Quality of life among community-dwelling older adults: evidence from a large population-based study in rural Sri Lanka

Against the backdrop of the ever-increasing aging population in Sri Lanka and the scarcity of local evidence on quality of life (QoL) among rural elderly, this study was conducted to assess the QoL of the community-dwelling older adults in rural Sri Lanka. This cross-sectional study was conducted among community-dwelling older adults (60–74 years) in a selected rural setting in Sri Lanka. K-means cluster analysis was used to stratify participants into 'low' and 'high' levels of QoL and then significant associations between these clusters and underlying socio-demographic and self-reported health related factors were estimated using bivariate and subsequent multivariable binary logistic regression models. The final sample consisted of 3573 community-dwelling older adults (response rate 97.8%). The mean (SD) age of the sample was 66.7 (4.3) years and the majority were females (n = 2130, 59.6%). Amongst the six QoL domains assessed (physical, psychological, social, functional, environmental and spiritual domains), the highest and the lowest mean (SD) scores were reported for the functional [63.4 (16.9)] and the physical [52.9 (15.0)] domains, respectively. Aged 70 years or more, either unmarried/widowed/divorced, lower educational levels and having chronic illnesses were statistically significant associations of QoL (p < 0.05). The QoL among community-dwelling older adults in rural Sri Lanka is moderate. As having social support, absence of chronic diseases and good education level were found to be associated with better QoL, strengthening community-based interventions to improve these aspects by incorporating the evidence generated by other longitudinal studies is recommended.


Introduction
Worldwide, people are living longer. The rapid aging of the population is one of the major problems affecting the health, economy and social welfare of countries around the world irrespective of the development status of the country [1,2].
As the aging population increases rapidly and as older adults are vulnerable to many health, social, emotional and economic instabilities, population aging has gained an important place on the global agenda. A plethora of global evidence reveals that chronic illnesses have a negative impact on the quality of life (QoL) among older adults [3][4][5][6][7]. In addition, studies suggest that the QoL in older adults can decline as a consequence of multiple contributory factors such as reduction in autonomy, restrained independence, lack of social relationships, apart from declining health and functional status [8].
World Health Organization (WHO) defines QoL as an individual's perception of their position in life in the context of the culture and value systems in which they live and concerning their goals, expectations, standards and concerns [9]. There are several QoL assessment instruments developed across the globe. Apart from many generic instruments used widely such as the WHO Quality of Life instrument (WHOQoL-Bref), QoL Scale (QLOS), McGill QoL Questionnaire (MQOL) and Global Quality of Life Scale, there are instruments specific for the older populations such as Elderly Quality of Life Index (EQOLI) and Older People's Quality of Life (OPQOL) instrument.
Worldwide, there are a number of studies conducted on QoL in older populations and its determinants using diverse tools. The global evidence on QoL has steadily increased over the recent years with the cross-cultural adaptation and validation of assessment instruments in different languages [10]. There are several systematic reviews related to different aspects and associations of QoL of the older populations. Systematic reviews suggest significant associations between poor QoL of the older populations and; frailty [11], multimorbidity [3], dementia [12], cognitive dysfunctions [13], lower physical activity [14] malnutrition [15] as well as all-cause mortality [16]. However, owing to the subjectivity and multidimensionality of the concept of QoL, it is often difficult to make direct comparisons across different study findings.
Sri Lanka, a lower-middle-income country, has one of the fastest aging populations in Asia [17] and 15.1% of the Sri Lankan population is over 60 years of age [18]. Owing to this trend in demographic transition in Sri Lanka, it is projected that one in four Sri Lankans will be older adults by the year 2041 [19]. Consequently, Sri Lanka is expected to face profound challenges concerning health and social care service provision as well as the economic and social development of the country.
With regard to the local evidence on health and QoL in older adults, there are a few studies conducted in community settings as well as in healthcare settings. Local evidence suggests that not only the physical illness such as diabetes mellitus, hypertension, asthma and arthritis are very common among Sri Lankans aged over 60 years [20], but also mental health issues and other disabilities, visual and hearing impairment are common among Sri Lankan older population [21,22].
According to a study conducted in an urban setting in the Western Province, Sri Lanka, health-related QoL among elders is low while individual and household factors both played an important role in deciding the level of QoL [23]. Low QoL of the older population in Sri Lanka was reported in other studies conducted in Sri Lanka [24,25]. Male gender, better education, being married, living with a spouse, fewer health problems and perceived adequacy of income determined better QoL according to a health-centre based study conducted in the Western Province, Sri Lanka [4]. A community-based study conducted in the North Central Province, Sri Lanka revealed that living alone, low family income, the presence of chronic kidney diseases and poor self-rated health lead to poor QoL [26].
Even though there are a few studies conducted in different parts of Sri Lanka to assess the QoL of the older adults, these studies have often used a small sample; thus, limiting the generalizability. Further, most of the studies have been conducted either in urban or semi-urban settings. Given that approximately 80% of the Sri Lankan population is in rural areas, it is important to assess the health status in rural communities with a view to exploring any health inequities.
Evidence on health status, health care needs including QoL of the ever-increasing older population in the country is of utmost importance to facilitate planning health services for the older adults. Against the backdrop of the rapidly increasing aging population in Sri Lanka and the lack of large community-based studies on QoL of older adults in rural settings, this study was conducted in a selected rural setting, North Central Province, Sri Lanka to assess the QoL and its associations of older adults.

Methods
The Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) Guidelines has been used in the preparation of this paper.

Study design and setting
This community-based, cross-sectional study was conducted in the Thalawa Divisional Secretariat (DS) area, Anuradhapura District, North Central Province, Sri Lanka ( Fig. 1). Anuradhapura District is the capital city of the North Central Province, the largest province in Sri Lanka. The population of Anuradhapura District is approximately 860,000 with 94.1% of them being rural population [27]. Anuradhapura District has mainly an agriculture-based economy with 46.7% of the total workforce employed in agriculture [28]. The life expectancy at birth in Anuradhapura District is 70.5 years for males and 77.8 years for females in comparison to the national figure of 72 years for males and 78.6 years for females [29].
Anuradhapura District has 22 administrative divisions as DS areas and the Thalawa DS area is one of them [30].

Participants
The Thalawa DS area is sub-divided into 39 smaller administrative divisions named Garama Niladhari divisions. The total population in the Thalawa DS area is approximately 69,000 with 51% of females [31]. The population over 60 years is approximately 7300 in the area. The study population of this study consisted of all community-dwelling older adults aged between 60 and 74 years.

Measures
The questionnaire was primarily used as a self-administered questionnaire; however, for the participants, who could not read and write, a trained data collector facilitated the completion of the questionnaire. The questionnaire was developed and validated in Sinhala and it consisted of two sections. The first section included questions on basic demographic characteristics and the second section was the Quality of Life Instrument for the Young Elderly in Sri Lanka (QLI-YES), which is a questionnaire developed and validated in Sri Lanka, used to assess the QoL of the older adults aged 60-74 years [32]. The QLI-YES consists of 28 items, which assesses six domains, viz., physical domain (e.g. "To what extent do you feel that physical pain and discomforts interfere with what you need to do?"), mental domain (e.g. "How much of use were you to your family?"), social domain (e.g. "How satisfied are you with the recognition you receive from family and society?"), functional domain (e.g. "How satisfied are you with your ability to perform your day to day work?"), environmental domain (e.g. "How satisfied are you with the access to, and quality of health care received?"), and spiritual domain (e.g. "How satisfied are you with the opportunities and time you get to engage in religious activities?"). In the questionnaire, 4,6,8,4,3 and 3 items are targeted at assessing physical, psychological, social, functional, environmental and spiritual domains, respectively. The responses by the participant to each questionnaire item is recorded in a five-point Likert scale (1 = Not at all, 2 = A little, 3 = Moderately, 4 = Very much, 5 = Extremely).
The QLI-YES was found to be a valid and reliable instrument to assess QoL in the older adults. The validation study established satisfactory fit indices in confirmatory factor analysis and ensured criterion validity assessed in comparison with three instruments, viz., activities of daily living/instrumental activities of daily living (ADL/IADL), WHOQoL BREF and the Abbreviated Mental Test Score (AMTS). Further, the internal consistency was reported as Cronbach's α 0.93 [32].

Procedure
Ethical clearance to conduct the study was obtained from the Ethics Review Committee of the Faculty of Medicine and Allied Sciences, Rajarata University of Sri Lanka (Reference no: ERC/2019/06). The updated details of the adult population in the Thalawa DS area were obtained from the Voter's list, which provided details at household level. In the Thalawa DS area, there are ten Community Health Promotion Officers (HPO) employed in the Ayurvedic Community Health Promotion Service, whose primary responsibility is to conduct health promotion activities (including geriatric care) in their assigned areas and they are frequently engaged in filed activities often visiting households. These ten HPOs were trained for data collection and they distributed the questionnaires at the household level as well as in identified settings such as village level societies covering all eligible older adults in the area. Informed written consent was obtained before the completion of the questionnaire and in instances, where the participants could not read and write, informed verbal consent was obtained. As mentioned above, the questionnaire was prepared in a manner, which could be either administered by the participants themselves or facilitated by the data collectors if needed. Confidentiality of data collected was adhered to strictly and the anonymity of the participants was maintained.

Data analysis
Data analyses were carried out using SPSS version 25.0. The scoring of the QLI-YES was done according to the guidelines given by the original authors [32]. Accordingly, Fig. 1 Study setting the raw domain scores of the six domains were calculated by straightforward summative scaling of constituent items of respective domains. The negatively worded items were reverse scored prior to summation. Raw domain scores were transformed to a 0-100 scale and the higher scores denote higher quality of life. The transformation converted the lowest possible score to zero and the highest possible score to 100. Scores between these values represented the percentage of the total possible score of the domain achieved. Raw scores were transformed using the following formula.
The summary statistics for each domain score were computed with measures of central tendency (mean and median) and dispersion [standard deviation (SD)] and interquartile range (IQR)]. Further, these values were presented for different sub-groups based on socio-demographic characteristics. Bivariate correlation analysis was carried out to assess the relationship between the mean scores of the different QoL domains. In addition, K-means cluster analysis was carried out to compute two clusters of participants based on the standardized scores of the six domains of QoL. For the assessment of associated factors of QoL, a two-step procedure was followed. Categorical data related to associated factors were amalgamated rationally as dichotomous variables where necessary for the bivariate analysis. The crude odds ratios (OR) were calculated as the measures of effect with 95% Confidence Interval (95% CI) in the bivariate analysis using a series of binary logistic regressions. Associated factors that showed statistical significance at a p value less than 0.05 in the bivariate analysis were included in the multivariable analysis and binary logistic regression model with the backward stepwise elimination method was used. The model produced adjusted odds ratios (AOR) and 95% CI with the significance level for variables of interest.

Results
Altogether, 3652 participants aged 60-74 years were enrolled in this study and out of them, 3573 have completed the questionnaires. Thus, the response rate was 97.8%. Table 1 illustrates the basic characteristics of the sample. The mean age of the participants was 66.7 years (SD = 4.3 years). The majority of the sample was females (n = 2130, 59.6%), Buddhists (n = 3471, 97.1%) and married at the time (n = 2666, 74.6%). In the sample, 14.7% (n = 525) of participants did not have any formal education and approximately half of the sample did not have regular financial assistance (n = 1703, 47.7%).

Quality of life in older adults
The QoL scores of the six domains (in a 0-100 range scale) are given in Table 2. Amongst the six domains, the highest mean score was reported for the functional domain (63.4, SD = 16.9), whereas the lowest mean score was reported for the physical domain (52.9, SD = 15.0).
The results of the bivariate correlation analysis between the mean scores of the six domains of QoL are summarized in Table 3.
Correlation analysis revealed that all six domains had statistically significant correlations with each other (p < 0.001).
Two clusters of community-dwelling older adult participants were computed using the K-means cluster analysis based on the standardized scores of the six domains of the QoL, named low QoL (n = 1924, 53.8%) and high QoL (n = 1649, 0.46.2%). Further, the ANOVA test indicated that all six domains significantly contributed to determining the above clusters (p < 0.001) and the final cluster centres of the standardized scores of all six domains of QoL are illustrated in Fig. 2.
The statistically significant associated factors of QoL were assessed using multivariable analysis following the bivariate analysis and the results of both analyses are summarized in Table 4.
Multivariable analysis elicited several statistically significant associations of low QoL when controlled for other factors included in the model (p < 0.05). Community-dwelling older adults aged 70 years or more were more likely to report low QoL in comparison to those who aged less than 70 years (p < 0.001). Further, older adults, who were either unmarried/widowed/divorced, had greater likelihood to report low QoL in comparison to those who were married (p = 0.004). In addition, it was noted that older adults, who had lower educational levels with only up to primary education were were more likely to report low QoL in comparison to those who had studied beyond primary education (p < 0.001). Furthermore, community-dwelling older adults who had reported diabetes mellitus (p = 0.049), hypertension (p = 0.013), arthritis/joint diseases (p < 0.001), chronic kidney disease (p < 0.001) and bronchial asthma (p = 0.004) were also more likely to report low QoL in comparison to those who did not report those respective illnesses.

Discussion
The present study assessed QoL and its associated factors in a large sample of community-dwelling older adults living in a rural setting in Sri Lanka. To the best of our knowledge, this is the largest Sri Lankan study on QoL of the older adults conducted in a community setting. The present study used QLI-YES, which is a study instrument developed and validated in Sri Lanka; thus, it enabled to capture the culturally appropriate domains of QoL in the older adults in Sri Lanka.
The distribution of most of the socio-demographic characteristics of the sample was similar to the national figures of Sri Lanka. The study sample has a female predominance, which is compatible with the sex ratio among Sri Lankan older population [33]. According to the Population and Housing Census in Sri Lanka, the married proportion in the 60-74 years cohort was approximately 75%, which is consistent with the present study [27]. However, the distribution of religion was different from that of the national figure; yet, the distribution is more similar to that in the Anuradhapura District [27].
It is noteworthy to observe that approximately, threequarters of the older adults had reported at least one diagnosed chronic disease. This figure is higher than the corresponding figure of 55.2% reported in a national survey on self-reported health [20]. However, the same survey stated that a considerable proportion who denied having any chronic diseases have not undergone medical checkups for chronic diseases. It is also important to appreciate that Anuradhapura District has a high prevalence of chronic kidney disease of unknown origin, which could have led to the above difference.
The present study revealed that community-dwelling older adults experience moderate QoL in all six domains, viz., physical, psychological, social, functional, environmental and spiritual domains. A study conducted among a small sample of 140 older adults attending a Family Practice Centre in a semi-urban area, Western Province, Sri Lanka, has used the same study instrument to assess the QoL and reported similar scores for all six domains of QoL. Similar to the present study, the scores of all six domains were approximately 60 with spiritual and functional aspects having the highest scores while the physical aspect had the lowest score [4]. The original study on the development of the QLI-YES had reported higher mean scores for all six domains of QoL than the scores reported in the present study. These scores were reported by a sample of 200 older adults living in both urban and rural settings in Kotte, Western Province [32]. A previous community-based study conducted in Thalawa, Anuradhapura-the same study setting-concluded that the majority of the older people experience a moderate level of QoL. These study findings were based on the responses of 336 older people using the OPQOL questionnaire [26]. Though the difference in the assessment tools across the studies limits direct comparison of findings, the similarity in the level of QoL is noted.
On the contrary, a community-based study conducted in Colombo, Western Province, a more urban setting, concluded that QoL among the older adults was relatively low  using the WHO QOL-BREF questionnaire [23]. Further, another community-based study conducted among older adults in the Kandy District, Central Province reported low QoL and this study had used EQ-5D-3L to assess the QoL [24,25] In a similar vein, another community-based study concluded that the QoL of the older population in Jaffna District was found to be average using the WHOQOL-Bref questionnaire [21]. In addition to the local studies, studies conducted in India revealed that QoL in older adults are either poor [34,35] or average [36] in rural settings. In addition, evidence also suggests that older adults in rural settings reported relatively better QoL than older adults in urban settings in India [37]. The present study findings are also in line with these regional evidence.
While systematic differences in the assessment tools and methods used and the differences that stem out from the conceptualization of QoL across different settings may have contributed to the observed differences in QoL between these studies and the present study, contextual factors related to urban and rural settings could have also contributed substantially to the differences.
One of the possible reasons for the relatively higher QoL in the present study in comparison to urban settings could be its rural setting with its distinct agricultural lifestyle, which facilitates more opportunities for the older community for active social engagement and positive interactions. Evidence suggests that social participation is associated with lower levels of old-age morbidity and mortality [38]. Social gatherings, cultural and religious events, which are part and parcel of rural settings provide more opportunities for the elders to enjoy life with others, securing their social inclusiveness [39]. Such community-level organizations in Sri Lanka include farmers' societies, death donation societies, women's societies where mostly elderly communities lead and actively participate. In this context, the social capital, which considers ones social network as an asset [40], is expected to be stronger in this rural setting. Moreover, coresidence exists in rural areas more frequently, compared to urban areas where family solidarity is much stronger [41]. In contrast to the rural settings, the lifestyle in urban settings shuts out the opportunities for the older adults to have social support networks. Social isolation is more prevalent in urban areas causing many older adults to end up in elderly homes. According to the Department of Social Service records, institutionalized older adults have increased sizably in Sri Lanka within the past two decades, in urban areas [41]. It may also be argued that the perceptions, values, and norms on defining 'health and wellbeing' may be different in these communities, which anyway needs to be explored using more qualitative studies.
The findings of the present study on the significant associations of QoL among the older adults are in line with available local and global evidence. In this study, low QoL was associated with not having an immediate level of social support (marital status) [4,26,42,43], having chronic diseases [4][5][6][7] and lower education level [4,44,45].

Strengths and limitations
Given that the present study is the largest Sri Lankan study carried out in a community setting ncluding all different oldage groups, the findings of this study permit greater generalizability. In addition, it is important to note that due to the subjective nature of the QoL concept, the assessment of QoL is often carried out using self-reported measures and this study has used a study instrument developed and validated in Sri Lanka enabling to capture the QoL appropriate to the Sri Lankan context.
However, one limitation of this study is that the data are of self-reported illnesses which we believe can cause both over and underestimation of results. In addition, this study only included young elderly in the age range of 60-74 years; thus, the status of the older age groups is not represented in the study. The fact that this study did not assess healthrelated behaviours such as smoking, alcohol, diet and physical activity, which influence QoL, is one of the main limitations Furthermore, as this is a cross-sectional study, there is no possibility to attain a causal relationship between QoL and the studied factors.

Recommendations
As the findings suggests that having social support and good social relations [4,26,42,43], absence of chronic diseases [4][5][6][7] and having a good education level [4,44,45] were found to be associated with better QoL, it is of utmost importance to establish community-based interventions to improve these aspects. These interventions need to be developed by incorporating the evidence generated by studies suggesting the causal relationship of these associations with QoL. As Sri Lanka is experiencing population aging at a rapid pace, necessary policy decisions should be taken at all levels to facilitate timely action. It is well known that more opportunities for social networking are an inherent culture in rural communities compared to urban settings [46]. Strengthening the existing community networks in rural settings while identifying and establishing similar social network settings in urban areas would be an effective strategy to upgrade the QoL of the older adults in community settings. Educating the youth and school children about healthy aging and the importance of providing more opportunities for older population for active social engagement and participation would encourage the youth to create more opportunities for older adults for active social participation. This would be advantageous to utilize the vast experiences of the older population effectively. Further, it would sensitize the youth to be prepared for healthy aging in the future. More qualitative and comparative studies to assess the significance of utilizing health assets to achieve healthy aging are highly recommended. As chronic disease burden was found to be very high among the older population, limiting factors for health care access for the older population should be identified and rectified. Furthermore, underutilized health care services established specifically for older adults should be streamlined and popularized in the community as well.