have functioned well. No student refused to participate, the number of eliminated questionnaires was low, the proportion of survey leaders that reported disturbances was not high and nearly all survey leaders reported that the students were interested in the study and worked seriously. All of this is indicatives of the fact that student co-operation was satisfactory. None of the reliability and validity measures suggest any methodological problems in the Finnish study. As a whole, data would appear to be representative and comparable with other ESPAD data.

France The French study was coordinated by Dr. Marie Choquet at Institut National de la Santé et de la Recherche Médicale (INSERM) and François Beck at Observatoire Français des Drogues et des Toxicomanies (OFDT).

Population The target population consisted of students born in 1987 in all types of education including private establishments and schools with adapted teaching (EREA). Moreover, students in DOM TOM and overseas territories: West Indies, Guyana, and Bourbon Island were not included in the sampling frame.

Sample and representativeness The French study covered all grades from 6 to 12. The Ministry of Education conducts a population

Appendix I

census of the population of pupils each year in September. It was estimated that the large majority of the students born in 1987 were distributed in grades 9 and 10. The French schools are classified according to “zone d’éducation prioritaire, ZEP” i.e. schools with need for reinforced educational action. The sample of 450 schools was drawn from the computerised list of schools, which was updated at the end of November 2002, as a stratified random sample of schools proportionate to school size. The strata represented type of school, type of area (urban/rural) and educational characteristics (priority zone or not). From each selected school two grades were selected by simple random sampling, where the head master identified two classes with a teacher’s name closest to L in the alphabet, resulting in a

243

sample of 900 classes. The gender distribution in the different types of schools was 50% girls in public and 48% in private junior high schools, and 55% in both types of senior high schools. The sample, which covers all age groups from 11 to 19, was considered to be self-weighted.

Field procedure The headmasters were contacted and informed that the schools had been drawn for the ESPAD 03 survey. They were asked to appoint a person to supervise the data collection (school doctor or nurse). A serious complication that arose during the data collection period was that a strike came into force for school doctors and school nurses in France. However, the research team Inserm U472 was well known among school doctors and nurses, and most of them (400/450) accepted to perform the data collection for the ESPAD study. They received a phone call with the relevant information about the survey. The students were invited to participate in the survey and to complete the questionnaire during a lesson. The supervisor of the data collection informed the students in a standardised way, reading from a paper and he/she also read the text on the front page of the questionnaire. After completion the students were asked to seal the questionnaire with two stickers and to put it in a box. Neither teachers nor headmasters were present in the classroom during completion of the questionnaire. Data was collected between March 17 and May 18, which gave an average age of 15.8 years.

Questionnaire and data processing Two versions of the questionnaire were used in the French study, of which the short version was used in grade 6 and 7 and in classes labelled as SEGPA (General education and professional adapted sections). One specific detail regarding differences in the questionnaire was that the French version did not include “or some other hallucinogens” to the specific question on LSD. It was considered to overlap with the specific question on “magic mushrooms”. Other changes were made, i.e. the question on drinking beer at last drinking occasion, where the indicated volumes were changed to better relate to the usual drink size in France. However, the version used for the 1987 birth cohort included only 56% of the ESPAD core questions. Moreover, some module and own questions were inserted in the core section of the questionnaire. In total, the questionnaire included 52% ESPAD core, 5% ESPAD module and 43% own questions. The module

244

questions used was a selection from the ESPAD modules A–D. Some important methodology questions, such as the ESPAD honesty questions were omitted. A translation and back-translation of the questionnaire was done and resulted in some adjustments in relation to the French context. The questionnaire was pre-tested in two schools, with 115 participating students from different grades. As a result the questionnaire was modified into a final version with a better presentation of the questions and in some cases simpler wording. Before data processing 205 (1.2%) of the questionnaires were excluded because they were obviously not seriously answered. SAS statistical package was used and programmed according to the suggested SPSS syntax.

School and student co-operation In France passive parental consent is required for students below the age of 18. A non-response was considered as a passive consent. Overall, very few parents prevented their child from participation (1.2%) Unfortunately, the implementation of the survey in France was affected by some serious problems. The main problem was the strike that caused a loss of 50 schools (27 in which the headmaster refused to do the survey, 18 because of boycott, 5 because health staff were on strike, in total 100 classes, or 11%). However, differences between participating and non-participating schools were examined and no significant differences were found in relation to geographical or school characteristics. From the classroom reports it was apparent that no disturbances occurred in 62% of the classrooms. The disturbances noted were giggles or eye makings, which accounted for over half of them. The data collection leaders estimated that in 96% of the classes a majority of the students were interested in the survey and worked seriously. The response rate was 91% and the average time to complete the questionnaires was 45 minutes.

Reliability and validity Reliability measured by inconsistency rates between two questions in a single administration was not possible to do because of a change of format in one of the questions. The French students were asked to write the age at which they first tried a drug – not tick an alternative as in the ESPAD questionnaire. If they were not concerned, did not remember or did not want to answer they wrote nothing (there was no modality “never”.

Appendix I

The proportion of unanswered questions was low in general. For alcohol it increased somewhat from lifetime (3%) to 12 months or 30 days (5% for both). The same pattern was found in relation to the variables “been drunk” with 2% for lifetime prevalence to 7% for 12 months or 30 days, and cannabis use (1 to 4%). The average on lifetime use of other drugs including cannabis was 2%. The average number of unanswered core and module questions was 3%. The same measure for “own questions” was not possible to calculate since a skip sequence was introduced later on in the questionnaire. The inconsistency rate between lifetime, 12 months and 30 days prevalence rates was highest for alcohol use (5%) and “been drunk” (2%) followed by cannabis use (2%) and use of inhalants (0.3%). The two questions on the possible unwillingness or not to admit cannabis and heroin use were not asked. 7% of the students reported that they had heard about the dummy drug, in France called “mop” instead of “relevin”. However, less than 1 percent reported use of this fictitious drug.

Methodological considerations The French study is based on a good representative sample covering all grades in which students born

in 1987 can be found. The study encountered serious difficulties in the form of a strike among health staff, some of which were due to supervise data collection. Combined with other types of refusals the loss of classes in the sample was 14%. The French questionnaire was to a large extent modified and it deviates from the common ESPAD version. In total, the questionnaire included 52% core ESPAD, 5% module ESPAD and 43% own questions. The reliability and validity measures are incomplete, since the inconsistency check between two questions in a single administration is impossible to effect due to a change in the format in one of them, and the “honesty” questions about cannabis and heroin were excluded from the questionnaire. It is unfortunate that the study encountered difficulties and that it deviates in some respects from the common ESPAD methodology. The French team has, however, tried to analyse the loss of classes in the sample and found no significant difference between them on geographical and school characteristics. The fact that the proportions of unanswered questions are low in general and that other measures of validity and reliability show very low values suggests a good data quality despite the problems.

Germany Dr. Ludwig Kraus at the Institute for Therapy Research (IFT) in Munich was responsible for the German ESPAD study. This was the first time that Germany participated in the ESPAD project. The study was done in 6 out of 16 federal states (Bundesländer). They were Bavaria, Brandenburg, Berlin, Hesse, Mecklenburg-Western Pomerania and Thuringia.

Population The target population consists of all students in the 6 Bundesländer born in 1987. The study was limited to students in grades 9 and 10. School is com pulsory up to the age of 18. It has been calculated that 92% of all youngsters born in 1987 were enrolled in school at the time of the data collection.

Sample and representativeness The school system differs between Bundesländer. However, all grade 9 and 10 classes in regular types

Appendix I

of schools were included in the sampling frame. Non-regular schools such as special schools for retarded students or vocational schools were excluded from the study. These schools are calculated to include about 8% of all students. Of all students born in 1987 about 84% were to be found in the grades of the sampling frame. The sample is representative for students born in 1987 in grades 9 and 10 in the six participating Bundesländer. Data were weighted for grade and class type. Moreover, since the Bundesländer vary in size, data representing the six Bundesländer together were also weighted.

Field procedure In each Bundesland a person working at the Ministry of Education was responsible for co-ordination and data collection. School principals in selected schools were informed by the co-ordinators, who

245

also were responsible for distributing the material to the schools. Data were collected in the classrooms by teachers who were not in charge of the selected class. After completing the questions the students put his/her form in a large class envelope. The envelope, which also contained the classroom report, was sealed by the teacher in front of the class before it was sent to the field institute for data entry. The average time to complete the questionnaire was 40 minutes. Data were collected in March and April, which gives an average age of 15.7 years.

Questionnaire and data processing All core questions were asked except the question about the consumption of cider (Q11) (since it is almost not prevalent in Germany). For the questions about the consumption on the last drinking occasion (Q10 and Q12–Q14) response categories were changed to numerical responses of standardised measures. Similar changes to numerical responses were also made for the binge drinking question (Q17) as well as for the question about drunkenness frequency during the last 30 days (Q19c). The questionnaire included the Deviance module as well as the first question of the Mainstream module. In addition to this two own questions were added about alcohol consumption. The translation of the questionnaire was done in close collaboration with the Swiss and Austrian principal investigators. No pre-testing was carried out. Data entry was double checked.

School and student co-operation Out of the 557 classes that were selected, questionnaires from 34 were not returned. The reason for this is not known. Another 15 classes refused to participate, of which 8 were selected for another study. These 15 classes were replaced. Overall 440 (4%) students had not received parental permission or refused to participate. Of the total number of relevant questionnaires (11,122) only 79 (0.7%) were excluded. 5,110 of the remaining 11,043 students were born in 1987. The response rate was 89%. In the German country report it is stated that the “students’ co-operation may be considered as good”. Information from the classroom reports show that no disturbances were reported from 82% of the classes and in most of the others (15%) this only happened from a few students. Talking between neighbouring student (a free text answer in the German questionnaire) was the most common dis-

246

turbance, which was reported from 8% of the participating classes. In nearly all participating classes (96%) it was reported that “all”, “nearly all” or “a majority” of the students were interested in the survey. Nearly all survey leaders (99%) answered that they thought that “all”, “nearly all” or “a majority” of the students worked seriously.

Reliability and validity The inconsistency rate between two questions in the questionnaire was highest for the variable “been drunk” (6%) followed by inhalants (3%). It was lower for cigarettes (2%) as well as for different illicit drugs (0–2%). Missing data rates were low for all kinds of questions. It was 2% for alcohol consumption and own questions and lower for all other categories of questions. For the questionnaire as a whole 1% of the questions were unanswered. The rates of inconsistent answers to the questions about use in lifetime, last 12 months and last 30 days was highest for alcohol consumption (3%) and “been drunk” (2%) and even lower for inhalants and cannabis (0–1%). Of all students 4% reported that they “definitely not” would have admitted possible use of cannabis. The corresponding figure for heroin was 9%. On the same question 24% answered that they had already said that they had used cannabis, which is close to the lifetime frequency figure (27%). About one out of 10 students (11%) gave the answer that they had heard about the dummy drug relevin. However, only 0.4% said they had used it. It is commented in the German country reports that there is no indication that the reliability or validity may differ between subgroups, different kind of schools, geographically or otherwise.

Methodological considerations The sampling procedure seems to have functioned well. There were only rather few sampled classes (6%) that did not participate in the survey. The results seem to be representative for students born in 1987 in grades 9 and 10 in regular schools in the six participating Bundesländer. The student co-operation was good even though 4% of the students did not receive parental permission or refused to participate. Only few questionnaires were excluded. The classroom reports indicate a high interest from the students. None of the reliability or validity measures indicate any major problems.

Appendix I

The only aspect that is judged to influence the possibilities to compare the German results with data from other ESPAD countries is the fact that open-ended categories with numerical responses were used instead of fixed answering categories on the question about alcohol consumption at the last drinking occasion (Q10 and Q12–14), the binge drinking question (Q17) and the question about the frequency of drunkenness during the last 30 days. To stress that this difference in the wording of the

answering categories might influences the possibility to compare with other ESPAD data, results from these questions are put below a line at the bottom part of the result tables. The overall impression is that the German study is well done. However, because of the use of numerical responses instead of fixed answering categories on six questions, the results on these six questions are judged not to be fully comparable with data from other ESPAD countries.

Greece The Greek study was conducted under the auspices of the University Mental Health Research Institute (UMHRI) and was co-ordinated by Anna Kokkevi Ph.D., Assoc. Professor at the Athens University Medical School, in collaboration with Manina Terzidou M.Phil., Head of the Greek National Focal Point (REITOX Network – EMCDDA). UMHRI also conducted the 1999 ESPAD study in Greece. Some data from a 1993 national study were included in the 1995 ESPAD report.

Population The target population consisted of all school students who’s 16th birthday occurred in the calendar year 2003 and were registered in secondary education, i.e. junior and senior high schools, situated on the mainland of Greece and on the islands of Crete and Evia. It is common practice in Greek surveys to exclude the smaller insular areas from the sampling frame, due to the logistical problems arising from the large geographical dispersion of the Greek islands and to limited financial resources. A rough estimate of the percent of children born in 1987 still in school was made by comparing the number of births that year with the number of students enrolled in all grades of all secondary schools in the school year 2001–2002. This suggests that, theoretically, all children of the actual age cohort were in school. The proportion of this age cohort included in the sampling frame (that is, excluding the smaller islands) was estimated to be 93%.

Sample and representativeness The sampling methodology was identical to that employed in the 1999 ESPAD study. Thus the sample was a nationally representative stratified clustered probability sample where the sampling

Appendix I

units were schools. The geographical strata were 1) Athens, 2) Thessaloniki, 3) other urban areas (municipalities of 10,000 registered inhabitants or more) and 4) semi-urban and rural areas (municipalities and communities of less than 10,000 registered inhabitants). In all strata the schools were randomly selected with probability proportional to their size, and classes were randomly selected within each school. The average class size in the sample was 23.4 students, which was very close to the national average. The sample consisted of 221 schools and 427 classes from junior and senior secondary education. In the former category, students born in 1987 were found in 78 schools and classes and in the latter in 104 schools and 204 classes. The sampled student population was considered representative for the age cohort under study and to be self-weighted.

Field procedure In autumn 2002, UMHRI addressed an official letter to the Ministry of Education requesting permission to conduct the study. The Ministry subsequently communicated its approval to the Regional Offices of Secondary Education (responsible for the schools drawn in the sample), and the latter in turn informed the school directors regarding the study and their expected role in the survey. A month prior to the field work, UMHRI sent an official letter to the school directors informing them of the study and the time-frame within which it would be carried out. The co-ordinators of the research assistants contacted the schools to make appointments for the implementation of the study. The administration of questionnaires took place in the classrooms and was supervised by a research

247

assistant. No teacher was allowed to stay in the classroom except in a few cases (1%), when the teacher insisted on doing so. The study was introduced to the students as one that was being conducted internationally that aimed to identify their health-related needs as a group. It was emphasised that the University of Athens was conducting the research and that the school staff had no connection with it or its results. Instructions regarding the completion of the questionnaire were given to the students, for example, to read carefully the introduction and to refrain from asking questions regarding the content of the questionnaire items. When the students had completed the questionnaires they were put into a special folder that safeguarded the anonymity of the respondents. Data were collected in March–April 2003, which gives an average age of 15.8 years.

Questionnaire and data processing All but one of the ESPAD core items (Q11 on cider) were included in the Greek questionnaire. Minor modifications (e.g. month of birth) were made and the module C (Psychosocial) was included. In addition some national questions were placed at the end of the questionnaire. The 1999 Greek questionnaire was carefully checked for discrepancies or up-dates against the 2003 English ESPAD version. The translation and re-translation was only done this time for parts of the questionnaire. Re-translation was carried out by an in-house social scientist that was not working on the ESPAD study. A computer check to detect possible coding or scanning errors was conducted. The checking process included cases of 1) unanswered items, 2) extreme values, 3) missing values and 4) errors in questionnaire code numbers. SPSS version 11 for Windows XP was used for data processing.

School and student co-operation The majority of the schools were willing to participate in the study. Only 5 schools refused to participate. The next school in the sampling frame of schools replaced these. Following these replacements, the ultimate response rate for schools reached 100%. However, 13 classes (2.9%) did not participate in the survey for various reasons, mainly because of other interfering activities. 12 students openly refused to participate in the study on the day of the administration. Overall, the students were extremely co-operative and interested in participating in the survey.

248

Based on the classroom reports from the collaborators, in the majority of classes (55%) there were no disturbances. In most of the cases where there were disturbances, only only a few students caused them. The most common type of disturbances was loud comments, sometimes stemming from the content of the questionnaire. The questionnaire items that caused most queries from the students were Q3 (activities), Q4 (absence), Q6 (ever smoked), Q20 (drunkenness scale) and Q22 (ever heard of). The level of student comprehension was overall very good; only a few cases of students of non-Greek origin requested clarification. The average time taken to complete the questionnaire was 52 minutes. The response rate was 88%.

Reliability and validity The consistency between two related questions in a single administration indicated quite high reliability, one question being the self-reported lifetime prevalence for the drug and the second question the age at first use of the drug. The highest inconsistency was observed in relation to questions on use of inhalants (6%), while the corresponding value for alcohol use was 5% and for cigarette smoking (3%). The figure for other variables was 1% or lower. The proportion of unanswered questions about various drugs was low overall (1%). For lifetime questions the highest proportion was observed in relation to alcohol (2%), while all other values were lower. A small increasing tendency for questions on 12 months or 30 days use was reported for “been drunk” (from 1 to 2%) or cannabis and inhalants use (from 0 to 1–2%). The inconsistency rate between lifetime, 12 months and 30 days use of any alcohol was 7%, while for “been drunk” it was 3% and for marijuana or inhalants use it was 1%. The two questions about possible reluctance to admit cannabis and heroin use, respectively, reveal that 7% answered that they had already said so in the questionnaire that they had used cannabis, compared to the actual prevalence rate of 6%. The proportion that answered that they would “definitely not” admit cannabis use was 4% and the same for heroin. Only 3 students (0.2%) reported use of the dummy drug “relevin”, while 9% reported that they had heard about this “drug”.

Methodological considerations The Greek study was based on a similar methodology to that employed in 1999. It was estimated that

Appendix I

93% of the target age group would be included in the sampling frame. This figure is based on the calculation of the population size in the insular areas (except Crete and Evia), which, as in the earlier studies, were excluded from the sampling frame. There is no information available about the impact of this exclusion on the results of the study, but it can be expected to be rather small. The only other possible failure to sample students from the target age group is in the case of those who were below the third grade of junior high school. The number of such cases is unknown, but was assumed to be minimal as this implies that students must have twice repeated a grade. As recommended by the research protocol, data collection for the 2003 study was conducted in the spring term (March–April) of the 2002/2003 school year. However, for the previous survey in 1999, due to repeated student walkouts during the spring term of 1998/1999, data collection was carried out in the autumn term of the following school year (1999/2000) (note: the 1983-born students were

consequently found in their next grade). This variation between the two data collection periods means that the 1999 ESPAD population consisted of students who were about 7 months older than their counterparts in 2003 (mean age 16.3 years in 1999 compared to 15.7 in 2003). The age variation between the samples of the two surveys should be taken into account when trend analyses are carried out involving the Greek data. The implementation of the survey in schools seems to have functioned well, and the students were overall extremely cooperative and interested, except in a very few cases of students who refused to participate. Low values on most of the methodological measures indicate a good quality of data. Very few students (4%) answered on the “honesty” question that they would not admit using cannabis or heroin had they done so. In all the Greek study seems to have functioned in accordance with expectation and seems to have provided reliable and valid data.

Greenland Paarisa (Ministry of Health) and Charlotte Lycke (Statistics Greenland) were responsible for the ESPAD data collection in Greenland. Thomas Andersen (Statistics Greenland) carried out the analyses and reporting. Greenland also participated in the ESPAD study in 1999.

Population The target population consists of all students in Greenland born in 1987. Of all 923 Greenlanders born in 1987 812 were estimated to be at school during the spring of 2003, i.e. 88%.

Sample and representativeness No sample was drawn since the total target population was so small. Students born in 1987 could be found in grades 9–11 in 73 of the 86 secondary schools and one special school (students from the special school were excluded from the survey). Consequently all these three grades were included in the data collection. It is estimated that nearly 100% of all students born in 1987 were to be found in the three participating grades. Since no sample was done the sampling procedure does not call for any weighting procedure.

Appendix I

Data are considered to be representative for Greenlandic students born in 1987.

Field procedure After an introduction that participation was voluntary and that full anonymity was guaranteed the students answered the questionnaire under the same conditions as a written test. When the students had finished the questionnaires they put them in individual envelopes, which were sent to Statistics Greenland together with the classroom reports. All students in grades 9–11 were asked to answer the questionnaire. However, the results in the ESPAD report only include data from those born in 1987. Teachers were survey leaders. The average time to complete the questionnaire was 69 minutes. Data were collected in March, which gives an average age of 15.7 years.

Questionnaire and data processing The questionnaire was translated from Danish to Greenlandic by a professional interpreter. The questionnaire was not pilot tested. All ESPAD core questions were asked except two. In Q11 cider was removed since cider is not sold in

249

Greenland. In Q3a “motorcycle” was changed to “snow mobile”. The Greenlandic questionnaire also included the Integration and Mainstream (3 questions) modules as well as 8 own questions. It is stressed in the Greenlandic country report that the answers of the Greenlandic students to some questions are not comparable with the answers of the students in other countries. One example is Q34, which is about perceived risk of different substances. Many Greenlandic students are unfamiliar with some of the drugs, which is indicated by a large number of students answering “do not know”, which heavily “has influenced” the proportion that have answered “great risk”. Another example is Q38 about the influence of heavy drinking. This “comparability problem” was “caused“ by different methodological factors, including the fact that the concept “heavy drinking” was translated with “drinking alcohol”. The fact that the results of these questions are not comparable with data from other countries will be commented in the result chapter.

eight students were inconsistent on the drunkenness as well as the inhalant questions. Missing data rates were rather high for all drug related questions; 12–13% for cannabis, inhalants, been drunk and alcohol consumption, 8% for tranquillisers and sedatives, anabolic steroids and other illegal drugs and 5% for cigarettes. Looking at the questionnaire as a whole 10% of the questions were left unanswered. The rates of inconsistent answers to questions of use in lifetime, last 12 months and last 30 days were high for the variables alcohol consumption and been drunk (10–12%) but lower for cannabis and inhalants (2–3%). A high proportion (30%) answered that they definitely not would have admitted possible cannabis use. The corresponding figure for heroin was 46%. On this “willingness question” 25% of the students answered that they had already said that they had used cannabis, which is close to the frequency figure (28%). Rather few students (5%) answered that they had heard about the dummy drug relevin and only 0.2% said that they had used it.

School and student co-operation It has been calculated that 68% of all students born in 1987 in Greenland answered the questionnaire. However, it is not known whether any of the 73 schools refused to participate or whether a full class did not participate for some reason. About two thirds (68%) of the survey leaders did not notice any disturbances during the data collection, while 30% reported that this happened only among a few students. The most commonly reported disturbance was loud comments (42% of all classes) followed by “other kinds of comments (37%). All survey leaders (100%) reported that “all”, “nearly all” or “a majority” of the students were interested in the study (93% answered “all” and “nearly all”). The corresponding figures on the question whether the students worked seriously were equally high (97 and 93% respectively). It is not known how many questionnaires that were excluded in the scrutinising process.

Reliability and validity The inconsistency rate between two questions in a single administration was highest for the variable been drunk (16%) followed by inhalants (11%). It was lower for cigarettes and cannabis (6–7%) and even lower for other illicit drugs, anabolic steroids and tranquillisers and sedatives (0–1%). In the Greenlandic Country report it was stated that only

250

Methodological considerations No sampling was done and all students in grades 9–11 in all 73 Greenlandic schools were supposed to answer the ESPAD questionnaire. Unfortunately no information is available about the number of schools or classes that refused to participate. Based on the assumption that 88% of the 1987 birth cohort were at school it was estimated that 68% of all students born in 1987 answered a questionnaire, which indicate that some schools or classes did not take part in the survey. The reliability inconsistency measures of reported lifetime use of different substances on two different questions, show rather high figures (compared with other countries) for four variables (been drunk, inhalants, cigarettes and cannabis use). The inconsistency figures are also high for some of the validity measures of inconsistency between lifetime, last 12 months and last 30 days prevalence figures. Compared to other ESPAD countries these figures are high for two out of four variables (been drunk and alcohol consumption). It should be noticed though that even in the worst case 84% of all students were consistent on these variables. The proportion of unanswered questions in the questionnaire as a whole is 10%, which is among the highest among all ESPAD countries. In addition to this it should be noticed that the proportion of students that definitely not would admit possible

Appendix I

cannabis (and heroin) use is extremely high in Greenland. Nearly all survey leaders answered that the students were interested in the survey and that they worked seriously, which indicate that the data collection ran smoothly. Different available reliability and validity measures indicate some concern about the Greenlandic data. Even though the data collection procedure did

not contribute to these concerns, they are there anyhow. Some measures indicate that the reliability and validity probably are a little lower in Greenland than in most other countries, which is of importance to keep in mind when comparing the Greenlandic results with data from other countries. Hence, some caution is recommended when the answers from the Greenlandic students are compared with data from other ESPAD countries.

Hungary Professor Zsuzsanna Elekes and Dr. Borbala Paksi at the Behaviour Research Institute, at the Budapest University of Economics and Public Administration, were responsible for the Hungarian study. Hungary participated both in the 1995 and the 1999 ESPAD surveys.

Population The population consisted of students born in 1987, who in 2003 were taught in elementary or secondary education. As in earlier studies, the population estimates were to be based on the previous year’s statistics, since no later data were available. However, according to these data it was estimated that 91% of the target population were taught in grades 8 to 10.

Sample and representativeness In Hungary education at grade 8 level is given through two types of schools. The majority of students attend classes at an elementary school, a smaller number are in secondary education. At grade 9 and 10 levels, there are three types of schools: grammar, training and specialised secondary schools. According to educational statistics from previous year, the percentage of 16 year-oldstudents in the sampling frame of grades 8, 9 and 10 was expected to be 8, 48 and 40% respectively (32.6% as an average). In Hungary 91% of the 1987 birth cohort was expected to be found in any of these grades. Taking into consideration the expected percentage of 16-year-old students in the multitude frame, the net sample size corresponding to the ESPAD requirements was 2800/0,326=8,589 students. The expected rate of sample loss was added to the estimate (based on earlier studies 3.5% for schools and 10.2% for students). The sample of 386 classes

Appendix I

was drawn as a stratified random cluster sample. To be able to analyse data from Budapest schools separately, these schools were over-represented by 100% (46 classes). Each class had the same probability to be drawn, independent of school size. However, mean class size in the study population was 25.7, while the corresponding value in the sample was 26,4.

Field procedure The schools included in the sample were contacted in February to inform the director and to ask for permission to perform the survey. Qualified interviewers and university students from the departments of sociology and social policy collected data, 80 in total. The teacher was asked to leave while the students filled out the questionnaire. They had got detailed instructions as how to answer questions from the students etc. Each student put his/her own questionnaire in an envelope placed at the front desk, which was sealed in front of the students. Only research assistants were present in class while the students answered the questionnaire. Data was collected between March 5 and 20, 2003, which gives an average age of 15,7 years.

Questionnaire and data processing All ESPAD core questions, except one for cider, were included in the questionnaire. Parts of the modules B and C were added. It was decided that the few questions from module B (B2 and B3) thematically belonged to the first section of the questionnaire and they were introduced there. An independent translator back-translated the questionnaire into English. It was piloted among approximately 100 students from all relevant types of education. Special attention was paid to the students in grade 8, which were included in the sample

251

for the first time. As a result all extra (own) questions were omitted to reach a format suitable for the students. In order to compensate for the over-representation of grades 9 and 10 in Budapest, and the loss of sample due to flu epidemics, the database was weighted in relation to school type and grade. Data was logically controlled and errors were corrected after check-up with the questionnaires. The number of invalidated questionnaires for students born in 1987 was 50 (1.6%). The SPSS programme version 11 was used for data processing.

School and student co-operation 21 of the selected classes refused to co-operate. The willingness to participate was higher in the countryside than in Budapest. In addition, there were 3 classes in which it was not possible to collect data (contact failure, school didn’t exist etc.). Of the 21 classes 16 were replaced, with others from a supplementary sample, but 8 classes were lost. In the classrooms two incidents of a student’s refusal was reported. However, in nearly 75% of the classes the survey leaders did not observe any disruptions and in another 20% only a few students made difficulties. In the majority of classes the students were interested in the survey and 90% of the survey leaders believed that the students took their task seriously. Moreover, in the majority of the classes the students found the questionnaire interesting. Only in a few classes (1.4%) they criticised or had problems in understanding the questionnaire. Average time to fill out the questionnaire was 48 minutes.

Reliability and validity Reliability measured by the consistency between two questions in a single administration was overall low. The highest was found in relation to ques-

tions on cannabis use (5%), while for smoking cigarettes, “been drunk” and use of tranquillisers or sedatives it was 4% on each. For other variables it was lower (2% or less). The inconsistency rate between lifetime, 12 months and 30 days use was highest for any alcohol use (4%) and “been drunk” (2%). For other variables (cannabis and inhalants respectively) it was 1%. The Hungarian researchers point to the fact that compared to previous ESPAD surveys, they found the most significant change in the case of cannabis as the rate of inconsistent answer among all informants has more than tripled in the years after 1995. However, at the same time the prevalence rates have increased too. When asked about their willingness to admit cannabis use 12% claimed that they already had answered that in the questionnaire, which is somewhat lower than the prevalence figure (16%). However, 6% answered that they would definitely not admit any such use. On a similar question about heroin use 7% gave this answer. The use of the fictitious drug “relevin” was reported by less than 0.5%, while 7% thought that they had heard about it.

Methodological considerations The Hungarian study covered this time the grades 8 to 10, which was an improvement from earlier studies focused on grades 9–10 only. Based on the experiences of quite many national studies it was carried out with meticulous methodology and the outcome was reported in detail. Also the reliability and validity measures point at a good quality of data. On the honesty questions, however, the percentage claiming that they already had declared the use of cannabis was somewhat lower than the actual proportion that did so (12 vs. 16%). On the other hand rather low proportions answered that they definitely would not admit any such use (6%).

Iceland Dr. Thoroddur Bjarnason, University of Akureyri was the principal investigator for the Icelandic ESPAD study. The data collection was conducted in collaboration with co-investigator Stefan Hrafn Jonsson and other researchers at the Icelandic Centre for Social Research and Analysis. Iceland also participated in the two ESPAD studies in 1995 and 1999.

252

Population In Iceland adolescents born in 1987 were found in 10th grade of compulsory school. The very small proportion of 1987 born students that were found in grade 9 (18 students) are not included in the target population. At the time of the data collection about 99% of the 1987 birth cohort was enrolled in school.

Appendix I

Sample and representativeness In the whole country, a total of 4,121 students were registered in 10th grade in 132 schools at the time of the survey. Instead of drawing a sample, all students attending 10th grade were targeted for participation in the 2003 ESPAD survey. Of all 1987 born students 99.5% were to be found in grade 10. The survey represents the population of grade 10 students born in 1987.

Field procedure Prior to the survey, a letter was sent to all 132 schools that included grade 10. The principals were asked to appoint a teacher as a contact person for the ESPAD survey. The contact teachers were asked to send a list of all classes in the school to the research team. Using these class lists, the research team prepared a survey package for each 10th grade class in the country. The packages contained the appropriate number of questionnaires and confidentiality envelopes, a letter to the teachers and a classroom report. For each school, all classroom packages were placed in a box, along with a letter to the contact person. In the capital area, these boxes were transported by research assistants, who in some cases also administered the questionnaires. Elsewhere the boxes were sent by certified mail and the survey was administered by school authorities. Data were collected March 8–28 under the same conditions as a written test. The average age of the students was 15.7 years and the average time to answer the questionnaire was 55 minutes. A total of 61 questionnaires filled out by 10th grade students who were not born in 1987 were omitted from further analysis.

Questionnaire and data processing Two versions of the ESPAD questionnaire were used. Form A closely followed the ESPAD standard questionnaire and included almost all core items. The major exception was that Q27a–n was only included in form B. Form B deviated somewhat from the ESPAD standard questionnaire and only included some of the core items. The latter form was used for some methodological purposes and for substantive research questions independent of the ESPAD survey. The questionnaires included the Deviance modules as well as a majority of the questions in modules A (Integration) and C (Psycho-social measures). With few exceptions the order of these questions followed the ESPAD core and module ques-

Appendix I

tions. Form A also included 27 country-specific questions and form B 43. The new ESPAD items were translated into Icelandic by the research team, read externally for linguistic accuracy and then translated back into English. The questionnaire was pre-tested in one grade 9 class and in a school programme for adolescents with substance use problems. The test resulted in some minor changes in wording and street names. The questionnaires were scanned. The optical data processing system was programmed to prompt for unusual entry when more than one mark was found for a question allowing only one answer. Random checks were conducted throughout the scanning process to assume consistent quality. Questionnaires were flagged if they met some specific criteria. All flagged questionnaires were collected and examined in one session by the research team. Data were not weighted.

School and student co-operation No schools or classes refused to co-operate, but 3 small schools with a total of 42 students in 10th grade did not return the questionnaires. No student who was present refused to answer the questionnaire. The response rate was 81%, which is the lowest response rate obtained in the Icelandic annual school surveys since 1992. The flu season may have played a major role. According to the classroom reports 12% of the students were sick on the day of survey administration. In the scrutinising process 26 out of 3,348 (0.8%) questionnaires were rejected. According to the data collection leaders, no disturbances were reported in 72% of the classes. Another 23% said that there were some disturbances among a few students. The most commonly reported disturbance was giggles or eye makings (16%). In nearly all participating classes (96%) the data collection leaders reported that “all”, “nearly all” or “a majority” of the students were interested in the survey (88% answered “all” or “nearly all”). The corresponding figures were even higher on the question whether the students worked seriously (100 and 96% respectively).

Reliability and validity Reliability as measured by the inconsistency rate between two questions in a single administration was not extremely high for any variable. The high-

253

est was found for inhalants (7%), while the figures were lower for other substances (0–3%). The inconsistency rate for use of alcohol, been drunk, cannabis and inhalants was about 1%. Five per cent of all students indicated that they would definitely not have admitted using cannabis and 8% said that they would definitely not have admitted using heroin. On the question about the willingness to admit drug use 15% answered that they had already said that they had used cannabis, which is more or less equivalent to the prevalence figure (13%). Of all students 11% answered that they had heard of the dummy drug relevin. However, only 0.3% said that they had used it.

Methodological considerations Since no sampling was done there are no sampling problems. Data were collected by research assis-

tants in some schools in the capital area and by teachers in the rest of the country. In practice the use of different kinds of data collection leaders in different parts of the country would not appear to influence the outcome as a methodological study has demonstrated that these two modes of administration do not produce different results in Iceland (Bjarnason, 1995). Student co-operation as well as school co-operation was satisfactory. The reliability and validity measures do not indicate any major methodological problems. The Icelandic ESPAD study seems to have been conducted without any important concerns. As a whole data seem to be representative for students born in 1987 and comparable with other ESPAD data.

Ireland Dr. Mark Morgan, St. Patrick’s College, Dublin was responsible for the Irish ESPAD study. Ireland also participated in the ESPAD data collection in 1995 and 1999.

Population The population consisted of students born in 1987 in all fifth grade classes in postprimary school. It is estimated that 93% of children born in 1987 were at school at the time of the data collection.

Sample and representativeness There are three types of schools: Single-sex secondary, mixed secondary and vocational and community schools. The schools were divided into these three strata. In the first sampling step schools were selected within these strata proportionate to the number of schools in the sampling frame. 120 schools were sampled. In the second sample step two classes were randomly sampled from each of the schools. It is estimated that about 67% of all 1987 born students were to be found in grade 5. The sample is representative of students in grade 5 born in 1987. Data are not weighted.

Field procedure The selected schools were contacted and, after having agreed to participate, the headmaster was asked to identify a liaison teacher to take responsibility

254

for the performance of the survey in the school. The questionnaires were mailed to the liaison teachers. Included with the questionnaire were guidelines for the administration of the survey. Only students who were born in 1987 were asked to go to the room in which the study was conducted. This is reported to have worked well. After instructions were given the questionnaires were answered under the same conditions as a written test. The students put their forms in individual envelopes. The average time to answer the questionnaire was 37 minutes. The data collection was done in April, which gave an average age of 15.8 years.

Questionnaire and data processing All ESPAD core questions were asked. The questionnaire also contained the Deviance module but no optional or own questions. No pre-testing was deemed necessary due to the previous experience with the ESPAD survey in 1995 and 1999, which proved to be satisfactory. The first 10% of the questionnaires were entered twice. Since this showed a 99% consistency, single data entry was used for the rest of the data.

School and student co-operation Out of 120 sampled schools 12 did not participate. They were not replaced by mutual schools. The

Appendix I

major reason for not participating was that many schools already had participated in other school surveys. In the remaining schools 196 out of 216 classes participated. The reasons for not participating varied but they were in general based on other activities that would have made the study difficult to complete in time. All present students answered the questionnaire, i.e. no one refused to participate. The response rate was 96%. Seventeen questionnaires (0.7%) were omitted following the scrutinising process. No major problems were reported by the survey leaders. A very large majority of them (97%) reported that they did not notice any disturbances during the data collection. All of them (i.e. 100%) answered that “all” or “nearly all” students were interested in the study. Also on the question whether or not the students worked seriously 100% reported that “all” or “nearly all” did so.

Reliability and validity The inconsistency rate between two questions in the questionnaire was low for all substances (1%). Missing data rates were highest for the variables alcohol consumption and been drunk (4–5%), while the proportions for other substances were lower (0–3%). For the questionnaire as a whole 2% of all questions were left out. The rates of inconsistent answers to the questions about use in lifetime, last 12 months and last 30 days were low for all drugs measured (1%). Of all students, 5% reported that they would “definitely not” admit possible use of cannabis. The corresponding figure for heroin was 10%. On the question about “the willingness to admit cannabis use” 36% answered that they had already said that they have used cannabis. The lifetime preva-

lence figure was very similar (39%). About one of 7 students (14%) gave the answer that they had heard about the dummy drug relevin. However, only 0.4% said that they had used it.

Methodological considerations In both sampling steps (first of schools and then of classes) each school/class had the same probability to be sampled, which, in principle, could result in an oversampling of students from small schools and classes. However, since there is not a huge variation in the size of Irish schools and since classes within a school usually are of the same size, there is reason to assume that the sample is adequate and representative for students born in 1987, who attended the 5th grade. However, it should be noted that grade 5 only accommodates about 67% of all students born in 1987. Consequently, the answers cannot be generalised to 1987 born students in other grades. The number of schools and classes that did not participate are not large and reasons given for nonparticipation do not indicate any important bias of the results. No student refused to participate, only a few (17) questionnaires were omitted and nearly all survey leaders reported a data collection without any disturbances with students that were interested and worked seriously. Hence, the student co-operation seems to have been good. No reliability and validity measures indicate any important methodological problems. As a whole, the Irish study seems to have functioned very well without any major problems. However, it must be kept in mind that data are only representative for 67% of the 1987 born students that attended grade 5.

Isle of Man Isle of Man is an internally self-governing dependent territory of the British Crown. It is not part of the United Kingdom, but is a member of the British Commonwealth. Dr Andreea Steriu, from the DHSS – Directorate of Public Health, Isle of Man, and Dr Jane Powell, Dr Patrick Miller and Professor Martin Plant, all from the University of the West of England, Bristol were responsible for the survey. Isle of Man did not participate in the earlier ESPAD studies.

Appendix I

Ethical approval was given to the study on condition that individual schools were not identified from the data and mentioned in the report.

Population The population consists of all students living in the Isle of Man who were born in 1987. Of all the young people born in this particular year a minimum of 80% are estimated to have been in school at the time of the data collection.

255

Sample and representativeness The Island’s Chief Registrar’s Report on births, marriages and deaths for 1987 shows that 729 students were born in the Island in 1987. To these, the 2001 Census identified a further 177 students born in 1987 that migrated to the Island after 1987 to give a total of 906. Students born in 1987 were to be found in grades 10 and 11 in a total of seven schools. No sampling was done. 1,974 questionnaires were distributed and 1,672 were returned. A total of 748 were identified as born in 1987. A further 27 were discarded and 721 were entered for data analysis. Data were entered for 721 students from a cohort of 906 young people that were born in 1987. This would suggest that ESPAD 03 was representative for all 1987 born students in the country, with participation of 80% of all persons born in 1987 and who lived in the Island in 2003. It has been estimated that about 95% of all 1987 born students in the Isle of Man were to be found in the two participating grades.

Field procedure Each school was contacted through the Head Teachers who in turn had delegated liaison officers. The questionnaires were distributed to all schools by local project managers from DHSS – the Directorate of Public Health. The data collection was conducted during tutorials or health education classes under examination conditions. Each student was provided with an individual envelope to place the completed questionnaire. Data were collected in the class rooms under the supervision of a teacher. All students in participating classes answered the questionnaire, i.e. also students not born in 1987 (all in all 1,672 students). However, the results in this report were limited to 721 students born in 1987. The envelopes were returned to the co-ordinating agency, DHSS. The survey was administrated during a period of five weeks (March 31 – May 3, exclusive of Easter break), which gave an average age of 15.8 years. The average time to complete the questionnaire was 60 minutes.

Questionnaire and data processing Isle of Man used the same questionnaire that was used in the UK study. It included all core questions as well as the three modules of Integration, Mainstream and Psycho-social measures. In addition to this, one question was added about alcohol and seven others that were related to changes in the

256

legal classification of cannabis. Since the UK questionnaire was used no translation or field testing was done. All data of Isle of Man were checked alongside the UK data for accuracy and implausibility. Data were self-weighted.

School and student co-operation All seven schools with grade 10 and 11 students participated. Eleven students, of the 1683 present in participating classes at the day of the data collection, refused to participate. The overall response rate was 85% when considering participation of all grade 10 and 11 students. About two thirds of the students who were not at school, were absent due to illness. If one also include others that had “acceptable reasons” this figure rises to about 96%, which implies that 4% were absent due to truancy. Of the 748 questionnaires that were answered by 1987 born students 27 (3.6%) were excluded. The classroom report was not used. However, very few disturbances were reported by the survey leaders. When this happened it was nearly always giggles or eye makings to the class mates. Only one student was reported to have made loud comments. All students but one were judged to have been interested in the survey and worked seriously. No comments of specific problems were reported. The overall assessment of student comprehension was judged to be “good”.

Reliability and validity Inconsistency rates between two questions in a single administration, which is used as a reliability measure, was low for all substances (0–1%) with the exception of inhalants (7%). The proportion of unanswered questions about different drugs varies between 0 and 3%. No figures are available for core, module and own questions but looking at the questionnaire as a whole, 2% of the questions were not answered. No student reported inconsistent answers to the questions of use in lifetime, last 12 months and last 30 days for alcohol consumption, being drunk, cannabis and inhalants. Seven percent answered that they would definitely not have admitted using cannabis and 12% gave the same answer for heroin. On the same question 37% answered that they had already said that they had used cannabis, which is marginally lower than the prevalence figure (39%). Of all students 16% answered that they had heard of the

Appendix I

dummy drug Relevin. However, only 0.6% said that they had used it.

Methodological considerations Since no sampling was done there are no sampling problems. The proportion of eliminated questionnaires was 3.6%. Even though this is not a relatively high figure as such, it is worth observing that it is one of the highest of all ESPAD countries. Eleven students (0.7%) refused to participate, which is a low figure as such, but is still rather high compared to other countries. The ESPAD classroom reports were not used. However, nearly no survey leaders reported any important disturbances during the data collection. As a whole, student as well as school co-operation seem to have been satisfactory. A rather significant number of students provided inconsistent answers for inhalants (7%) but not for other substances (0–1%). No inconsistencies were reported for lifetime, last 12 months and last 30 days prevalence on questions about alcohol consumption, being drunk, cannabis and inhalants.

The fact that no student showed any inconsistency on all four variables must be seen as rather extreme. The data processing was done by the UK ESPAD team under the same conditions as for the UK data. Hence, there is no reason to believe that there were any “technical problems” related to the formulation of these figures. No reliability or validity measures raise any important question marks, which indicate that the Isle of Man study has been done without any major methodological problems. No separate figures on the number of unanswered questions are available for the categories core, module and own questions. However, since only 2% of all questions were unanswered there is reason to believe that the figures would be any different for these “sub groups”. The Isle of Man ESPAD study seems to have been conducted without any important concerns. Reported data seem to be representative for all students born in 1987 and comparable with other ESPAD data.

Italy Dr. Fabio Mariani at the Institute of Clinical Physiology, Italian Research Council, Pisa was the principal co-ordinator for the Italian survey. Italy also participated in the 1995 and the 1999 ESPAD studies.

Population In Italy the ESPAD survey was conducted in the whole country: North, Centre, South and Islands. It covered all grades of high school from 1 to 5 (students aged 14 to 19). Only students attending the second grade were included in the analyses for the ESPAD project.

Sample and representativeness As in previous surveys, the Italian sample was drawn as a multistage stratified random sample. The stratification of the 103 Italian provinces was based on 3 variables: geographical area (north, centre, south and islands), population density and SMAD index, which is a drug abuse monitoring system that classifies the Italian provinces in relation to high, medium and low levels of drug use prevalence. The next stratum was created in relation to schools within each province type: Lyceums

Appendix I

(classic, scientific, linguistic, pedagogic), artistic institute and vocational institute (technical and professional). Finally, 1% of the classes in each school stratum were randomly (simple random) drawn. The artistic schools were oversampled (7%) in response to national interest. However, out of the number of schools initially drawn, 12 refused to participate and were replaced by randomly drawn schools. The size of schools was not considered for stratification as the Italian school system guarantees a rather homogeneous number of students per school and per class (average number of students per school is 500, and per class 25. In Italy, 93% of the 1987 birth cohort was present in high schools. In addition, analysis of distribution by geographical area by school and by sex did not show any anomalies in the selection factors; hence the sample was considered representative of the whole birth cohort.

Field procedure Contact was established via telephone with the health teacher or CIC staff (Consulting and Infor-

257

mation Centre for juvenile distress). If none of these were found, the school headmaster was contacted. Materials for the survey were mailed to the contact person in each school. Data was collected in the classroom in the presence of a teacher. Printed information for the survey leader (teacher) was provided, and he/she was advised to read aloud the instructions (same as on front page of the questionnaire) to the class. When the questionnaires were completed, each student put their questionnaire in a separate envelope and sealed it. The data collection leader sent the class envelope including the classroom report to the National Research Council. Data collection period was from the end of March until end of April 2003.

Questionnaire and data processing Almost all ESPAD questions, but no extra country specific questions were included in the Italian questionnaire. The questions 11 and 12 were excluded since cider or alcopops are not available in Italy. Question number 5 (average grade last term) was modified to better fit the Italian grading system. In addition, the drug Ketamin was added to the list of drugs, since it’s use has been recorded among Italian adolescents. No module questions were added to the Italian version of the questionnaire. The parcels with completed questionnaires were opened at the National Research Council. The questionnaires were scrutinised following a checklist for exclusion. As a result, 83 questionnaires were excluded from the analyses. Finally the data was entered into the computer, using the programme Filemaker 5.5. For the analyses SPSS 11.0 was used. The sample was considered to be self-weighted, except for the overrepresentation of artistic schools for which a weight was inserted into the data file.

School and student co-operation Of the 336 schools (and classes), which accepted to participate in the survey 324 sent back the questionnaires to the research institute. This implies a loss of 3.5% of the sample. Of the non-participating schools 5 did not do so because the assigned teacher failed to fulfil his/her task, 5 schools had technical problems within the schools and two because of loss of questionnaire within the postal services. No student refused to participate in the study. The teachers’ comments revealed that cooperation was excellent for the majority of the students. According to the classroom reports more than half of the teachers (56%) reported no disturbances

258

at all during completion of the questionnaire. Of those where some disturbances occurred the majority concerned giggles or eye makings to the classmates. A large majority of the survey leaders (94%) found that a majority of the students were interested in the survey and that they worked seriously (98%). The average time to complete the questionnaire was 40 minutes. The response rate was 98%.

Reliability and validity The inconsistency rate between two questions in a single administration was generally low and the highest was found in relation to the questions on drunkenness (6%), smoking use of cannabis and use of inhalants (5% each). Other variables with inconsistent answering pattern were use of tranquillisers or sedatives (4%), use of heroin (3%) and amphetamine or LSD use (2%). The differences for other variables were lower, around 1%. The missing data rate was also overall low, especially in relation to lifetime prevalence. For any alcohol and having been drunk it was 1% and for smoking cigarettes it was even lower (0%). However, for other variables related to illicit drug use it was somewhat higher (2% on average). As can be expected, the 12 months and 30 days prevalence of drinking alcohol or having been drunk showed an increasing rate of missing data, but still relatively low (3% on both variables for alcohol and 2% for been drunk). For use of cannabis or inhalants the missing data rose from 2 to 3% on both. The rates of inconsistent answering in relation to lifetime, last 12 months and 30 days prevalence was 5% for any alcohol and 3% for having been drunk, 1% for cannabis use and 0% for use of inhalants. The questions related to students’ willingness to admit drug use reveal that 21% answered that “I already said I that have used it”, compared to the prevalence rate of 27% for cannabis users. The proportion who answered that they would “definitely not” admit such use was not very high; 4%. The corresponding figure for heroin use was somewhat higher, 7%. Use of the dummy drug “relevin” was reported by 1%, while 11% thought that they had heard of it.

Methodological considerations The Italian sample was drawn in the same way as in earlier ESPAD surveys in Italy, which would appear to provide a representative sample of all types of high schools, in which the absolute major-

Appendix I

ity of the students born in 1987 are taught. According to the classroom reports the survey seems to have functioned very well in the Italian schools. The response rate seems to be unusually high, but an inquiry among the responsible Italian researchers confirm that this is often the situation in Italian schools at this time of the school year. The methodological measures such as inconsistency rates and missing data rates indicate no important problems. Not all cannabis users indicated

on the honesty question that they had used it, but on the other hand this is something that also has been observed in many other ESPAD surveys. However, very few answered that they definitely would not admit any such use (4%). The same figure related to heroin was only somewhat higher (7%), which is indicative of good quality data. Thus, the Italian survey seems to provide reliable and valid data.

Latvia Mrs. Ilze Koroleva, Institute of Philosophy and Sociology, University of Latvia was the principal co-ordinator for the Latvian ESPAD study. Latvia collected data also in the 1995 and 1999 ESPAD studies.

Population The target population consisted of all students born in 1987 in Latvian schools, including Russian-speaking students. In 2003 87% of young people born in 1987 were at school.

Sample and representativeness Two types of schools were represented in the study; one was comprehensive and the other vocational schools, all together 1,148 schools. Participating grades were 8–10 in comprehensive schools and grade 1 in vocational schools. Student attending evening sessions at comprehensive schools (0.6% of the birth cohort) and schools for students with serious disabilities (5%) were excluded from the sampling frame. A proportional stratified cluster sample was used. For each of the 4 participating grades the schools were stratified by five levels of urbanisation. For grades 8–10 in compulsory schools there was also a division in Latvian and non-Latvian speaking schools. Taken together this resulted in 33 strata. Sampling units were classes and in each stratum classes were drawn via a simple random sample proportionate to the size of each stratum. All together 436 classes in 351 schools were included in the sample. Data were weighted. The sample was judged to be representative for all students born in 1987.

Appendix I

Field procedure Principals in the sampled schools were contacted by telephone. They were informed on the objectives of the survey and asked to nominate a contact teacher. When more than one class was sampled in a school the contact teachers were asked to arrange the data collection on the same day in all classes. Research assistants administrated the data collection. One reason for such was that “students tend not to trust teachers on such sensitive issues”. The teacher who should have taught the class at the time of the data collection was present, but not active, in the classroom. This helped to avoid disturbances and made it easier to obtain consent from the schools. The questionnaires were answered in the classrooms under the same conditions as a written test. The students put their questionnaires in individual envelopes, which they sealed and were collected by the research assistants. The questionnaires and classroom reports were returned to the research institute where they were checked. Russian speaking students answered a questionnaire in Russian. All students in participating classes took part in the data collection. However, the analysis only includes students born in 1987. The average time to answer the questionnaire was 49 minutes. Data were collected in March, April and early May, which gave an average age of 15.8 years.

Questionnaire and data processing All ESPAD core questions were included as well as the modules of Integration and Psycho-social measures. Three own socio-demographic questions were added together with 12 drug related questions. Question 23 was incorrectly formulated and was excluded from the analysis. Some of the few added drug related answering categories will be com-

259

mented in asterisks of a few tables. A professional interpreter translated the new questions from English to Latvian and Russian. Since most questions had already been used in 1999, no translation – back-translation process was deemed to be necessary for the few new questions. No pre-test was done. No double entering of data was carried out. However, logical consistency checks were run and checked by going back to the original questionnaires.

students answered “definitely not” on the question “If you had used marijuana or hashish, do you think you would have said so in this questionnaire” (and the corresponding question about heroin). On the same question 16% answered that they had already said that they had used cannabis, which is the same figure as the lifetime frequency figure. Rather few (6%) reported that they had heard about the dummy drug relevin and only 0.1% answered that they had used it.

School and student co-operation

Methodological considerations

Of the 436 sampled classes 14 refused to participate. Information related to the student co-operation was based on the data from all 7,533 participating students, i.e. also those 4,697 students that were not born in 1987. In the participating classes 16% of the students were absent. No present student refused to participate in the study. The scrutinising process resulted in the exclusion of 88 (1.2%) questionnaires. Of the survey leaders, 67% did not report any disturbances and 27% that disturbances were found only among a few students. The most important disturbance was giggles or eye makings, which were reported by about one fifth (22%) of the data collection leaders and loud comments by 14%. Some survey leaders reported that the questionnaire was too repetitive. A large majority of the survey leaders (94%) reported that “all”, “nearly all” or “a majority” of the students were interested in the study (79% answered “all” or “nearly all” students). The corresponding figures were similar on the question whether the students worked seriously (95 and 79% respectively).

The sample was drawn as a proportional stratified simple random sample of classes and thus the risk for oversampling of small classes was inherent in the procedure. However, since separate samples were drawn in a large number of strata (33) and the sizes of the classes vary little within the strata, there is reason to believe that this issue did not cause any major sampling problems. As a whole the sampling procedure seems to have functioned well and the results are considered representative for Latvian students born in 1987. Data were collected by research assistants and not teachers unlike the exercise conducted in 1999. However, even though this is seen as an improvement, the effects of such are deemed not to be of sufficient magnitude to distort comparability between data from the 1999 and 2003 surveys. Only a few sampled classes (3%) did not take part in the survey, which is indicative of good school co-operation. No student refused to participate and the proportion of excluded questionnaires was acceptable (1.2%). Disturbances were reported from one third of the classes. Of all survey leaders 79% reported that “all” or “nearly all” students were interested in the survey and the proportion was the same on the question of whether or not the students worked seriously. Even though these figures are rather high they are a little lower than in most other countries. However, as a whole student co-operation seems to have been satisfactory. Rather many students (13%) however gave inconsistent answers to two questions in relation to drunkenness and for many variables the figures are slightly higher than in 1999. However, looking at all reliability and validity measures the survey seems to have been conducted without any major methodological problems. The overall impression is that the Latvian study has functioned pretty well and that data are comparable with data from other ESPAD countries.

Reliability and validity The inconsistency rates between two questions in a single administration were highest for the variable been drunk (13%). It was lower for cannabis, inhalants and cigarettes (5–7%) as well as for tranquillisers and even lower for other illicit drugs and anabolic steroids (1–3%). Missing data rates were low or very low for drug related questions (varying between 0 and 3%). In the questionnaire as a whole, the proportion of unanswered questions was low (2%). The rates of inconsistent answers to questions of use in lifetime, last 12 months and last 30 days were quite low; 2% on alcohol questions and 0–1% on the questions about cannabis and inhalants. For cannabis as well as heroin about 12% of the

260

Appendix I

Lithuania Dr. Aleksandra G Davidaviciene at the Education Development Centre, Ministry of Education and Science was responsible for the Lithuanian ESPAD 2003 study. Lithuania also participated in the 1995 and 1999 ESPAD studies.

Population The target population consisted of all students in Lithuania born in 1987. In the Spring of 2003 approximately 96% of the 1987 birth cohort was at school.

Sample and representativeness Students born in 1987 were found in grades 8–10 (or grades 1 and 2 in gymnasiums) of academic schools. The population of 1987 born students in grade 1 of vocational schools was so small (3%) that this school type was excluded from the sample. All schools in the country were stratified according to type of academic school (basic, secondary or gymnasium), teaching language (Lithuanian, Russian or Polish) and geographic location (urban or rural). The sample was a proportional stratified cluster sample. In each strata a systematic sample of classes was done. In the first step schools were selected and in the second one class per school was sampled. The only exception was five large schools from which two grade 9 classes were sampled. The sample was selfweighted. It was representative for all Lithuanian students born in 1987 (with the exception of the small proportion attending vocational schools).

Field procedure The headmaster of chosen schools were informed of the study. Data were collected by teachers under the same conditions as a written test. The students were informed according to the standard ESPAD instructions. Following completion students put their questionnaires in individual envelopes, which were returned to the research institute together with the classroom reports. In sampled classes in which more than half of the students were born in 1987 all students in the class answered a questionnaire. When less than half was born in 1987 (which usually was the case in grades 8 and 10) only students born in this year were asked to participate in the study. The average time to answer the questionnaire was 44 minutes. Data were collected in March and April, which

Appendix I

gave an estimated average age of 15.7 years.

Questionnaire and data processing All ESPAD questions were asked together with the Integration and Deviance modules. The questionnaire also contained one question from the module Psycho-social measures. No other questions were included. The questions that were new in 2003 were translated from English to Lithuanian and then back translated. Even though some schools teach in Russian or Polish all students answered a Lithuanian questionnaire (simply because this was preferred by the students). The questionnaire was not pretested. However, pretesting was conducted prior to the 1995 and 1999 surveys without any indication of any inherent problems. Data were not weighted.

School and student co-operation The school co-operation was very good. No schools or classes refused to participate. However, the questionnaires for one class were lost during transportation. No present student refused to answer the questionnaire. The response rate was 87%. Of the absent students about 70% were home because of illness. All together, 91% of the absent students were not at school because of sickness, authorised leave and other “acceptable reasons”. In the scrutinising process 451 questionnaires were rejected because the respondents were not born in 1987. Five questionnaires were eliminated for students belonging to the target group. A large majority of the data collection leaders (72%) did not report any disturbances during the data collection and another 24% answered that they only noticed disturbances from a few students. The most important disturbance was giggles or eye makings, which were reported from 17% of all classes. In nearly all participating classes (97%) the sur vey leaders reported that “all”, “nearly all” or “a majority” of the students were interested in the survey (87% answered “all” or “nearly all”). The figures were of the same magnitude on the similar question whether the students worked seriously (99 and 88% respectively).

Reliability and validity The inconsistency rate for two questions in a single administration was highest for the variables been

261

drunk (6%) and cigarettes (3%) while it was 0–2% for other drug variables. Missing data rates on some drug related questions were very low (0%) and the figure was the same for the questionnaire as a whole. The rate of inconsistent answers to questions about lifetime, last 12 months and last 30 days was low for all four variables (0–1%). For both cannabis and heroin 10% of the students answered that they would definitely not have admitted possible use. On the same question 11% of the students answered that they already had said that they had used cannabis. This figure is a bit lower than the answer to the lifetime prevalence question (14%). Only a few students (0.2%) reported that they had heard about the dummy drug relevin and nearly no one answered that they had used it.

Methodological considerations The sampling procedure functioned well. No schools, classes or students refused to participate. No major problems were reported in the data collection and the same may be said about the reliability and validity measures. The only measure for which a figure was a little high was about the unwillingness to admit cannabis use. The figure (10%) is higher than in many other countries but not extremely high. It is also worth noting that the corresponding figure was even higher in ESPAD 95 and 99. The Lithuanian study seems to have been conducted without any significant methodological problems. Data seem to be representative for Lithuanian students born in 1987 and comparable with the results from other ESPAD countries.

Malta Sedqa – agency against alcohol and drug abuse – Malta, in collaboration with the Guidance and Counselling Services, Department of Education Malta were responsible for the Maltese study. Malta also participated in the 1995 and the 1999 ESPAD surveys.

Population The total population of the 1987 born students participated in the survey. They attended one of the three types of schools: General Secondary Schools, Junior Lyceum and General Schools. There were 65 such schools comprising of 49 General Secondary, 11 Junior Lyceums and 5 General schools.

Sample and representativeness A class list was collected from all three different types of schools that cater for students born in 1987. As the total number of students born in 1987 was approximately 5,600 and most of them were in the fifth grade (or equivalent) half of them were needed for the ESPAD project sample. However, given that the total number of students was below 10,000 it has been suggested that in such cases a total population survey would be advisable since complexities involved with sampling would far outweigh those related to logistics. Therefore, total population sampling was adopted for the 2003 ES-

262

PAD survey, as has also been the case in 1995 and 1999 surveys.

Field procedure First contact with every school was made via a formal letter from the Guidance and Counselling Services of the Department of Education. Following a briefing meeting with guidance teachers and counsellors, a final meeting was held prior to the actual survey between the school co-ordinators and teachers who supervised the participating students in their respective classes. Since school for fifth formers normally finishes earlier to allow ample time for students to study and prepare themselves before sitting for their MATSEC examinations (equivalent to Ordinary Level Examinations), the Maltese survey was conducted earlier than in other countries. The main reason was the positive results achieved in the response rate of the 1999 study when compared to the 1995 survey, which was conducted on the same day as in other participating countries. The questionnaires were sealed in packs and numbered appropriately. They were distributed to all co-ordinators of each school one-day prior to the survey. The time allotted for the completion of the questionnaire was mid-morning in order to include any latecomers. Teachers compiled the class report

Appendix I

data. When the students had completed the questionnaire, each student placed the questionnaire on a table at the far end of the room face down. A students’ representative placed the questionnaires in an envelope provided and sealed it together with the class report and thereafter deposited the sealed pack at the office of the head of the school, for transportation to the team of researchers. The survey was conducted during one day in all schools: January 22. This means that the average age of the Maltese students was 15.6 years.

Questionnaire and data processing All core segments of the questionnaire were included in the Maltese version except for questions on magic mushrooms and cider that were omitted. As regards the optional segments, 2 of the 5 questions of the Psychosocial module and all questions related to the Deviance module were included. The questionnaire was translated into Maltese and then translated back to English by another researcher from the collaborating consortium. The two English versions were subsequently compared and a final Maltese questionnaire (and an English one for non-Maltese speaking) was concluded. Before processing the data, all questionnaires were scrutinised and 15 were removed due to what seemed to be invalid data.

School and student co-operation All schools and classes participated with the exception of one school with three classes. The refusal was due to a significant number of students with learning difficulties (illiteracy) and thus the time allotted for the questionnaire was deemed too short. In 83% of the classes no disturbances were observed and where any disturbance was reported it regarded solely giggles and eye contacts. The majority of the students showed interest in the study. A very small number of classes reported lack of interest, mainly due to the length of the questionnaire and some problems with technical words that were not understood by the students. Almost all classes reported that the vast majority worked seriously. In a suggestion that was forwarded by a teacher, the use of pictures to indicate alcohol measurement was recommended so that students would clearly understand the quantities of alcohol in question. The response rate was 81%.

Appendix I

Reliability and validity The reliability as measured by the inconsistency rate between two questions in a single administration was rather good, highest for inhalants (10%) and “been drunk” (7%), while it was lower for cigarettes (3%), cannabis (2%), tranquillisers (2%) and anabolic steroids (1%). The missing data rate on drug questions was low. For the lifetime variables it was highest for alcohol (3%), while for all other variables it was 1% or less. However, for alcohol consumption and “been drunk” it was higher for 12 months and 30 days prevalence (about 4%). The rate of inconsistent answers between lifetime, 12 months and 30 days prevalence questions was rather low; 5% for any alcohol, 3% for “been drunk” and 1% for cannabis and inhalants use. As for the “honesty questions” regarding admitting the use of cannabis or heroin, 13% of the students answered that they would definitely not admit if they had used cannabis and a few more (15%) wouldn’t admit heroin use. On the other hand the lifetime prevalence figures for cannabis and heroin use denotes the same proportions indicated in the question where students answered “I already said that I have used it” (10% and 2% respectively).

Methodological considerations The Maltese study was done in the same way as earlier studies within the ESPAD project. Since the island is rather small, as already indicated earlier a total survey was considered the best option. The implementation of the survey seems to have been successful and very few disturbances were reported from the classrooms. The methodological measures such as inconsistencies between two questions in a single administration and inconsistencies between lifetime, 12 months and 30 days prevalence show very low figures. However, on the honesty questions a number of students indicated that they were reluctant to reveal the use both of cannabis and heroin (13 and 15% respectively), which might suggest the possibility for underreporting. However, this is contradicted by the very fact that exactly the same proportions that had said earlier in the questionnaire that they had used cannabis and heroin answered “I already said that I have used it”. The overall assessment of the Maltese study is that it provides reliable and valid data.

263

The Netherlands The Dutch ESPAD study was conducted by Karin Monshouwer and Saskia van Dorsselaer for the Trimbos Institute. The Netherlands also participated in the 1999 ESPAD data collection exercise. However, for methodological reasons the 1999 data from the Netherlands were not considered to be directly comparable with those from other ESPAD countries. Hence, data from the Netherlands were presented separately in the result tables of the 1999 ESPAD report.

Population The population consists of all students in grades 3 and 4 of regular secondary education born between August 1, 1987 and July 31, 1998. The reason for this particular choice of the target population, which differs from the one used in other ESPAD countries, is that the data collection in the Netherlands was done in October–November, i.e. about 6–7 months later than in most other countries. The redefinition of the target population results in an average age of the Dutch ESPAD students (15.7 years) which however is similar to the average age in a large majority of the ESPAD countries. It has been calculated that about 92% of persons born between August 1, 1987 – July 31, 1988 attended a Dutch school at the time of the data collection.

Sample and representativeness Schools were stratified in four strata according to the level of urbanisation. In proportion to the size of each stratum, schools were sampled randomly via a systematic sample from a list of all schools in each strata. Every fourth school was assigned as a school where a third grade class should be sampled. In all remaining schools a fourth grade class should be sampled. Of all students in the target population 92% were estimated to be found in these two grades. Schools that agreed to participate in the study sent lists of all grade 3 or 4 classes. These lists were used to draw a sample of one class per school. The sample is judged to be nationally representative for all secondary school students born between August 1, 1987 and July 31, 1988. Data were weighted on age, gender, grade and school level.

Field procedure The data collection was lead by staff members from Regional Health Services, research assistants

264

and researchers from the Trimbos Institute, all together 29 people. All survey leaders received a half day training session prior to the survey. The material was sent to the Regional Health Services and research assistants. For each class there was an envelope with questionnaires, a written instruction for the data collection leader and a classroom report. The teachers were asked to leave the room or to take a place in the back of the room during the data collection. After completion, the questionnaires of all students were put in a large class envelope together with the classroom report. The envelopes were sent to the data-entry service. Data were collected in October and November, which gave an average age of 15.8 years. The average time to complete the questionnaire was 31 minutes.

Questionnaire and data processing The Dutch questionnaire included all ESPAD core questions with the exception of the consumption of cider (Q11) (since cider is not a popular beverage). In addition to this four new questions were included. Three questions were culturally adjusted to such a degree that might limit the provision of comparability with data from other ESPAD countries. First: in Q9c in which it was stated that “spirits” did not include pre-mixed drinks. Second: NSTC was used as a dummy drug in Q27 and Q28 (instead of relevin). Third: In Q33 “coffee shop” was added as a possible place to buy cannabis. The ESPAD questionnaire was translated from English to Dutch and then back translated by another interpreter. The questionnaire was pre-tested in three classes which resulted in some minor adjustments in the wording of some questions.

School and student co-operation Out of the 268 sampled schools 76 (28%) did not participate. In the remaining schools data were collected from 189 of the 192 sampled classes. Participating and non-participating schools were compared for school size and proportion of immigrant students (students born in a foreign country or who had one or both parents born outside the Netherlands). No significant differences were found. No present student refused to participate. The response rate for all students in participating classes was 93%. Ten questionnaires (0.5%) were eliminated

Appendix I

following the scrutinising process. 19% of the survey leaders reported some kind of disturbances during completion of the forms. The most common disturbance was “other kinds of comments”, which was reported by 18% of the data collection leaders. The question about students interest in the survey was not asked. However, in all participating classes it was reported that “all”, “nearly all” or “a majority” of the students worked seriously (96% answered “all” or “nearly all”). At an evaluating meeting with all survey leaders no major difficulties were reported in the data collection procedure.

Reliability and validity The inconsistency between two questions in a single administration, which is a reliability measure, was not extremely high for any variable. The highest was found for the variables tranquillisers or sedatives, been drunk and inhalants (4–6%), while the figures were lower for other substances (0–3%). The inconsistency rate for the variables been drunk and alcohol consumption were about 2%, while it was lower for cannabis and inhalants (0%). Six per cent of all students indicated that they would definitely not have admitted cannabis use and about 9% gave the same answer on reporting possible heroin use. On the question about willingness to admit drug use, 23% answered that they had already reported that they had used cannabis, which is a little lower than the prevalence figure (28%). Of all the students, 13% answered that they had heard of the dummy drug NSTC. However, only 0.9% said that they had used it.

Methodological considerations For pragmatic and historical reasons the data collection in the Netherlands took place 6–7 months later (October–November) than in other ESPAD countries (in which data were collected during the winter and spring). To “compensate” for this the target population was redefined as students born between August 1, 1987 and July 31, 1988. This results in an average age of 15.7 years, which is the same as that found in most other ESPAD countries. The situation was similar to that in the 1999 data collection exercise in which it was possible to compare the results from students defined in a similar

Appendix I

way with students defined according to the ESPAD protocol (Hibell et al. 2000). There were only some minor differences between the two groups and they were all in the expected direction. The conclusion drawn was that the definition used in the Dutch study seemed to be the most appropriate for ESPAD comparisons. It seems relevant to make the same assumption with respect to the 2003 survey. The sample of schools seem to have been done adequately even though it probably gave an overrepresentation of small schools. However, this was compensated for in the weighting process. The sample is judged to be nationally representative for secondary school students born between August 1, 1987 and July 31, 1988. Of the sampled schools 76 out of 268 (28%) did not want to participate. This is rather high compared to other ESPAD countries. A comparison between participating and non-participating schools did not show any differences for the variables school size and proportion of immigrant students. The Dutch researchers explained that compared to similar school surveys in the Netherlands the response was very good. Also there appear to be sufficient grounds to assume that the relatively high number of non-participating schools did not influence the results to such a degree that the comparability with other ESPAD countries is not warranted. Three questions in the Dutch questionnaire were culturally adjusted. In the few cases where this might pose difficulties with the ability to make comparisons with other ESPAD countries, this factor is addressed in the result section. Student co-operation would appear to be good. No student refused to participate and only a few questionnaires were omitted. The data collection seems to have functioned without any major obstacles. No reliability or validity measures indicate any important methodological problems. As a whole, data from the Dutch survey seem to be comparable with data from other ESPAD countries. However, it might be worth keeping in mind that the data collection was done at a different time of the year, that the target population is defined differently (even though the mean age is about the same) and that relatively many schools did not want to participate in the survey.

265

Norway Astrid Skretting at the Norwegian National Institute for Alcohol and Drug Research was the principal investigator for the Norwegian study. Norway also participated in the 1995 and the 1999 ESPAD surveys.

Population The target population consisted of all students in grade ten in secondary (compulsory) schools in Norway born in 1987. Nearly 100% of children born in 1987 were enrolled in school in March 2003. Nearly all of them were to be found in grade 10.

Sample and representativeness The sampling frame was all 2,525 grade 10 classes in Norway. They were divided into 87 strata according to a combination of county and a form of municipality. In the stratified cluster sample, classes were sampled with a simple random sampling technique within each stratum proportionate to the size of the stratum. The sample consisted of 265 classes. Since the sample of classes within each stratum was not proportionate to class size, students in small classes in some cases may have been overrepresented in the sample. However, it should be borne in mind that class size does not vary to any great extent within each stratum. The sample is estimated to be a representative nation-wide sample of students born in 1987 attending grade 10.

Field procedure Via letters taken home by the students, parents were informed in advance on the conduct of the study and thus had the possibility through which to prevent their child from participating in the said survey. However, very few students did not participate as a result of parent refusal. The questionnaires and instructions were sent to the sampled schools. The data collection was done under the same conditions as a written test and the completed questionnaires were collected in individual envelopes by a teacher, who then sent them back to the institute responsible for the conduct of the study. The questionnaires were scanned into a computer. The average time to complete the questionnaire was 36 minutes. Data were collected in March– April, which gave an average age of 15.7 years. The questionnaires of the few grade 10 students that were not born in 1987 were excluded from the survey.

266

Questionnaire and data processing All core questions in the ESPAD questionnaire were asked as well as the questions that formed part of the Integration and Deviance modules. A few own questions about alcopops and spirits were also included in the questionnaire. The questionnaire was translated by the Norwegian ESPAD researcher. It was not translated back and was not piloted. Data are weighted for geographical distribution.

School and student co-operation Of the 265 sampled classes 60 did not participate in the survey and they were not replaced. The proportion of non-participating classes (23%) was a bit higher than it was in the 1999 data collection (14%). The Norwegian ESPAD researcher expressed the view that the increase was mainly caused by the significant number of requests to schools to participate in school surveys. Hence, it was judged that students in non-participating classes do not differ significantly from participating students in regards to their alcohol and drug habits. The response rate was 87%. Explicit information on the number of students that refused to participate was not available. However, no data collection leader reported any refusals. Very few questionnaires were excluded following the scrutinising process (0.3%). Of the 205 survey leaders, 150 returned a classroom report. About four fifths (81%) did not report any disturbances, while 18% answered that this was the case with only a few students. The most common disturbance was giggles or eye makings that was mentioned by 10% of the teachers. In the vast majority of the classroom reports (96%) it was mentioned that “all”, “nearly all” or “a majority” of the students were interested (89% answered “all” or “nearly all”). The figures were about the same on the similar question whether the students worked seriously (99 and 93% respectively).

Reliability and validity Reliability as measured by consistency between two questions within a single administration showed that the rate of inconsistency was highest for cigarette smoking (5%). For questions about alcohol, inhalants and illicit drugs the inconsistency rate were smaller (0–3%). Missing data rates on drug questions varied be-

Appendix I

tween 4 and 7%, with the exception of cigarettes where the figure was lower (1%). Looking at the questionnaire as a whole, 3% of the questions were not answered. The rates of inconsistent answers to questions about lifetime, last 12 months and last 30 days were low for all variables (0–1%). The proportion who would definitely not admit cannabis use was 3% and the same was true for heroin. The proportion who answered in the affirmative “I already said that I have used it” (i.e. cannabis) was 9%, which is the same as the prevalence figure. Eleven per cent of the Norwegian students answered that they had heard of the dummy drug relevin. However, only 0.4% said that they had used it.

Methodological considerations Within each stratum classes were drawn with the same probability, which could have resulted in an overrepresentation of students from small classes. However, since class size within each of the 87 strata did not differ very much this was judged to have negligible impact on the representativeness of the sample. Hence, the sample is considered to be representative of students born in 1987 attending grade 10. The parents were informed about the study in advance, which may have created the opportunity for discussion prior to data collection either between the students or at home between parents and the students. If such discussions occurred one cannot exclude that they may have negatively influenced the willingness to give true answers. However, since the study was done anonymously and since there was no information available from the

data collection leaders that the validity might have been negatively affected, it seems reasonable to assume that contact with the parents did not have any adverse consequences on the outcome of the study and thus comparisons with other ESPAD countries are acceptable. Compared to other ESPAD countries a significant number of the sampled classes (23%) did not participate in the data collection. They are spread all over the country and there are no indications that students in non-participating classes can be expected to have significantly different alcohol and drug habits. However, it must be noted that this conclusion is not based on any systematic follow up. The response rate was acceptable (87%), the proportion of unanswered questions low (0.3%) and the classroom reports did not indicate any important disturbances during the data collection. Hence, student co-operation seems to be satisfactory. The proportion of unanswered questions in relation to illegal substances (4–7%) is higher than in most other ESPAD countries, which could be seen as an indicator of underreporting. However, the proportion that definitely not would have answered honestly about possible cannabis use is among the lowest (3%), so there are no clear foundations for such a conclusion. Moreover, the Norwegian figures for the reliability and validity measures do not infer any major methodological problems. As a whole the results seem to be representative and comparable with other ESPAD data. However, the rather high proportion of non-participating classes is an uncertainty that should not be dismissed.

Poland Janusz Sieroslawski, Institute of Psychiatry and Neurology, Warsaw was responsible for the Polish study. Poland also participated in the 1995 and 1999 ESPAD studies.

Population The population consists of students born in 1987 attending third grade of the gymnasium. It was assumed that 95% of this age cohort were enrolled in school in March/April 2003.

Appendix I

Sample and representativeness List of schools were obtained from the Ministry of Education. They contained information about the number of classes in each school. The sampling unit was class. The sampling frame constituted of lists where the name of the schools appeared as many times as the number of classes within each school. The sample was drawn as a systematic random sample with a probability proportionate to school size. In addition, extra classes were drawn from two cities (Warsaw and Pozan)

267

and three regions (Mazowieckie, Lodzkie and Zachodniopomorskie) that wanted to have data for their own cities and regions. For this reason data were weighted. 390 classes were sampled, with one each in 390 schools. Of all students born in 1987 92% were estimated to attend grade 3 in the gymnasium. The sample is judged to be representative for all Polish students born in 1987.

Field procedure For the administration of the data collection Poland was divided into six areas. Administration and data collection were performed by all together 124 research assistants, who were specially trained for this task. The assistants were told to collect data under conditions similar to a written test. Instructions to the students were read aloud in each class and each student could also read it before answering the questionnaire. After completion each student put his or her questionnaire in an individual envelope. No teacher was allowed to stay in the classroom while the survey was done. All material was taken to the research institute by the research assistants. The average time to answer the questionnaire was 37 minutes. Data were collected in May–June, which gives an average age of 15.9 years.

Questionnaire and data processing The questions that were new in 2003 were translated to Polish and then back-translated to English, which did not result in any important changes. The questionnaire contained all ESPAD core questions as well as questions of the Integration module. The same own questions were asked as in the 1995 and 1999 surveys. The questionnaire also included one new question. The questionnaire was tested via interviews with six students, which did not indicate any problems in understanding the questions Data were weighted to correct for the oversampling of some cities and regions.

students (0.1%) refused to answer the questionnaire. The number of eliminated questionnaires was 52 (0.9%). No serious problems or disturbances were reported form the data collection. Of all survey leaders 54% did not report any disturbances at all, while 36% answered that this happened with a few students only. The most important disturbance was loud comments, which was reported from nearly half of the survey leaders (49%). In a large majority of the classes (90%) the data collection leaders reported that “all”, “nearly all” or “a majority” of the students were interested in the study (81% answered “all” or “nearly all”). The proportions that answered that the students worked seriously were 92 and 74% respectively.

Reliability and validity The inconsistency rate between two questions in a single administration was highest for the variables been drunk, tranquillisers, cigarettes and inhalants (6–8%) followed by cannabis (4%). The corresponding figure was lower for anabolic steroids and other illicit drugs (1%). Missing data rates were rather low (1–2%) for all categories of substance use variables. No information is available about the proportions of unanswered questions in the questionnaire as a whole. The rate of inconsistent answers to questions about use in lifetime, last 12 months and last 30 days were 5–6% for all four drug related variables. For cannabis 8% answered “definitely not” on the question “If you had used marihuana and hashish, do you think that you would have said so in the questionnaire?” The corresponding figure for heroin was about the same (10%). On the “honesty question” 28% answered that they had already said that they had used cannabis, which is higher than the reported proportion (18%). Twelve percent answered that they had heard about the dummy drug relevin, while 1.0% said that they had used it.

Methodological considerations School and student co-operation Only six out of 390 schools did not participate. The major reason was that it was not possible to collect data during the time of the data collection. The six schools that did not collect data were not replaced. It is stressed in the national report that there were no problems with the willingness of the schools and classes to conduct the survey. The response rate was 85%. Only five present

268

The sample seems to have been done without any problems. There are rather many survey leaders that reported some kind of disturbance during the data collection. A plausible explanation to this from the Polish ESPAD researcher is that the research assistants were trained to note all disturbances, which made them very observant. It was also commented that the survey leaders were trained to handle situ-

Appendix I

ations with loud comments from the students. Hence, there is reason to assume that the disturbances during the data collection were not more serious in Poland than in other ESPAD countries. Very few students refused to participate, the proportion of skipped questionnaires was not high and the response rate acceptable. The reports of the survey leaders don’t indicate any serious problems during the data collection. Hence, the student cooperation seems to have been satisfying. The number of refusing schools and classes was low and there are no problems reported in the co-operation with the schools. Thus, there is reason to assume that the school co-operation was good. The inconsistency rates are a little higher in Poland than in most other ESPAD countries, especially for the variable tranquillisers and sedatives without a doctor’s prescription, which call for some