A cross-sectional study on quality of life among the elderly in non-governmental organizations’ elderly homes in Kuala Lumpur

This cross-sectional study was conducted in eight elderly homes in Kuala Lumpur between September to November 2014. These non-profit homes are run by non-governmental organizations. Residents of these homes are mostly elderly citizens abandoned in hospitals in Kuala Lumpur. Other residents are neglected or poor elderly citizens without a shelter. Residents are referred to these homes by the hospitals, family members, neighbors, and sometimes by the Department of Social Welfare Malaysia.

Unlike nursing homes and retirement homes that charge monthly residential fees, these homes provide residential services without a fee. They rely enormously on philanthropy of the public and they do not provide nursing services. Hence a basic criteria for admission into these homes is the ability to perform activities of daily living. These elderly homes provided accommodation in either the ward-type accommodation or the twin-sharing accommodation. The twin-sharing accommodation had two people (two beds) in a room, the ward-type accommodation had over two people (over two beds) in a room. In addition, unlike the ward-type, residents in the twin-sharing accommodation can close their door when they need privacy, and they can easily control who goes in and out of their room. On the other hand, the ward-type accommodation was like a large hall with several residents. The size of the homes varied from 16 to 59 beds. The number of staff also varied from 4 to 10. These homes had shared toilets and bathrooms, a kitchen, common dining area for meals, a gathering place for recreational and leisure activities like watching TV, playing cards, listening to music among others. These homes organize routine indoor exercise sessions in the morning and some are fitted with exercise equipment with the stationary bicycle the most common. Medical care is usually provided by volunteers who visit occasionally and are most times limited to physical examination without radiological and laboratory investigations. However these homes when they can afford it endeavor to take residents to hospitals. Occasionally outdoor activities are organized by the homes or volunteers and could include visit to a park, place of worship, or to a local theatre to see a movie. The homes lacked sufficient space for gardens, and wardrobes and libraries were also absent.

Using a standard deviation of 10 for quality of life scores in elderly population, margin of error as 1.5, and a confidence level of 95 %, the minimum sample size for this study was determined to be 170. The total sample size became 203 after adding a non-response rate of 19 %. During the post study survey 13 elderly homes were identified in Kuala Lumpur. A spot map was used to identify the geographical locations of the homes. The homes were numbered serially on the map in a clockwise fashion beginning from the home farther north. Using the lottery method, simple random sampling was used to select eight homes. Eight homes were selected because it was more than half of the identified elderly homes, thus would be a good representation of the target population, and also because the number of residents in the eight homes was up to the estimated sample size. A total of 203 residents were selected by stratified (proportionate) random sampling. Firstly the proportion of participants to be selected from each home was determined by dividing the number of residents in that home by the total number of residents in the eight homes, then multiplying the fraction by 100. Secondly the total number of participants to be selected from each home was determined by dividing the proportion of residents to be selected in each home (obtained from the first step) by 100, then multiplying the fraction by the estimated sample size (203). Participants in each home were then selected randomly from a sampling frame of each home using the lottery method. The inclusion criteria included those aged 60 years and above, those who were able to understand English, Malay, or Chinese, and those who gave written consent to participate. Residents aged less than 60 years, unable to communicate, and those with cognitive impairment as indicated by their files in the homes were excluded from the study. Retirement homes, private homes, and homes that provided nursing services were not included in this study.

A structured close-ended questionnaire was used in this study. The first part included questions on socio-demographic variables such as; age, gender, ethnicity, religion, marital status, number of children, level of education, previous employment sector, pension, socio-economic status, outdoor leisure activity, smoking status, alcohol status, physical activity, duration of stay in home, type of accommodation, co-morbidities, and self-rated health status. Outdoor leisure activity was defined as leisure activities outside the home such as a visit to a park, movie theatre, among others at least twice a month excluding visit to health centers for medical treatment. Physical activity was defined as 150 minutes of moderate intensity exercise or at least 75 minutes of vigorous- intensity exercise weekly [18], for type of accommodation, two types of accommodation were available; the twin-sharing accommodation had two people (two beds) in a room while the ward-type accommodation had over two people (over two beds) in a room. For self-rated health, participants were asked whether they perceived their general health as good or poor. Economic status was determined by asking participants to rate their current economic status as good, intermediate or poor. This method was used since participants were currently unemployed. It has been used in a previous study [19], and it is known to be a reliable means of obtaining such information as people can correctly grade their economic status with reference to standard of living in their community and compare themselves with others.

Quality of life was measured with the validated World Health Organization Quality of Life Instrument-Brief Version (WHOQOL-BREF). It evaluates perceived quality of life using 26 items categorised into Physical domain (7 items), Psychological domain (6 items), Social Relations domain (3 items), and Environment domain (8 items). Two items evaluates perception of general health and quality of life. Each item is ranked on a 5-point Likert scale. Higher scores indicate higher quality of life [5]. The English version of the instrument was validated in a previous study [20] and was found to have Cronbach’s alpha ranging from 0.68 to 0.82 across the four domains. The instrument also demonstrated good validity. The Malay version was validated in a previous study [21], and the Cronbach’s alpha ranged from 0.64 to 0.80 across the four domains. The Chinese version of the instrument have also been validated in a previous study [22], the Cronbach’s alpha ranged from 0.67 to 0.78 across the four domains. The multidimensional Scale for perceived social support (MSPSS) was used to measure social support. This instrument was developed and validated by Zimet et al. in 1988 [23]. It measures perceived availability of support. It has 12 items, and it measures social support from three sources of support: Family, Friends and Significant others (special person). Each source of support has 4 items. Each item is ranked on a 7-point Likert –Scale from 1: very strongly disagree to 7: very strongly agree. On the MSPSS, higher scores indicate higher level of social support. A total score of 69–84 indicate high level of perceived social support, 49–68, and 12–48 indicate moderate and low levels of social support respectively [24]. The English version was validated in previous studies [23, 25] and was found to have Cronbach’s alpha of up to 0.92. The instrument also demonstrated good validity. The Malay version was validated in a previous study [26], and was found to have a Cronbach’s alpha of 0.89. The Chinese version was validated in a previous study [27] and it demonstrated good psychometric properties with Cronbach’s alpha of 0.92 and test retest reliability of 0.71. All language versions of study instruments used in this study have demonstrated good psychometric properties. Data collection was by face to face interview. The questions were administered in Chinese, Malay and English languages by a trained native speaker of each language. A pilot study was conducted before the actual study commenced.

The research and ethical committee of International Medical University Malaysia approved this research in 2014 (Project Identification Number; MScPHI1/2014(09). Permission was obtained from the administrators of each home to access the homes. All participants were given detailed information about the study objectives and confidentiality of the information, and a written consent was signed by each participant.

Statistical Package for Social Sciences (SPSS) version 20.0 for windows was used to analyze the collected data. The results of continuous variables were expressed as means and standard deviations, while categorical variables were expressed as proportions and frequencies. Some continuous variables were dichotomized; social support was dichotomized into low level of social support and medium/high levels of social support, marital status was dichotomized into married and single (unmarried, separated, and widowed), level of education was dichotomized into none/primary and secondary/tertiary education.

T-test was used for univariate analysis, and only variables with P-value less than 0.05 was added to the multiple linear regression model. This model was used to identify the predictors of quality of life and P-value less than 0.05 was considered statistically significant.

The age of the study participants ranged from 60 to 95. As shown in Table 1, of the 203 participants, 64.5 % were males and 35.5 % were females, 46.3 % were aged 60 to 69 years, and 53.7 % were aged 70 years or more. The mean age was 71.5 (±6.8). Majority of the respondents were Chinese (87.1 %). Most of the respondents were married (61.6 %) and 38.4 % were single, 57.6 % of the respondents had 1 child and 42.4 % had 2 children or more. Majority of the respondents had secondary education (46.8 %), while 39.9 % had primary or no education. Most of the respondents previously worked in the private sector (61.1 %), while 29.5 % were self-employed, and 3 % were unemployed. Most of the respondents were not receiving pension (90.6 %). Only 6.4 % perceived their economic status as good, 8.4 % reported an intermediate, and 85.2 % reported poor economic status. Most of the respondents engaged in outdoor leisure activities (78.3 %), and 31.5 % were physically active. Smoking and alcohol drinking was reported by 23.6 % and 7.9 % of respondents respectively. Majority (54.2 %) had lived in an elderly home for less than 2 years. Most of the respondents lived in ward-type accommodation (90.1 %) and 9.9 % lived in twin-sharing accommodation. Majority (82.8 %) had at least one chronic co-morbidity. Only 5.4 % reported high level of social support, while low and medium levels of social support was reported by 47.8 % and 46.8 % respectively.

Table 2 shows the mean scores for the quality of life domains. The mean (SD) for the physical domain was 14.3 (±2.7), 13.7 (±2.5) for the psychological domain, 10.8 (±3.4) for the social domain, and 13.0 (±2.5) for the environment domain.

The univariate analysis in Table 3 shows that among the socio-demographic variables, gender was significantly associated with all domains of QOL (p < 0.05). Age was significantly associated with the physical domain (p = 0.010). Level of education was significantly associated with the physical domain (p = 0.006), psychological domain (p = 0.009) and social domain (p = 0.007). Economic status was significantly associated with the physical domain (p = 0.037), psychological domain (p = 0.008), and social domain (p = 0.019). Duration of residence was significantly associated with psychological domain (p = 0.001), social domain (p = <0.0001) and environment domain (p = <0.0001). Type of accommodation was significantly associated the psychological domain (p = 0.001), social domain (p = 0.081), and environment domain (p = 0.004).

Tables 4, 5, 6 and 7 show the results of the multivariate analysis. Variables significantly associated with the physical domain of quality of life included; age (p = 0.021), gender (p = <0.0001), economic status (p = 0.037), outdoor leisure activity (p = <0.0001), chronic co-morbidities (p = 0.020) and social support (p = 0.013).

The psychological domain was significantly associated with gender (p = 0.003), level of education (p = 0.04), economic status (p = 0.013), outdoor leisure activity (p = 0.036), chronic co-morbidities (p = 0.045) and social support (p = <0.0001).

The social domain was significantly associated with gender (p = 0.039), level of education (p = 0.030), outdoor leisure activity (p = 0.012), duration of residence (p = <0.0001), chronic co-morbidities (p = 0.011) and social support (p = <0.0001). Chronic co-morbidities (p = 0.044) and social support (p = <0.0001) were significantly associated with the environment domain.