Journal of Epidemiology and Preventive Medicine

A Prevalence Study of the Activities of Daily Living (ADL) Dependency among the Elderly in Four Districts in Selangor, Malaysia

Download PDF

Published Date: October 18, 2015

A Prevalence Study of the Activities of Daily Living (ADL) Dependency among the Elderly in Four Districts in Selangor, Malaysia

Sharifa EBW. Puteh1*, Intan MA. Bakar1, Boekhtiar Borhanuddin1, Khalib Latiff1, Rahmah M. Amin2 and Rosnah Sutan1

1Department of Community Health, Faculty of Medicine, the National University of Malaysia, Malaysia

2Faculty of Medicine and Health Sciences, Universiti Sultan Zainal Abidin, Kuala Terengganu, Terengganu, Malaysia

*Corresponding author: Sharifa EBW. Puteh, Department of Community Health, Faculty of Medicine, the National University of Malaysia, Malaysia, Tel: +603-9145-5891; Fax:  +603-9145-6600; E-mail: sh_ezat@ppukm.ukm.edu.my  

Citation: Puteh SEBW, Bakar IMA, Borhanuddin B, Latiff K, Amin RM, et al. (2015) A Prevalence Study of the Activities of Daily Living (ADL) Dependency among the Elderly in Four Districts in Selangor, Malaysia. J Epid Prev Med 1(2): 110. Doi: http://dx.doi.org/10.19104/jepm.2015.110

 

Abstract

 

Introduction: Similar to many other developing countries worldwide, Malaysia is currently experiencing a demographic transition towards an ageing population. This situation leads to an increasing number of disabled individuals in the population. The present study investigated the epidemiology of activities of daily living (ADL) dependency among the elderly in one of the highest urbanized state in Malaysia i.e. the state of Selangor.

Objective: To determine the prevalence of ADL dependency (which also reflects disability) among the elderly population in Selangor and its associated factors.

Methodology: This is a clinic-based, cross sectional-study done in four urban and rural districts in Selangor. One hundred and seventy five (175) subjects aged 60 years and above were selected. They were administered the Modified Barthel Index (MBI) to assess self-reported ADL dependency. The Elderly Cognitive Assessment Questionnaire (ECAQ) and the Geriatric Depression Scale (GDS) were used as screening tools.

Result: A total of 175 elderly subjects enrolled. The majority of the respondents were Malays and Muslims (60%). Prevalence of mild dependency was 14.9%; moderate dependency was 9.1%; severe and total dependency was at 1.1 % each. Based on the multivariable analysis: those who have more than one chronic disease have significantly lower odds to be dependent (Adjusted Odds Ratio (AOR): 0.09, 95% CI 0.03-0.34) while those with high social support, have significantly higher odds to be dependent (AOR: 7.65, 95% CI 1.60-36.57).

Conclusion: Elderly having more than one chronic disease seems to protect against being dependent. Whereas, good social support from friends is associated with higher risk towards dependency.

Keywords: Barthel Index; Elderly; Dependency

Top ↑

Abbreviations

ADL: Activities of Daily Living; ECAQ: Elderly Cognitive Assessment Questionnaire; GDS: Geriatric Depression Scale.

 

Introduction

 

Ageing is a natural course of dynamic biological changes that is also subject to social construction in the determination of its reality and meaning.  Hence, the definition of old age (based on changes in chronological age and functionality) varies across societies and cultures. For example, in many developing countries, the elderly is defined as those who are 60 years old and above. This definition was based on the decision made in the World Assembly on Aging held in Vienna in 1982 [1,2]. This differs from the definition used by most developed countries that choose the cutoff-point of 65 years and above.

Starting from the twentieth century, the proportion of older population throughout the world is continuously increasing [3]. This demographic change becomes a challenge worldwide, in both developed and developing countries. In the year 2013, the world ageing population was 11.7 per cent (841 million) and expected to rise to 21.1 per cent (2 billion) in 2050. This can be explained by the projection that between the year 2010 and 2050, life expectancy at age 60 will increase from 20 years to 22 years worldwide on average from 23 years to 26 years in more developed regions and from 19 years to 21 years in less developed regions. Throughout this period, the population growth of the elderly will be dominated by the less developed regions, whereby their number in these regions will triple from 554 million in 2013 to 1.6 billion by 2050.

Malaysia is experiencing a demographic transition as many other countries worldwide. This transition includes the increase of longevity, lowering of mortality, declining fertility and a healthier living environment. Currently, the percentage of the elderly in Malaysia stands at eight per cent (2,308,720 individuals), which is still lower than the 10 per cent needed for the definition of an ageing population [4]. However, there are already signs towards the inevitable situation as the number of the elderly is a rising trend. The older population increased from 539,118 (5.2% of population) in 1970 to 745,152 was (5.7% of the population) in 1980 [5]. The situation is also contributed by the increasing trend of life expectancy at birth and reduction of child birth rate among the total population of 28.9 million Malaysian in 2011 [4].

The phenomenon of population ageing in Malaysia predictably leads to an increasing number of disabled populations. This is supported by several studies that confirmed the association between disability and ageing [6,7]. The increment has led to a rise of the national health burden, whereby there was an increasing trend of total expenditure of the gross domestic product from 3.1% in 2000 to 4.4% in 2010 [4].

Disability is a complex phenomenon. It reflects the interactions between a person’s body, features of the society and environmental factors that give an impact on how he or she lives. In 2001, 191 member states of the World Health Organization agreed to adopt the International Classification of Functioning, Disability and Health (ICF) as the basis for the scientific standardization of data on health and disability world-wide [8]. According to the Malaysian law, an individual with disability is defined as a person who has “long term physical, mental, intellectual or sensory impairments which in interaction with various barriers that may hinder their full and effective participation in society” [9]. These disabilities  can be assessed through the activities of daily living (ADL), which  are routine tasks performed by each individual on a daily basis that are essential to independent living without any assistance, and involves self-care [10]. A person’s ability in performing ADL is important in determining their long term care and coverage of their needs [2,11,12]. An individual with disability will have dependency that requires more help to do house chore, as well as higher needs during nursing care and hospitalization. Hence, functional dependency will determine the health care needs among the elderly. In order to assess this dependency, various assessment tools for ADL have been developed. One of these tools will be further described in the methods section.

A few studies suggested that some socio-demographic factors, living lifestyle and medical illnesses have influence on the ADL among elderly respondents [13-15]. The majority of studies highlighted that medical illness status of respondents is a major influence of ADL among the elderly. Most of these studies suggested that disability or dependency among older people is higher among women, increases with age and presence of chronic diseases [6,16-18]. The prevalence of elderly disability also varies based on locality, such as in the general population, nursing home population, rural/ suburban/ urban, developing or developed countries.

The objectives of this study were to determine the prevalence of ADL dependency using Modified Barthel Index (MBI) among the elderly in both rural and urban population in the state of Selangor (highly urbanized state in Malaysia), as well as to determine its associated factors.

Top ↑

Methods

 

Study design and sampling

This was a cross-sectional study using self-administered consented questionnaires, implemented in the primary care clinic setting. Ethical approval was obtained from UKM Research and Ethics committee as well as the National Medical Research Register (NMRR) Ministry of Health Malaysia. The NMRR registration number for this study is NMRR-12-320-10948.

Four districts in the state of Selangor were purposively chosen for the study. The districts of Gombak and Kuala Selangor represent suburban and rural community, whereas the districts of Sepang and Petaling represent urban community. The number of health clinics in each district is as follows: eight in Gombak, five in Kuala Selangor, three in Sepang and five in Petaling. From each districts, two health clinics were purposively selected based on the highest total number of registered elderly patients in 2010. The estimated number of potential respondents in all these clinics per year was approximately to be around 500.

The sampling of the respondents was based on purposive sampling.  The inclusion criteria were as follows: (i) elderly patients (age 60 years and above) who were staying in the selected districts; (ii) attended the outpatient services of the selected health clinics in 2010; and (iii) have normal scores for cognitive function and depression via screening using the Elderly Cognitive Assessment Questionnaire (ECAQ) and the Geriatric Depression Scale (GDS), respectively. The calculation of the minimum sample size (N) based on logistic regression model was done using the rule-of-thumb formula suggested by Peduzzi et al.— N = 10k/p, where k is the number of independent variables and p is the prevalence of the condition of interest (based on the categorical dependent variable) in the analyses [19]. Hence, the calculated minimum sample size needed for simple binary logistic regression using this formula was 50 subjects, where k = 1 and p = 0.2 (based on the assumptions that about 20 per cent of the subjects would have some form of disabilities). For the multiple binary logistic regressions with two predictors (i.e. k = 2), by using the same formula and assumption (p = 0.2), the minimum number of subjects needed for the final parsimonious model was 100 subjects.

Data collection and study tool

Data collection was done using a self-administered questionnaire. It consist of 12 sections including socio-demographics profile, living arrangement, work status, income, social network, health status and the Modified Barthel Index (MBI) used to measure Functional Status among the participants. The Elderly Cognitive Assessment Questionnaire (ECAQ) and the Geriatric Depression Scale (GDS) sections served as concurrent screening tools for the inclusion criteria. The stringent criteria allow the elderly selected are of good cognitive function (to exclude the probable dementia participants) and devoid of depression that may hinders their ADL response reliability of recall ability. Hence, this leads to a low number of selected respondents’ fulfilling the above criteria’s.

The MBI is used in our study to measure the level of disability based on the activities of daily living (ADL) that reflects the dependency in daily activities among the elderly. MBI basically measures functional status that corresponds to societal standards and priorities [20]. It is a measure that focuses on multiple indicators of “the good things in life”, achievements, priorities and ADL. The major domains include employment and education; marriage and spousal relations; sexuality; other major social relations such as friendships; leisure activities; spirituality and religion; healthcare; equipment and accessibility; and personal caregivers. MBI covers ten domains of functioning (activities): bowel control, bladder control, as well as help with grooming, toilet use, feeding, transfers, walking, dressing, climbing stairs, and bathing. The internal consistency of this inventory yields a Cronbach’s alpha value of 0.90, which is excellent [21]. The format is a 10-item scale where each activity is given a score for five levels of dependency ranging from 0 (unable to perform task) to a maximum of 15 (fully independent) [20].  In the MBI scoring, each activity is given points ranging from 0 (unable to perform task) to a maximum of 5, 10, or 15 points (fully independent). A total score is obtained by summing the points for each item. Total scores may range from 0 to 100, with higher scores indicating greater independence and fully independent as a Barthel Index score at 100. Based on the studies stated prior to this, functional disability can be defined as a Barthel Index score of 95 (which corresponds to 1 decrease in an item on the Barthel Index) or lower. Besides that, the severity of disability can be categorized into four levels: slight dependency (a score of 95), moderate dependency (a score of 65-90), severe dependency (a score of 25–60) and total dependency (a score of 0–20). The latter corresponds approximately to a bedridden state, with decreased functionality in at least eight activities.  In this study, the analysis will categories in two group of independent and dependent only due to low number of participants. Self-assessment of current status based on the past 48 hours is preferred due to unreliability of recall ability among the elderly.

The Elderly Cognitive Assessment Questionnaire (ECAQ) is used to determine the respondents’ cognitive levels. It is derived from items in the Mini-Mental State Examination and Geriatric Mental State Schedule. ECAQ has been shown to be a valid routine screening tool for assessment of cognitive impairment among older people living in developing countries for relatively low education level population. It is a reliable and valid scale that shows a sensitivity of 85.3%, specificity 91.5%, positive predictive value of 82.8% and overall error rate of 10.5% [22].  It consists of ten items that assess long term memory, orientation and recall. Score ranges from zero to ten, whereby a score of seven or more indicates normal memory, score of five or six indicates borderline and score of four and below indicates probable dementia [11,23,24]. The Geriatric Depression Scale (GDS) (short form version) that was used in our study consists of 15 questions to screen for depression among the elderly [25]. The internal consistency of this inventory is excellent with the Cronbach’s alpha value of 0.94. The criterion validity of this inventory is also good, whereby the main effect for the classification variable based on normal, mildly depressed or severely depressed was highly significant. The score range from zero to 15, with the score of six or more indicates depressive symptoms [11,26,27].

Main variables

The primary dependent variable for the study inferential analysis was the ability for ADL among elderly, which will be categorized into having no dependency (MBI score of 100) and having dependency (any other MBI scores) in daily life. The nine main independent variables of interest were the age, ethnicity, religion, education level, living status (i.e. married/having a partner or not), living arrangement (i.e. staying alone or not) , presence of chronic disease(s), social support from family and social support from friends. The secondary dependent variable was the dependency level based on ADL (based on the four types of dependency classification of MBI that was discussed prior to this). Besides that, other variables of interest  the specific medical conditions that included hypertension, diabetes mellitus, heart disease, arthritis, hypercholesterolemia, hearing problem, vision problem, urinary problem and other diseases (which included cardiovascular accident, depression, anxiety, sexual problem, asthma and gout).

Data analysis

The data were analyzed using SPSS version 21 software. The significance level for all the inferential statistics in this study was set p < 0.05 (two-tailed). Descriptive data with categorical variables were analyzed using percentage and their accompanying confidence interval. Descriptive data with continuous variables were analyzed using mean (plus their standard deviation (S.D.)) or median (plus their interquartile range (IQR) value), depending on the normality of the data. The normality of the data was assessed using the Kolmogorov-Smirnov test.

Bivariate statistical analyses were done as part of the procedure of the multivariable analysis, in terms of the crude associations. For the bivariate statistical analyses, Chi-Square test and simple binary logistic regression analyses were done separately for each association between the primary dependent variable and the nine independent variables that were potential predictors. In reference to the Chi-Square test, if there were any violation of assumptions, Yates’ correction or Fisher’s exact test would be done where necessary.

For the multivariable statistical analysis of this study, multiple binary logistic regression analysis based on the step-wise method was used for the determination of the main effects of each nine potential predictors on the primary dependent variables [28]. The appropriate assumption tests were done for the multivariable analysis (e.g. multivariable normality (proven through unstandardized residual normality analysis); linearity (proven through the no significant interaction of the continuous predictor and its natural log transformation term); and multicolinearity (proven through the examination of Variance Inflation Factor (VIF) value that should not be more than 10). Any variables that violated the assumptions would not be entered into the multivariable model. For the stepwise method (i.e. forward likelihood ratio method), the decisions about the order in which predictors are entered into the model are based on a purely mathematical criterion. Two-way interaction was done only between predictors that were significant in a separate final model after the preliminary model was identified. Evaluation of the validity of the model was done via assessment of the goodness-of-fit through the Hosmer-Lemeshow test. The best model step was chosen based on this test to represent the best possible multivariable model.

Top ↑

Results

 

Of the total 175 eligible subjects who answered the questionnaires, only 120 of them completed it properly. It was not known why these 55 subjects did not answer the social support section. The number of eligible respondents is considered low as not many respondents managed to pass through the screening tools and preceded with the MBI. Nevertheless, the overall sample was representative of the older population in Selangor. Table 1 shows characteristics of the 175 study subjects (including the subjects who did not complete the questionnaire).

Table 1: Characteristics of study population by gender (N = 175)

 

The respondents ranged between the ages of 60 to 85 years. The overall mean age was at 67.93 ± 5.4 years, whereby men (n = 84) have slightly younger mean age at 67.56 ± 5.0 years than women (n = 91), with mean age of 68.27 ± 5.8 year. 88.6 % of the subjects (n = 155) could be classified as young old (60 to 74 years old) and the other 11.4 % (n = 20) could be classified as older old (75 years old and older). 60 % of subjects were Malays and Muslim, whereas 84 % had some form of education. Women (34.1%) were more likely to have living partner compared to men (10.7%) and 7.7 % of the women were living alone compared to men (2.4%). 59 % of the participants had at least one self-reported medical condition, with the proportions among men and women being 48.8% and 69.2%, respectively. The descriptive results of the prevalence rates for ADL dependency using the MBI score based on the characteristics of subjects is shown in Table 2. Overall, 24.2 % of them reported needing help in at least one of the ten ADL. The prevalence of ADL dependency based on the MBI score for mild dependency was 14.9%; for moderate dependency was 9.1%; and 1.1% each for severe and total dependency. A large percentage at 73.7% of the respondents, were independent of any help for their daily activity. This reflects the good and able condition of most attendees to the clinics. The severe and total ADL dependents are scarcely seen attending clinics as their disability may forbid them from doing so. Hence, only small amounts of severely dependent elderly and of ADL capacity are approached. Table 2 summarizes the severity of disability status based on MBI according to the ten potential predictor variables of ADL dependency.

 

Table 2: Prevalence of disability status among the elderly in the study based on level of activities of daily living dependency according to Modified Barthel Index (N = 175)

 

Support from family (high and low family support) showed both to be similar at almost 50% each. It shows that most half of the elderly perceived family support as low. However, high friends support was seen at a high rate of 71.5% than low friends support at 28.5%. The bivariate analyses (Table 3 and Table 4) showed that there was significant difference of ADL dependency status in terms of ethnicity, religion, social support from friends and chronic disease status. Based on the simple binary logistic regression, it seemed that being Chinese (Odds Ratio (OR): 3.70, 95% CI 1.52-8.99), being non-Muslim (OR: 2.64, 95% CI 1.18-5.90) and having high social support from friends (OR: 10.46, 95% CI 2.35-46.54) were risk factors for ADL dependency among the subjects. On the other hand, elderly having one or more chronic diseases seemed to be a protective factor against ADL dependency at OR: 0.10 (95% CI 0.01-0.81) and OR: 0.09 (95% CI 0.02-0.31), respectively.

Table 3: Chi-Square analysis of activities of daily living (ADL) dependency with its factors (N=120) Note: The variable of living arrangement cannot be analyzed as there is violation of assumption—singularity (i.e. zero value in one of the subgroup) a Chi-Squared test with Yate’s correction          bFisher’s exact test           Significance level: * p < 0.05

 

Table 4: Simple binary logistic regression analysis of variables associated with activities of daily living (ADL) dependency according to Modified Barthel Index (N=120) Note: Each logistic regression was done by using ADL independence as the reference group for the outcome. The variable of living arrangement cannot be analyzed as there is violation of assumption—singularity (i.e. zero value in one of the subgroup) Significance level: * p < 0.05

 

Table 5: Multiple binary logistic regression of influencing factors of activities of daily living (ADL) dependency (N=120) Note: Each logistic regression was done by using ADL independence as the reference group for the outcome. Only eight independent variables were included in the stepwise method (age group, gender, ethnic, religion, education level, and presence of chronic disease, social support from family and social support from friends). The variable of living arrangement cannot be analyzed as there is violation of assumption—singularity (i.e. zero value in one of the subgroup). The final model has good model fitness-based on the non-significant Hosmer-Lemeshow’s test and a high Nagelkerke’s R-squared value of 37.6%.  Significance level: * p < 0.05

 

Table 5 presents the final model of multivariate analysis that tried to identify the independent predictors of ADL dependency. In general, there was no violation of assumptions (i.e. adequate ratio of cases to predictor variable [20:1]; presence of linearity of logics, absence of multicolinearity; absence of or outlier in the solution and presence of independence of error). Only a single variable (living arrangement) was removed from the initial model due to the presence of singularity (i.e. one of the cell has a value of zero). The model was valid in terms of goodness-of-fit, whereby the Hosmer-Lemeshow test was still not significant in step number two of the stepwise process. None of the variables in the final have a significant two-way interaction, and therefore no interaction term needs to be included in the final model. In the best-fit final multivariable model, the data showed that: (i) those who have more than one chronic disease have significantly lower odds to have ADL dependency (Adjusted Odds Ratio (Adj. OR): 0.09, 95% CI 0.03-0.34); and (ii) those with high friend social support have significantly higher odds to have ADL dependency (Adj. OR: 7.65, 95% CI 1.60-36.57). This is interesting as family supports are not seen to be significant predictor towards ADL dependency. Based on the Nagelkerke R-squared value of 0.376, we can interpret that 37.6% of the variance in the model can be explained by the two variables. This percentage can be considered as high.

Top ↑

Discussion

 

Comparison with previous ADL dependency studies in Malaysia

Previous studies on dependency among the elderly in Malaysia differ from ours in terms of the measurement of ADL dependency and the characteristics of samples. For example, a population-based study of 765 older people in the state of Melaka in 2008 used the 10-item Barthel Index [27]. The 2008 study found that these researchers found that 24.7 % of subjects aged 60 years and above were ADL dependent, if compared to 49.6 % of those aged 75 years and above. These results slightly differ from our study that found the prevalence of ADL dependency to be 30.0% in the former and 33.3% in the latter. This means that the subjects aged 75 year plus in our study of urban-rural population mixture have less ADL dependency compared to previous study’s rural population. This difference may suggest that disabilities among the elderly of rural areas are worse than those from the urban areas in Malaysia. This reflects a persistent theme of health inequity in Malaysian healthcare setting.

Another study among elderly Malays in rural areas of Peninsular Malaysia in 2006 found that prevalence of dependency among older people was 22.8%, whereby men were more dependent than women [29]. This is similar to the descriptive result in our study, albeit not statistically significant. However, a contrary result can be seen in another study in Malaysia, based on the 2003 Mental Health and Quality of Life of Older Malaysians Survey to assess self-reported disabilities among the elderly based on instrumental activities of daily living (IADL) and activities of daily living (ADL). The strength of this study is the use of multi-staging sampling in all the 13 states of Malaysia based on urban-rural stratification. This study found that the disability in 13 or more IADL or ADL components was 22.8%, with women having higher dependency prevalence (31%) than men (14%) [14]. Our study found almost similar number of both gender facing ADL dependency, however males have higher number of mild dependency and females have more moderate dependency. This study may representative of the older population in Selangor as the distribution of the female higher than male as life expectancy of female higher than male with 77.0 and 72.3, respectively [30].

International comparisons of prevalence of ADL dependency

The use of different ADL measurements in studies from other countries causes some difficulty in comparing the results. Most of these international studies used the original Barthel Index, rather than MBI as used in our study. For example, investigators from Singapore found a much lower dependency prevalence based on the 10-item Barthel Index— 11% in people age 60 years and above; 14% in people age 65 and above; and 26% among those aged 75 and above [6].  These figures were much lower than the figures in our study, which may be explained by the status of developed nation for Singapore if compared to Malaysia.  Another study from the developed nation of Japan also portrayed some rather similar findings, comparable to Singapore. In this study, the prevalence of ADL dependency (based also on Barthel Index) among those aged 65 years and above was 20.1% [17]. This dependency was found to be higher with age, among women and those presenting with medical illness. For comparison with another developing middle-income nation, another study from Nigeria will be more relevant. Based on the findings in this study, more than one quarter (28.3%) of the elderly in rural area of Nigeria have at least one area of dependency based on the ten ADL in the Barthel Index [11]. This ADL dependency prevalence reaches up to 98.9 % among those aged 75 years and above. The ADL dependency was also higher among women than men, which is different from our result.

There are several possible reasons why the ADL dependency prevalence is higher in Malaysia and other developing countries compared to those reported from the developed countries. The existence of socio-demographic differences between developing countries and developed countries might partially explain this discrepancy. In our study, there were more old people with low educational levels (48% have no form of education and primary education level only). Besides that Malaysian elderly population were more economically active, whereby the 2010 Malaysian census reported that there are people in the 60-64 years age groups were working in the agricultural-related industry. Another possible reason is due to cultural differences. The traditional Malaysian culture emphasizes that old people should be taken care of by their own family members. Regarding private nursing homes— an institutional care centre is only accessible to those who can afford them. Admission to funded shelter homes is usually the last choice to older people who have no heir, no shelter on their own or those who are poor. Therefore, majority of the elderly in Malaysia continue to stay with their spouse or other family members. On the other hand, institutionalization in the developed countries is closely related to disability in terms of ADL [31] and lack of family support. For example, a study done among institutionalized older people in nursing homes in Japan found the ADL dependency prevalence at 38.5% [2]. Therefore, the higher prevalence of ADL dependency in our study may be due to the lower number of publicly institutionalized elderly in Malaysian population as these government funded institutions are strictly reserved for the elderly without homes, heirs or family.

Ethnic and religious differences

In terms of ethnic variation, we found that Chinese elderly had significant odds of ADL dependency, followed by Indians and Malays in the bivariate analysis. Our findings are different from previous international studies on ethnic variation and ADL dependency in previous study in Malaysia that found no significant difference between ethnicity [11, 14]. The observed differences among the ethnic groups in our study might be attributable to the different types of education level. We found that almost three quarter (71%) of the older Chinese had higher educational level (secondary and tertiary level) as compared to 49.5 % of the Malays and 40.0 % of the Indians. Several studies have found that religiosity and spirituality play important roles in helping people to maintain physical functioning as they grow older or regain functioning after an illness. Religious beliefs have the potential to influence the cognitive appraisal of negative life events in a way that makes them less distressing. For people with medical illness, these beliefs are particularly useful in protecting them against dependency [32]. Besides that, our study found that Muslims were less likely to have ADL dependency. This may be due to the fact that all the Malays in our study can be considered as Muslim in Malaysia. The question of whether they are practicing Muslims or nominal Muslims is an entirely different issue. We postulate that the protective effect of being a Muslim attributable to the higher observance of compulsory religious rites (praying and meditating) among them, which may be considered as a form of “informal therapy”.

ADL dependency, health status and social support from friends

 Two of the most surprising findings in our study are that the presence of chronic disease is a protective effect against dependency, whereas receiving friends’ social support is a risk factor for dependency. The population in our study that suffered more than one disease but interestingly lived without any ADL dependency is high at 90.4%. Our literature research did not prepare us for these counter-intuitive findings, there was no similar finding out there that we know of. We postulated that the presence of chronic disease may motivate subjects to have a higher health-efficacy, which motivates them to seek treatment at nearby health facilities, thus mitigating effects of disabilities. Having chronic diseases, may encourage elderly to gather help and treatment at health facilities. On why high friends’ social support is associated to ADL dependency, there is a probability that as individuals became more dependent, they attracted or gathered more social support from friends. For example, a friend might not be concerned if a person’s disability is not severe, as he might attain the ability to take care of himself. This is totally different in the situation where a person might be bed-ridden. In this scenario, friends around him would be more willing to provide some forms of assistance, as part of our culture social altruism.

Limitations and Recommendations for Improvement of Study

This study has several limitations that may be rectified in future studies. First and foremost is the fact that the sample size needed for the testing of multivariable model is greatly limited. The smaller-than-expected sample size our study might have also underestimated the prevalence of ADL dependency. The analyses might have yield more significant findings, if the study was to include more areas in the sampling frame. Another issue is that the generalization of finding is only relevant to community-dwelling older people as in many previous studies, which systematically excluded the elderly in institutionalized care. This study also did not look at elderly institutions. Hence, the external validity of this study was limited by the purposive sampling employed in the study that caused selection bias to happen. However, due to the limited resources (time and man power) for the operation of the study, this bias was unavoidable in practice. To improve on this limitation, future studies should be done through a form of “tracing” of potential subjects based on a dedicated registry for the elderly based on the latest census. Besides that, the inherent weakness of any cross-sectional designs causes the inappropriateness of drawing a causal inference between health-related variables and ADL dependency. To improve this, we recommend that future studies should utilize the cohort design, whereby the registry suggested prior to this prospectively and continuously update for changes of physical illness and disabilities in the specified elderly cohort including those in elderly institutions. Future studies must also utilize the qualitative method to explore the contradictory findings in this study as the real situation may be too complex for current quantitative measures.

Top ↑

Conclusion

 

We found that ADL dependency is a common situation among older people in Malaysia. Our rates appears to be much higher than in developed countries but comparable to other middle-income developing countries. Based on our study, there are counter-intuitive findings that suggest that the presence of chronic diseases might be a protective factor against ADL dependency. This statement is to be taken with caution. As the population in our study that suffered more than one disease but lives without any ADL dependency is high at 90.4%. Whereas high social support from friends might be a risk factor for ADL dependency. Further studies are warranted to re-examine these findings in order to understand whether these results are just statistical flux or portrayed the real situation in the field.

Top ↑

Acknowledgements

 

The authors would like to thank the Dean and Hospital Director UKM Medical Centre and the Director of Health Malaysia for permission to publish this paper.

Top ↑

References 

 

  1. United Nations. Vienna International Plan of Action on Aging.  New York. United Nations publication,1983.
  2. Nakazawa A, Nakamura K, Kitamura K, Yoshizawa Y. Association between activities of daily living and mortality among institutionalized elderly adults in Japan. J Epidemiol. 2012;22(6):501-7.
  3. United Nations. World Population Ageing 2013. New York. United Nations publication, 2013.
  4. World Health Organization. World Health Statistics. World Health Organization, Geneva, Switzerland, 2013.
  5. Ahmad WIW, Ismail Z. Population ageing and religious participation among rural elderly in Terengganu, Malaysia. Journal of US-China Public Administration 8:968-977.
  6. Ng TP, Niti M, Chiam PC, Kua EH. Prevalence and correlates of functional disability in multiethnic elderly Singaporeans. J Am Geriatr Soc. 2006;54(1):21-9.
  7. Reyes-Ortiz CA, Ostir GV, Pelaez M, Ottenbacher KJ. Cross-national comparison of disability in Latin American and Caribbean persons aged 75 and older. Arch Gerontol Geriatr. 2006;42(1):21-33.
  8. World Health Organization. Towards a Common Language for Functioning, Disability and Health ICF. World Health Organization, Geneva, Switzerland 2002.
  9. Persons With Disabilities Act 2008 (Act 685): Laws of Malaysia (2008), Kuala Lumpur, Malaysia.
  10. Lawton MP, Brody EM. Assessment of older people: self-maintaining and instrumental activities of daily living. Gerontologist. 1969;9(3):179-86.
  11. Abdulraheem IS, Oladipo AR, Amodu MO. Prevalence and correlates of physical disability and functional limitation among elderly rural population in Nigeria. J Aging Res. 2011:369894. doi: 10.4061/2011/369894.
  12. Boggatz T, Dassen T. Ageing, care dependency, and care for older people in Egypt: a review of the literature. J Clin Nurs. 2005;14(8B):56-63.
  13. Wilson D, Consolvo S, Fishkin K. In-home assessment of the activities of daily living of the elderly. In Extended Abstracts of CHI 2005: Workshops - HCI Challenges in Health Assessments. 2005.
  14. Siop S, Verbrugge LM, Hamid TATA. Disability and quality of life among older Malaysians. Population Association of America, 2008 Annual Meeting Program, Presented in Session 173: Disability and Active Life Expectancy in Asian Populations. 2008.
  15. Latifah D, Al-Saji AA, Winn T, Hamzah AM. Physical disability in instrumental's activities of daily living (IADLs) and its associated factors among elderly in Kota Bharu, Kelantan. National Drug Research Centre, Universiti Sains Malaysia. 2013.
  16. Tas U, Verhagen AP, Bierma-Zeinstra SM, Hofman A, Odding E, Pols HA, et al. Incidence and risk factors of disability in the elderly: the Rotterdam Study. Prev Med. 2007;44(3):272-8.
  17. Yoshida D, Ninomiya T, Doi Y, Hata J, Fukuhara M, Ikeda F, et al. Prevalence and causes of functional disability in an elderly general population of Japanese: the Hisayama study. J Epidemiol. 2012;22(3):222-9.
  18. Gupta P, Mani K, Rai SK, Nongkynrih B, Gupta SK. Functional disability among elderly persons in a rural area of Haryana. Indian J Public Health. 2014;58(1):11-6. doi: 10.4103/0019-557X.128155.
  19. Peduzzi P, Concato J, Kemper E, Holford TR, Feinstein AR. A simulation study of the number of events per variable in logistic regression analysis. J Clin Epidemiol. 1996;49(12):1373-9.
  20. Wade DT, Collin C. The Barthel ADL Index: a standard measure of physical disability? Int Disabil Stud. 1988;10(2):64-7.
  21. Shah S, Vanclay F, Cooper B. Improving the sensitivity of the Barthel Index for stroke rehabilitation. J Clin Epidemiol. 1989;42(8):703-9.
  22. Kua EH, Ko SM. A questionnaire to screen for cognitive impairment among elderly people in developing countries. Acta Psychiatr Scand. 1992;85(2):119-22.
  23. Rashid AK, Azizah AM, Rohana S. Cognitive impairment among the elderly Malays living in rural Malaysia. Med J Malaysia. 2012;67(2):186-9.
  24. Minhat HS, Rahmah MA, Khadijah S. Continuity theory of ageing and leisure participation among elderly attending selected health clinics in selangor. Med J Malaysia. 2013;12(3):51-58.
  25. Yesavage JA, Brink TL, Rose TL, Lum O, Huang V, Adey M, et al. Development and validation of a geriatric depression screening scale: a preliminary report. J Psychiatr Res. 1982-1983;17(1):37-49.
  26. Sherina MS, Rampal L, Mustaqim A. Factors associated with chronic illness among the elderly in a rural community in Malaysia. Asia Pac J Public Health. 2004;16(2):109-14.
  27. Hairi NN, Bulgiba A, Cumming RG, Naganathan V, Mudla I. Prevalence and correlates of physical disability and functional limitation among community dwelling older people in rural Malaysia, a middle income country. BMC Public Health. 2010;10:492. doi: 10.1186/1471-2458-10-492.
  28. Tabachnic BG, Fidell LS. Using multivariate statistics. (5th edn), Allyn & Bacon, Needham Heights, MA. 2007.
  29. Shahar S, Ibrahim Z, Fatah AR, Rahman SA, Yusoff NA, Arshad F, et al. A multidimensional assessment of nutritional and health status of rural elderly Malays. Asia Pac J Clin Nutr. 2007;16(2):346-53.
  30. Department of Statistics Malaysia (2014) Malaysia @ a Glance.
  31. Luppa M, Luck T, Weyerer S, Konig HH, Brahler E, Riedel-Heller SG. Prediction of institutionalization in the elderly. A systematic review. Age Ageing. 2010;39(1):31-8. doi: 10.1093/ageing/afp202.
  32. Koenig HG. Religion, spirituality, and health: the research and clinical implications. ISRN Psychiatry. 2012;2012:278730. doi: 10.5402/2012/278730. Print 2012.

Top ↑

Copyright: © 2015 Sharifa EBW. Puteh, et al. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.