Biosocial and disease conditions is associated with good quality of life among older adults in Eastern Nepal

The ageing population in most low-and middle-income countries is accompanied by an increased risk of non-communicable diseases culminating in a poor quality of life. However, the factors accelerating this poor quality of life (QOL) have not been fully examined. Therefore, this study examined the factors influencing the quality of life of Nepali older adults. Data from a previous cross-sectional study, conducted between January and April 2018 in eastern Nepal, was used. The analytical sample included 794 older adults aged ≥60 years, selected by a multi-stage cluster sampling approach. QOL was measured using the Older People’s Quality of Life tool; dichotomized as poor and good QOL. Other measures used included age, gender, ethnicity, religion, marital status, physical activity and chronic diseases such as osteoarthritis, cardiovascular disease, diabetes, chronic obstructive pulmonary disease (COPD), and depression. The factors associated with quality of life were examined using mixed-effects logistic regression.

may worsen their QOL relative to those benefiting from those programs. Previous studies on QOL among Nepalese elderly have been limited to the outpatient clinics 14 and nursing homes 15 in Nepal's capital city of Kathmandu. To achieve the SDG, it is important to identify the key factors influencing the health and QOL of the older population, and operationalise appropriate policy mechanisms to improve the QOL of the elderly. Therefore, this current study aimed to 1) identify the state of QOL among Nepali older adults and 2) highlight the factors associated with QOL.

Methodology Study designs and participants
Data from a previously conducted community-based cross-sectional study was used 16 . The study was conducted among older adults, aged ≥ 60 years, living in the rural region of Morang and Sunsari districts of Nepal. The data collection took place between January-April 2018. The sample size of 847 was calculated based on an unknown prevalence formula, i.e., Z 2 /4d 2 17 . Here, sampling error = 5.0%, and Z = 1.96 at 95%CI. Further, a design effect of two and 5% non-response rate was added. A total of 794 eligible participants agreed to be interviewed in the study resulting in a response rate of 93.7%.
Study participants were recruited from the community setting using a multi-stage cluster sampling approach. The detail on the sampling method has been fully described in our previously published work 16 .

Data collection and study variables
We used both semi-structured interviews and a validated survey questionnaire to collect data from the participants. The English version of the questionnaire was first translated to Nepali and then translated (forward-backward translation) back to English by two researchers to check the consistency of the instrument.
Outcome Variable QOL of the older adults was assessed by the Older People's Quality of Life (OPQOL) questionnaire 18 , which is a novel instrument specifically designed to measure the QOL of older adults 19 . The OPQOL questionnaire has 35 questions that asked the participant to indicate the extent to which he/she agrees with each item in the Likert scale response (i.e., "strongly disagree", "disagree", "neither agree nor disagree", "agree" and "strongly agree"). Each of the five possible answers is scored between one ("strongly disagree") and five ("strongly agree"). The 35 items of this instrument consider the following aspects of QOL: life overall, health, social relationships and participation, independence, control over life and freedom, home and neighbourhood, psychological and emotional well-being, financial circumstances, leisure, activities, and religion. The cumulative score of the 35 items, which ranged from35 to 175, provides the measure of overall QOL; with higher scores indicating a better QOL. In this study, Cronbach's alpha for the OPQOL instrument was 0.75, which indicates acceptable reliability of the tool.

Co-variates Measurement
Independent variables included were age; gender; ethnicity; religion; marital status; living arrangement; literacy status; occupation; monthly personal income; smoking habit; alcohol drinking habit; tobacco chewing habit; physical activity and presence of any co-morbidities. These co-variates are described in the published paper authored by Yadav et al. 20 .

Ethics
The study was approved by the Institutional Review Board of Nepal Health Research Council, Government of Nepal, Ministry of Health, Kathmandu. Prior to the interview, written informed consent was obtained from all literate participants, and thumb impressions were obtained from illiterate participants. Participants received an explanation about the study objectives and procedures and voluntary participation.

Statistical analysis
We employed descriptive, bivariate and multivariable regression models for this study. First, descriptive analysis was carried out to present the distribution of background characteristics.
Frequency, percentage, mean, standard deviation (SD) and range distributions are presented. For the bivariate analysis, the chi-square (χ2) test was performed to compare the percentage of participants with different QOL within different categories of variables at a 5% level of significance. Considering the nested nature of the survey data with possible variations among clusters (municipality), we performed a mixed-effect logistic regression model to assess the true association between the QOL and associated factors. Cluster variation was considered as a random effect and the rest of the variables were considered as fixed effects. The generalized estimating equation (GEE) was undertaken to estimate the parameters of the model while the exchangeable correlation structure within the clusters was employed. We retained in the final model only variables with a p-value of less than 0.25 in the bivariate model. Both unadjusted and adjusted odds ratios (ORs) are reported with 95% confidence intervals (95% CI). All analyses were performed using Stata v. 13.0 (Stata Corp, College Station, TX).

Descriptive statistics
A total of 794 older adults aged 60 years and above participated in the study. The mean age of the participants was 69.9 years; more than half (55.4%) were in their sixties ( Table 1). The male to female ratio was close to unity (50.4% and 49.6%, respectively). A greater proportion of the participants were Hindu (78.7%) and illiterate (80.1%). Nearly 38% of the participants were of indigenous origin, and 34% were from the Madhesi and other ethnic groups. About half of the participants (53.8%) were married at the time of the survey. Regarding occupational status, 54.2% of the participants were not involved in any income-generating activities. As such, around half of the participants had a family income of 5000 NRs or less. More than three-quarters of the participants had no physical activity at all (77.1%) and tobacco consumption history (76.8%), while only a quarter (25.1%) had alcohol drinking habits (Table 1). Depression (55.8%), osteoarthritis (41.7%), and chronic obstructive pulmonary disease (COPD) (15.4%) were the most prevalent health conditions among the participants ( Table 2).  The proportion of participants reporting a poor QOL was significantly higher among those aged ≥ 80 years (p-value = 0.009), female (p-value < 0.001), living in Sunsari district (p-value < 0.001), following Hinduism (p-value < 0.001), from Brahmin/Chettri/Thakur ethnicity (p-value < 0.001), unmarried (p-value < 0.001), illiterate (p-value < 0.001), unemployed (p-value < 0.001), earned ≤ 5000 Nrs./month (p-value = 0.000) and had no physical activity at all (p-value = 0.003) ( Table 1).

Chronic condition and QOL
The Chi-square test for the relationship between chronic conditions and QOL showed that participants with prevalent osteoarthritis (p-value < 0.001), COPD (p-value < 0.001), and depression (p-value < 0.001) had significantly poor QOL ( Table 2).

Summary of QOL indices
The numeric indices of QOL scores by different domains are presented in

Discussion
This study aimed to assess the QOL and its correlates among older adults in Eastern Nepal and found that seven in ten participants had poor QOL, which was significantly associated with age, socioeconomic status, religion, ethnicity, physical activity, osteoarthritis, and depression.
The overall poor QOL observed in this study is consistent with previous studies from Nepal's capital city of Kathmandu, where older patients in an outpatient clinic 14 and nursing homes 15 settings, had a lower overall QOL score. Previous studies, from international settings, are in line with our findings 21,22 . Further, a gradient decline in the odds of poor QOL was noted by increasing age group which is in line with a previous study from Nepal where age was inversely associated with QOL 23 . The declining QOL with age is plausible, given that older adults are at increased risk of chronic diseases and infection 6,24 . Furthermore, age is associated with a progressive decline in muscle mass, strength, power, and physical performance 8,9 . As a result, they have reduced mobility and functional capacity which ultimately influences the overall wellbeing and lowers the QOL at later life 8,9 .
A significant finding of this study is the role of socioeconomic status and its implications for QoL.
Better socioeconomic status, as indicated by literacy and higher income in this study, was associated with higher QOL among older adults in our study as well as others 25−27 . Socioeconomic status is considered as one of the driving forces for the existing health disparities globally 28 . Given the wellestablished relationship between socio-economic status and well-being, in terms of perceived health 29 , mortality, and morbidity 30,31 , the observed association with QOL was anticipated. Education increases health literacy and influences one's ability to make informed decisions about their health and healthy behaviours 32 . Likewise, income increases purchasing capacity, access to health care, and affordability of everyday need 33 . Together, education and income may determine one's social status and the psychosocial advantages gained through social networks 33  Previously, among older adults, several mediators such as better physical and mental health status increased exercise self-efficacy, increased physical self-worth, and reduced disability limitations, has been identified in the pathways between increased physical activity and QOL 35,36 .
In the context of Nepal, an individual's ethnicity has similar effects as their socio-economic status.
Hence, it is not surprising to find that ethnicity was associated with QOL. Physical and mental ailments were associated with lower QOL. Absence of osteoarthritis and depression was associated with higher odds of better QOL. Previously, low perceived QOL among patients with osteoarthritis is reported 44,45 . The pain and limitations of daily living activities resulting from osteoarthritis may explain the observed reduced QOL 45,46 .
Our findings of an inverse association between QOL and depression are consistent with previous studies from Nepal and globally 22,47,48 . A meta-analysis of 24 studies reported moderate improvements in QOL following treatments for depression 49 . Depression may lower the QOL by impairing physical and social functioning, and overall health 50 .

Strengths And Limitations
As with most studies, this study has some limitations. The participants were from eight rural municipalities of Morang and Sunsari district, Nepal; thus, the results can be only generalized to the studied setting. Secondly, social-desirability bias may have occurred as our findings relied on self-reported data. Further, the study adopted a cross-sectional design, which precludes any inferences of the cause-effect relationships. The most important strength of this study includes a large sample size with more than 90% response rate, strong methodology, and adoption of Older People's Quality of Life (OPQOL) questionnaire for the first time in Nepalese settings.

Conclusion
This study has provided statistical evidence of the factors influencing the good QOL among the elderly in Nepal. The current study shows that one in three respondents have a good QOL. In delineating the mechanism by which this happens, the results demonstrate that biosocial factors such as increasing age and ethnicity outside the upper caste decrease the likelihood of a good QOL. Similarly, disease conditions such as depression and COPD were found to be associated with a lower likelihood of a good QOL. However, compositional factors such as literacy and high-income levels were found to increase the likelihood of a good QOL. These results provide compelling evidence to develop and implement policies aimed at improving the conditions that catalyse poor QOL in this population.