Risk Factors of Falling among the Elderly in Taiwan: A Longitudinal

Risk Factors of Falling among the Elderly in Taiwan: A Longitudinal

[Original Vol.3 No.2Article] Hui-Chuan Hsu et al Taiwan Geriatrics & Gerontology Risk Factors of Falling among the Elderly in Taiwan: A Longitudina...

324KB Sizes 0 Downloads 5 Views

[Original Vol.3 No.2Article]

Hui-Chuan Hsu et al

Taiwan Geriatrics & Gerontology

Risk Factors of Falling among the Elderly in Taiwan: A Longitudinal Study Hui-Chuan Hsu1, Li-Jyun Jhan2 Abstract Objectives: This study aimed to analyze the prevalence of falls and repeated falls of national elderly samples and to explore the risk factors of falls. Methods: Data were obtained from the “Survey of Health and Living Status of the Elderly in Taiwan” conducted in 1996 (n=2,669) and 1999 (n=2,310). The related risk factors of falls included demographic variables, chronic diseases, disability and changes in physical function, depressive symptoms, exercise behavior, and use of assistive devices (walker/sticks or glasses). Logistic regression was used for analysis. Results: The prevalence of falls was 19.5%, with 12.3% single fall and 7.2% repeated falls. The risk factors of falls included being female, having disability, reduced ADL function, with depressive symptoms, using stick/walker but walking well, and not wearing glasses but not seeing clearly. Among the fallers, 73.8% reported fear of falls, and 45.1% attributed their falls to environmental causes. Conclusion: Physical and psychological factors as well as environmental factors may cause falls and injuries for elderly. Physicians in the fields of geriatrics or family medicine should pay more and closer attention to the possible risks of their senior patients; home safety and appropriateness of assistive devices for the elderly should be regularly evaluated and carefully maintained. (Taiwan Geriatrics & Gerontology 2008;3(2):141-154) Key words: falls, elderly, longitudinal study, risk factors


Department of Healthcare Administration, Asia University; 2Central-South Regional Alliance of DOH Hospital, Executive Yuan. Correspondence to: Hui-Chuan Hsu, Associate Professor, Institute of Healthcare Administration, Asia University. 500, Lioufeng Road, Wufeng Township, Taichung, 413, Taiwan. Email: [email protected]


台灣老年醫學 暨老年學雜誌

Risk factors of falls

第 3 卷第 2 期

Previous fall experience constitutes the risk of repeated falls [7,11,31-33].


Some studies focused on the Accidents and their adverse effects

protective factors of falls. Exercise may

constituted the fifth cause of death in

reduce the incidence of falls [2,14,25].

2005, and accidental falls ranked second

More family networks contribute to lower

among accident events [1]. The prevalence

incidence of falls for older community

of falls was 17.2-29.3% for male and

women [34].

13.4-50.8% for female [2-11]. Falls is also

Although many studies have explored

a major health risk for the elderly. Past

the risk factors of falls, studies of

research has found that the fall mechanism

longitudinal risk factors with national

and injury severity were related to the

elderly samples were scare. Furthermore,

impact of falls. Falls happened outdoors,

the personal risk factors are usually

the vertical and horizontal displacement of

difficult to change in the old age, but

body’s center-of-gravity before falls, as

environmental factors and appropriate use

well as forward and sideways falls result in

of assistive devices (such as glasses or

more severe injuries [12]. Older age, poor

sticks) are changeable relatively. How

balance and gait were related to inability

much falls related to such changeable

to get up after falls, and these fallers were

factors have not been well explored. In this

more likely to be hospitalized or

research, we analyzed the prevalence of

institutionalized in nursing homes, or

falls and repeated falls using longitudinal

having functional limitation longer [13].

data of Taiwan representative elderly

Risk factors to falls for the elderly

samples and explored risk factors of falls.

include advanced age, gender (female),

We hoped to provide more evidence

lower educational level, living alone,

concerning risk factors of falls among

arthritis, rheumatic disorders, stroke,

elderly and further contribute to falls

diabetes, foot ulcers, vision or hearing


problem, dizziness, gait balance, orthostatic hypotension, physical function,


urinary incontinence, depressive symptoms, cognitive impairment, fear of

Data and samples

falls, sleep disturbance, use of certain medications, low fall efficacy, and

Data were obtained from the “Survey

sedentary lifestyle [2, 7, 9, 14-30].

of Health and Living Status of the Elderly 142

Vol.3 No.2

Hui-Chuan Hsu et al

Taiwan Geriatrics & Gerontology

in Taiwan”, which was first launched in

caused injuries or not), and the times of

1989. Face-to-face interviews were

falls in the past year. “Repeated falls” was

conducted with a random sample of

defined as more than two falls. In 1999,

individuals (aged ≥ 60 in 1989) derived

those elderly who had fall experience were

from the entire Taiwanese elderly

further asked if they were afraid of falls

population, including those in institutions.

and if they would reduce going out owing

A three-stage proportional-to-size

to fear of falls. They were also asked to

probability sampling technique was used.

evaluate if their most severe fall was due

The first stage consisted of a stratified

to environmental factors or personal

sample of the administrative units

factors (such as dizzy, chest pain, frailty,

(townships); the second consisted of


blocks in the selected townships; and the

Health status examined included the

third consisted of two respondents selected

following. (1) Chronic diseases: The

systematically from the register in each

participants self-reported if they have the

selected neighbourhood. The initial

following diseases: hypertension, diabetes,

interviews were followed up by subsequent

heart disease, stroke, cancer, bronchitis,

face-to-face interviews in 1993, 1996, and

arthritis, ulcer, liver disease, cataract,

1999. The lost cases were due to death or

glaucoma, TB, kidney disease, gout,

loss of follow-up. In this paper, we only

anemia, hip fracture, other fractures, or

used the data of 1996 (n = 2,669) and 1999

major injuries. The prevalence of different

(n = 2,310). Incomplete cases were not

chronic diseases was calculated using the

analyzed. According to the results of the

1996 data. (2) Disability: Disability was

goodness-of-fit test of gender and age, the

measured by activities of daily living

lost cases were mostly of more advanced

(ADL) and instrumental activities of daily

age and of males.

living (IADL) in 1996 and 1999. ADL disability indicates any difficulties in


taking bath, dressing, eating, transferring, walking indoors, or going to toilet, with

The dependent variable “falls” was

the difficulty lasting for at least 3 months.

defined as any experience of falls in the

IADL disability means any difficulties in

past year (including any falls, such as falls

doing groceries, money management,

when walking, slipped, falls due to did not

taking public transportation alone, heavy

sit/stand well, falls due to dizzy, or falls

housework, light housework, or making

when lying down from bed, whether

phone calls for at least 3 months. Changes 143

台灣老年醫學 暨老年學雜誌

Risk factors of falls

in ADL/IADL were measured by

第 3 卷第 2 期

very well to very unwell).

comparing the disability status at two

Demographic variables included age,

waves and classified as ‘improved or no

gender, educational level (illiterate,

change’ and ‘worse’. (3) Depressive

elementary school, junior high school or

symptoms in 1996 and 1999 were

higher), marital status (single/married),

measured by the short version of the

and living arrangement (living with others

Center for Epidemiologic Study

or alone).

Depression Scale (CES-D) [35], and the Analysis

cut-point of having depressive symptoms was decided by using T-score

We used descriptive statistics,

transformation [36]. Changes in depressive symptoms were measured by comparing

Chi-square test, and logistic regression

the depressive symptoms at two waves,

models for analysis in this study. Logistic

and classified ‘improved or no change’ and

regression analysis was used to explore the


odds ratio of related factors to the risk of falls. The dependent variable was any falls

Health behavior and using assistive devices examined included the following.

happened, and the independent variables

(1) Exercise behavior was defined as doing

included the demographics, chronic

exercise 3 times a week and 30 minutes or

diseases, ADL/IADL disability and

more each time, categorized as yes/no.

change, depressive symptoms and change,

Exercise behavior was measured in 1996

using glasses or sticks/walkers and if they

and 1999. Changes in exercise behavior

used it well. The chronic diseases which

were measured by comparing the exercise

were significantly related to falls would be

behavior at two waves and categorized as

used in the full model of logistic

‘improved or no change’ and ‘worse’. (2)

regression analysis. In order to reduce the

Use of assistive devices in 1999: The

full model, we also used stepwise logistic

participants were asked if they wore

regression analysis by setting alpha of

glasses (including reading glasses and

0.05. Only the coefficients of retained

contact lens) or used walkers/sticks.

variables were reported in the reduce

Whether they wore glasses or used


walkers/sticks or not, all the participants Results

were also asked if they could see clearly (five degrees from very clear to very

Demographic characteristics for the

unclear) or walk well (five degrees from 144

Vol.3 No.2

Hui-Chuan Hsu et al

Taiwan Geriatrics & Gerontology

samples at baseline (in 1996) are displayed

symptoms. Among the participants, 36.2%

in Table 1. Table 2 shows the health status

had exercise and about 15.2% reported

and health behaviors, including the status

their exercise frequency reduced.

at 1996 and the change from 1996 to 1999.

Regardless whether they wore glasses or

During the three years, the health status

used sticks/walkers, 24.2% reported they

declined rapidly: 10.2% ADLs, 17.3%

could not see clearly, and 19.5% had

IADLs and 18.8% with more depressive

walking trouble. In 1999, of the 2,157 participants, 420 (19.5%) had falls experience, 265 single

Table 1. Demographic characteristics of the sample at baseline (1996) Characteristics Age 67-74 75-84 85+ Gender Male Female Education Illiterate Elementary school Junior high school + Marriage No spouse Have spouse Living arrangement Live with others Living alone

fall and 155 repeated falls. Table 3 shows


the falling characteristics of the fallers.

63.1 32.0 4.9

Among those who had fall experience,

55.9 44.1

reduced their outdoor activities and 45.1%

39.3 41.2 19.5

caused by environment factors but not

73.8% reported fear of falls, 65.5% perceived that their most severe fall was personal factors. Table 4 shows the odd ratios of risk

41.3 58.7

factors to falls for the elderly after three

88.6 11.4

years. In the full model (including all the

Table 2. Health status and health behavior of the samples in 1996 and the change 1996-1999 Health status and behavior % Health status and behavior % ADL disability at baseline 6.9 Wearing glasses 1999 47.7 normal 93.1 No glasses 52.3 ADL worse 1996-1999 10.2 Using stick or walker 1999 18.7 No change or improved 89.8 No use 81.3 IADL disability 48.1 Seeing clear Normal 51.9 Unclear, no glasses 17.6 IADL worse 1996-1999 17.3 Unclear, wearing glasses 6.6 No change or improved 82.7 Clear, wearing glasses 41.1 Depressive symptoms 42.8 Clear, no glasses 34.7 No depressive symptoms 57.2 Walking well Depressive symptom worse 18.8 Walking trouble, no stick 10.2 No change or improved 81.2 Walking trouble, use stick, 9.3 Exercise behavior 36.2 Walking well, use stick, 10.3 No exercise 63.8 Walking well, no stick, 70.2 Exercise behavior reduced 15.2 No change or improved 84.8 Note: N=2,669 in 1996 and 2,302 in 1999. The samples in both 2 waves were included in change items.


台灣老年醫學 暨老年學雜誌

Risk factors of falls

第 3 卷第 2 期

Table 3. Reducing characteristics of the falling in 1999 Persons (%) Fallers 420 (100.0) Falling times Single fall 265 (63.1%) Repeated falls 155 (36.9%) Fears of falls # No fear 109 (26.2) Afraid 307 (73.8) Reduce outdoor activities because fear of falls No 107 (35.0) Yes 199 (65.0) Self-reported falls due to environments# No 246 (54.9) Yes 202 (45.1) Note: # Only falling samples answered the question of fear of falls and environment risks, and for those reported fears of falls were asked about if reducing going out.

independent variables), the elderly who

and IADLs disability, reduced ADLs

had ADLs disability at baseline, IADLs

function, having depressive symptoms, and

disability at baseline, worse ADLs, using

using sticks/walkers. The elders using

sticks or walkers, not wearing glasses but

sticks/walkers but walking well, and those

not seeing clearly, and using sticks/walkers

not wearing glasses but not seeing clearly

and walking well, would have higher risk

also had higher risk of falls. Among the

to fall after 3 years. In the reduced model

fallers, 73.8% reported fear of falls, and

by stepwise logistic regression, the risk

45.1% attributed their falls to

factors included being female, ADLs

environmental causes.

disability at baseline, IADLs disability at

Disability and changes in physical

baseline, worse ADLs, having depressive

function were found to be related to the

symptoms at baseline, and using sticks or

incidence of falls. This finding was

walkers, were more likely to fall after 3

consistent with results of previous studies


[18,32]. In this longitudinal study we added the changes in physical function and found that dysfunction constituting higher


risk of falls. If physical function of the In this study of national elderly

elderly declines rapidly, that may indicate

samples in Taiwan, the prevalence of falls

risk of falls increasing. However, the

was 19.5%, with 12.3% of single fall and

number of ADLs disability can not

7.2% of repeated falls. The risk factors of

accurately reflect the dynamic change of

falls included being female, having ADLs

physical function in our analysis. One may 146

Vol.3 No.2

Hui-Chuan Hsu et al

Taiwan Geriatrics & Gerontology

Table 4. Risk of falls of the elderly by logistic regression Variables Age- 75-84 years old Age- 85 years old+ Gender- Female Education- elementary school Education- primary high school or higher Marital status—having spouse Hypertension Diabetes Heart disease Stroke Arthritis Cataract Kidney disease Anemia ADL disability 1996 IADL disability 1996 ADL change—worse Depressive symptoms 1996 Depressive symptom change- worse Using stick/walker 1999 No glasses and unclear sight 1999 Wear glasses and clear sight 1999 Wear glasses and unclear sight 1999 No stick/walker and walking trouble 1999 Use stick/walker and walking well 1999 Use stick/walker and walking trouble 1999 -2 Log likelihood Chi-square

Odds ratio of full model Odds ratio of reduced model 1.007 -0.718 -1.666*** 1.641*** 0.998 -0.917 -1.036 -0.861 -0.892 -1.249 -1.308 -1.187 -0.798 -1.185 -1.151 -2.510* 2.986*** 1.645*** 1.761*** 1.749* 1.864*** 1.186 1.367* 1.149 -2.905*** 2.710*** 1.212* -1.060 -1.066 -1.092 -1.306*** -0.987 -1576.571 1594.482 165.907 (d.f.=26) 147.996 (d.f.=6)

Note1: N=1,809. The independent variables (reference group) included: Gender (male), diabetes (no), heart disease (no), stroke (no), bronchitis (no), arthritis (no), cataract (no), kidney disease (no), anemia (no), ADL disability (normal), IADL disability (normal), change of ADL (no change/improved), change of IADL (no change/improved), exercise (no), depressive symptoms (no), change of depressive symptoms (no change/improved), whether wearing glasses and seeing clear (no glass and seeing clear), whether using stick or walker (no use). Note 2: Full model was applied the entered logistic regression for analysis; reduce model was applied stepwise logistic regression, which only retained variables are reported in the table.

have different dominants of ADLs

risks may be easier than to diagnose

dysfunction, but still showed no change

depression or high risk of falls. Another

according to our definition.

study showed that older people will suffer

Having depressive symptom was

from more serious depression after fall

found to have higher risk of falls, that was

injuries [37], indicating that depressive

similar to past findings [15,18,30,40].

symptoms may be both the cause and

Biderman et al. [15] suggested that there

result of falls. The elderly who are in risk

were some common risk factors shared by

of fall and depression need to be

falls and depression, and to detect these

monitored more often and more closely by 147

台灣老年醫學 暨老年學雜誌

Risk factors of falls

medical professionals and their family.

第 3 卷第 2 期

diseases were related to falls in bivariate

It was interesting to find that elderly

analysis, such as hypertension, diabetes,

who did not wear glasses but could not see

heart disease, stroke, arthritis, kidney

clearly and those who used sticks/walkers

disease, and anemia. Multivariate logistic

and walking well had higher risk of falls.

regression analysis also showed no

Using sticks and walkers should help the

significant relationship between the elderly

disabled or frail elderly walk and live

who had exercise habit and lower risk of

independently, but those who can walk


well using assistive devices were more

This study found that 45.1% of falls

likely to fall, indicating the danger faced

had causes related to environmental

by assistive devices users. The possible

factors. A previous study reported the

reasons included inappropriate assistive

perceived causes of falls for the

devices used, unfamiliar uses, paying less

community elderly were mainly balance

attention to environmental risks, or

and gait, which was 61.9%, followed by

environmental risks that cannot be

accident or environments, 15.8% [26]. The

avoided. Those elderly who did not wear

environmental factors leading to fall

glasses and could not see clearly had

included uneven surface, objects on

higher risk of falls. That highlight poor

surface/rug, icy surface, wet surface or

eyesight without assistive devices

slippery footwear, storage problem, lack of

constitutes a major risk of falls. Northridge

safety features in bathroom, and having

et al. [38] found that not only the frail

problem in transfer [26,38].

elderly had higher risk of falls; the

Another study conducted in Taiwan

vigorous elderly who lived in homes with

found that falls happened outdoors were

more hazards also more likely to fall.

not significantly more severe than those

Geriatricians should be more concerned

that occurred indoors [12]. For the

with the vision problem in elderly. In

community-dwelling elderly, their home is

addition, the illumination in the house may

the main setting of their daily lives, and

not be enough and has to be modified.

these environmental risks not only result in

Previous studies reported that some

inconvenience for the disabled elderly but

chronic diseases were related to incidence

may also constitute the risk of falls and

of falls [8,22,28,39,40]. However, in this

further injuries. We suggest that home

study, having chronic diseases were not

safety and modification should be assessed

found to be significantly related to falls in

and supplemented in social welfare or

multivariate analysis, although some of the

long-term care policies. 148

Vol.3 No.2

Hui-Chuan Hsu et al

In this study, 73.8% of the fallers

Taiwan Geriatrics & Gerontology

not included in the data, such as use and

reported their fears of fall, and 65% of

efficacy of medication, and sleep

them expressed that they would reduce

disturbance. Third, the fall experience and

their outdoor activities because of their

related variables were only measured in

fear. This proportion was much higher than

1999 but not available at baseline or

previous findings, like hospital patients

between 1996 and 1999. We thus cannot

aged 77 years old or above (fear rate of

analyze if previous experience of repeated

21.7%) [16] or community elderly women

falls would be related to future falls.

aged 72 or above (fear rate of 27.0%)[23].

Moreover, the impact and severity of falls

The severity and impact of falls were not

were not measured in the analysis. Fourth,

investigated, so we have no evidence of

the numbers of falls were according to the

whether the community elderly had more

participants’ reported experience in the

serious experience of falls. Fear of falls

past year. Therefore, the mild falls might

may not constitute a direct risk of future

not be reported, and there may be recall

falls, but it was found to be linked with

bias in the numbers of falls. Therefore, the

fall-related self-efficacy, which was one of

prevalence of falls might be

the risks of falls [16]. The fear of falls may


influence the daily activity, social

Personal physical and psychological

function, and quality of life for the elderly.

risks as well as environmental risks of falls

The prevention of falls through controlling

may cause falls and injuries for the elderly.

the internal risk factors and removing

The physicians of geriatrics or family

external environmental causes should be

medicine should pay more attention to

the basic strategy of reducing fear of falls.

possible risks of their elder clients, and the

Building self-efficacy for safe exercise and

home safety for the elderly should be

managing falls themselves are also

noticed. In addition, the use of assistive

suggested [31].

devices including glasses, sticks or

There are some limitations in this

walkers are common in the elderly, but the

study. First of all, the longitudinal data

appropriateness of those devices should be

were obtained years ago in 1996 and 1999.

examined carefully.

Since longitudinal data and national representative samples were not easy to


get, this old dataset still provided some This study is based on data from the

important information about the risk of

“Survey of Health and Living Status of the

falls. Second, some of the risk factors were 149

台灣老年醫學 暨老年學雜誌

Risk factors of falls

第 3 卷第 2 期

Elderly in Taiwan” 1996 and 1999,

nonsyncopal falls.J Am Med Assoc

provided by the Population and Health

1989; 261: 2663-8.

Research Center, Bureau of Health

7. Schwartz AV, Villa ML, Prill M,

Promotion. The interpretation and

Kelsey JA, et al: Falls in older

conclusions contained herein do not

Mexican-American women. J Am

represent those of Bureau of Health

Geriatrics Soc 1999; 47:1379.


8. Schwartz AV, Hillier TA, Sellmeyer DE, Resnick HE, Gregg E, Ensrud KE, et al: Older women with diabetes have


a higher risk of falls: a prospective 1. Department of Health. Health

study. Diabetes Care 2002; 25:

statistics, 2005.



9. Stalenhoef PA, Diederiks JP,

tm. [Feb 24, 2007 accessed]

Knottnerus JA, de Witte LP, Crebolder

2. Chan KM, Pang WS, Ee CH, Ding YY,

HF: The construction of a patient

Choo P:Epidemiology of falls among

record-based risk model for recurrent

the elderly community dwellers.

falls among elderly people living in the

Singapore Med J 1997; 38: 427-31.

community. Fam Pract 2000; 17:

3. Dargent-Molina P, Favier F, Grandjean


H, et al: Fall-related factors and risk of

10. Wu CS, Chen CP, Lee FC, Lee SD:

hip fracture: the EPIDOS prospective

Prevention intervention to elderly fall

study. Lancet 1996; 20: 145-9.

in the community. Project research

4. Lin MR, Tsai SL, Chen SY, Tseng SJ:

report granted by Department of

Risk Factors for Elderly Falls in a

Health, 1993 (No:

Rural Community of Central Taiwan.

DOH82-HP-27-5S05). [In Chinese]

Taiwan J Public Health 2002;

11. Yasumura S, Haga H, Nagai H, Suzuki

21:73-82. [In Chinese]

T, Amano H, Shibata H: Rate of falls

5. Liu WL: The identification of fall risks

and the correlates among elderly

among community elderly. Project

people living in an urban community in

research report granted by Department

Japan. Age Ageing 1994; 23:323-7.

of Health, 1991 (No: DOH86-TD-078).

12. Leung WS, Chi HT, Hu MH, Liu MT:

[In Chinese]

Fall mechanism and injury severity in

6. Nevitt MC, Cummings SR, Kidd S,

community dwelling older people.

Black D:Risk factors for recurrent

Formosa J Phy Ther 2005; 30: 105-15. 150

Vol.3 No.2

Hui-Chuan Hsu et al

[In Chinese]

Taiwan Geriatrics & Gerontology

18. Jørgensen L, Engstad T, Jacobsen BK:

13. Tinetti ME, Liu WL, Claus EB:

Higher incidence of falls in long-term

Predictors and prognosis of inability to

stroke survivors than in population

get up after falls among elderly

controls: depressive symptoms predict

persons. J Am Med Assoc 1993; 269:

falls after stroke. Stroke 2002; 33:



14. Carter ND, Khan K, McKay HA, Petit

19. Kenneth E, Covinsky E, Kahana B, et

MA, Waterman C, Heinonen A, et al.:

al: History and mobility exam index to

Community-based exercise program

identify community-dwelling elderly

reduces risk factors for falls in 65- to

persons at risk of falling. J Gerontol A

75-year-old women with osteoporosis:

Biol Sci Med Sci 2001; 56A: M253-9.

randomized controlled trial. CM A J

20. Koshki K, Luukinen H, Laippala P,

2002; 167: 997-1004.

Kivela SL: Physiological factors and

15. Biderman A, Cwikel J, Fried AV,

medications as predictors of injuries

Galinsky D: Depression and falls

falls by elderly people: a prospective

among community dwelling elderly

population-based study. Age Ageing

people: a search for common risk

1996; 26: 29-38.

factors. J Epidemiol Community Health

21. Lamb Se, Ferrucci L, Volapto S, Fried

2002; 56: 631-6.

LP, Guralnik JM: Risk factors for

16. Cumming RG, Salkeld G, Thomas M,

falling in home-dwelling older women

Szonyi G: Prospective study of the

with stroke: The Women’s Health and

impact of fear of falling on activities of

Aging Study. Stroke 2003; 34:

daily living, SF-36 scores, and nursing


home admission. Prospective study of

22. Lawlor DA: Association between falls

the impact of fear of falling on

in elderly women and chronic women

activities of daily living, SF-36 scores,

and chronic diseases and drug use:

and nursing home admission. J

cross sectional study. Br Med J 2003;

Gerontol A Biol Sci Med Sc. 2000

327: 805-7.


23. Murphy SL, Dubin JA, Gill TM: The

17. Hill EL, Cumming RG, Lewis R,

development of fear of falling among

Carrington S, Le Couteur DG: Sleep

community-living older women:

disturbances and falls in older people.

predisposing factors and subsequent

J Gerontol A Biol Sci Med Sci 2007;

fall events. J Gerontol A Biol Sci Med

62A: 62-6.

Sci 2003; 58A: 943-7. 151

台灣老年醫學 暨老年學雜誌

Risk factors of falls

24. Richardson DR, Hicks MJ, Walker RB:

第 3 卷第 2 期

30. Whooley MA, Kip KE, Cauley JA,

Falls in rural elders: an empirical study

Ensrud KE, Nevitt MC, Browner WS:

of risk factors. J Am Board Fam Pract

Depression, falls, and risk of fracture

2002; 15: 178-82.

in older women. Arch Int Med 1999;

25. Stalenhoef PA, Diederiks JP,

159: 484-90.

Knottnerus JA, Kester AD, Crebolder

31. Andresen EM, Wolinsky FD, Miller JP,

HF: A risk model for the prediction of

Wilson MMG, Malmstrom Tk, Miller

recurrent falls in community-dwelling

DK: Cross-sectional and longitudinal

elderly: a prospective cohort study. J

risk factors for falls of falling, and falls

Clin Epidemiol 2002; 55:1088-94.

efficacy in a cohort of middle-aged

26. Talbot LA, Musiol RJ, Witham EK,

African Americans. Gerontologist

Metter EJ: Falls in young, middle-age

2006; 46: 249-57.

and older community dwelling adults:

32. Tinetti ME, Inouye, K, Gill TM,

perceived cause, environmental factors

Doucette JT: Shared risk factors for

and injury. BMC Public Health 2005;

falls, incontinence, and functional

5: 1186-1147.

dependence: unifying the approach to

27. Vellas B,Wayne SJ, Romero LJ,

geriatric syndrome. J Am Med Assoc

Baumgartner RN, Garry PJ: Fear of

1995; 273; 1348-53.

falling and restriction of mobility in

33. Wyller TB. Risk factors for serious fall

eldly fallers. Age Ageing 1997;

related injury in elderly women living


at home. Inj Pre 2004; 10: 308-13.

28. Volpato S, Leveille SG, Blaum C, Fried

34. Faulkner KA, Cauley JA, Zmuda JM,

LP, Guralnik JM: Risk factors for falls

Griffin JM, Nevitt MC. Is social

in older disabled women with diabetes:

integration association with the risk of

The Women’s Health and Aging Study.

falling in older community-dwelling

J Gerontol A Biol Sci Med Sci 2005;

women? J Gerontol A Biol Sci Med Sci


2003; 58A: 954-9.

29. Wallace, C, Reiber G, LeMaster J,

35. Radloff LS. The CES-D scale: a

Smith DG, Suillivan K, Hayers S, Vath

self-report depression scale for

C: Incidence of falls, risk factors for

research in the general population. App

falls, and fall-related fractures in

Psycho Measurement 1977; 1:

individual with diabetes and a prior


foot ulcer. Diabetes Care 2002; 25:

36. Kohout FJ., Berkman L, Evans DA,


Huntley JC: 1993 “Two shorter forms 152

Vol.3 No.2

Hui-Chuan Hsu et al

of the CES-D Depression Symptom

Taiwan Geriatrics & Gerontology

1995; 85: 509-5.

Index”. J Aging Health 1993; 5:

39. Hanlon JT, Landerman LR, Fillenbaum


GG, Studenski S. Falls in African

37. Scaf-Klomp W, Sanderman R, Ormel J,

American and white community-

Kempen GIJ: Depression in older

dwelling elderly residents. J Gerontol

people after fall-related injuries: a

A Biol Sci Med Sci 2002; 57:M473-8.

prospective study. Age Ageing 2003;

40. Huang SC, Chen TJ, Chou P.

32: 88-94.

Cumulative incident rate and

38. Northridge ME, Nevitt MC, Kelsey JL,

associated factors of falls among the

Link B. Home hazard and falls in the

elderly in Shih-Pai, Taiwan. Taiwan J

elderly: the role of health and

Public Health 2005; 24: 136-45.

functional status. Am J Public Health


[Original Article]


徐慧娟 1

詹莉君 2

目的:本研究目的在應用全國代表性長期追蹤樣本,分析老人跌倒盛 行率及其相關危險因子。 研究方法:資料來自國民健康局「老人生活與保健調查」1996年與1999 年長期追蹤資料,1996年為2,669人,1999人為2,310人。相關危險因子變 項包括人口變項、慢性病、身體功能障礙及其變化、憂鬱症狀、運動習 慣、使用柺杖或助行器以及是否行走正常、戴眼鏡以及是否看得清楚等 。以邏輯式迴歸進行分析。 結果:在1999年有19.5%的老人有跌倒經驗,其中7.2%為重複跌倒。 跌倒的相關危險因子包括女性、身體功能失能、身體功能惡化、有憂鬱 症狀、使柺杖或助行器但走路方便、未戴眼鏡且看不清楚等。有73.8%的 跌倒老人都害怕再次跌倒,有45.1%的跌倒老人認為是環境因素造成的。 結論:個人身體與心理因素和環境因素均可能造成老人跌倒與傷害。 老年人之家庭醫師或老年科醫師應對其可能的危險情形具有敏感性,老 年人輔具使用適當性與居家環境安全亦應注意。 ( 台灣老年醫學暨老年學雜誌 2008;3(2):141-154) 關鍵詞:跌倒、老人、長期追蹤、危險因子


亞洲大學健康暨醫務管理學系副教授、2 署立醫院中醫聯盟 通訊作者:徐慧娟 通訊處:413 台中縣霧峰鄉柳豐路 500 號 Email: [email protected]