Social Determinants of Health and Happiness of Older Adults in Ghana: Secondary Analysis of Ghana SAGE Wave 2 Longitudinal Data

Background Social determinants of health [SDOH] and happiness have received meaningful consideration as foundational concepts in the field of public health. However, the relationship between the SDOH and happiness of older adults have not received the requisite recognition in Ghana. This study examined the relationship between the SDOH and happiness of older adults in Ghana. Methods The study used data from the 2014/2015 Ghana Study on Global Ageing and Adult Health (SAGE) Wave 2. Data was analysed using the Structural Equation Modeling (SEM) technique to investigate the direct, indirect and covariances of the SDOH and happiness of older adults. Results The results showed positive relationship between the SDOH and happiness among older adults. The economic stability (β = 0.07), neighbourhood and built environment (β = 0.02, P < 0.001), access to quality education (β = 0.56, P < 0.01), access to healthy food (β = 0.48, P < 0.001) social and community context (β = 0.41, P < 0.05), and access to quality healthcare (β = 0.80, P < 0.001) had direct relationship with happiness of the older adults in Ghana. Conclusion This study shows that the conditions in which older adults were born, live, learn, work, play, worship, and age (SDOH) positively impact their happiness in later life. Neighbourhood and physical environment influence the effect of quality education on happiness of older adults. Social policies and interventions aiming at happiness of older adults should consider the social determinants of health and the mediating effects of food on happiness through quality education, and quality of healthcare system.


Introduction
Social determinants of health (SDOH) and happiness of older adults is a pivotal area in healthy ageing.
The past decades have received noteworthy consideration of the concept of happiness in the eld of public health [1]. However, the relationship between the SDOH and happiness concerning older adults is yet to be accorded adequate attention. Therefore, an evidential understanding of the correlation between these two vital concepts as far as older adults are concerned is appropriate and timely. This study drove on the premise that environments in which older adults are born, live, learn, work, play, worship, and age affect a wide range of health, functioning, and quality-of-life outcomes as well as happiness of older Imperatively, close relationships and interactions with family, friends, colleagues at work, and community members have key impact on people's happiness. Access to safer open social spaces for recreational purposes like the community centers, aged friendly drinking spots, church places, and places where older adults can gather to play games are very important characteristics of the social and community context that are directly related to happiness [31]. These enhance interpersonal relationships among older adults [31] and the sense of community which signi cantly in uence happiness [32]. Older adults who have positive relationships at home, at work, and in the community have less to worry about in life. Further, spending time with friends and good neighbours, community participation, and trust in people in the neighbourhood including the local police have signi cant relationship with happiness of older adults [33]. Also, religiosity and attending religious activities (social integration and engagement) have positive relationship with happiness [34,35]. An indication that the community serves as social environment that has profound effects on the kind of social support one could get [36,37]. Aged people are happier when they are socially integrated and engaged into the community where they live [38].
Studies in Ghana have shown that family social capital, including family sense of belonging, autonomy support, control, and social support, vary with self-reported happiness [39]. Happiness among Ghanaians is in uenced by several factors such as economic, cultural, social capital and health factors [40]. A wide range of scholarly work have contributed to the debate on SDOH and happiness. Yet these studies do not primarily target older adults. Also, the categorization of the SDOH by the Healthy People [2] as applied in this study does not include access to healthy food or diet which has been shown to have in uence on happiness. Lastly, none of the available studies have investigated the direct, indirect and covariance effect of the indicators of SDOH on happiness of older adults in Ghana, as far as our search revealed.
Premised on the foregoing, this study examined the relationship between the SDOH and happiness of older adults in Ghana. This is because the conditions in which they are born, live, learn, work, play, worship, and age could affect their happiness as previous studies have shown from different contexts.

Theoretical perspective
The Objective List Theory (OLT) strengthened the premise of this study. The theory holds that happiness consists of a human life that achieves certain things from a list of worthwhile pursuits which may include career accomplishments, friendship, freedom from disease and pain, material comforts, civic spirit, beauty, education, love, knowledge, and good conscience [41,42]. The OLT best explains relationship between happiness and the SDOH as the main aim of this study. Aspects that have applied the OLT include the argument that people's objectivist judgments would recognise knowledge as an element of happiness because it involves appropriately justi ed beliefs about meaningful truths [43]. Parker [44] asserts that the most worthwhile lives are those high in various objective goods which principally include happiness and meaning. Although these studies somewhat justify the tenets of the OLT, they apply the theory either making detail explanations of the theory itself or scarcely apply it to older adults in terms of their happiness. Therefore, the relationships between happiness and the SDOH concerning older adults is left of out. The study theorised that happiness of older adults is in uenced by the SDoH as categorised by Healthy People [2]. In addition to the ve indicators by Healthy People, access to healthy food was included and tested in this study. Figure 1 presents details of the theoretical framework.

Data source
The data used for this study was obtained from the 2014 Ghana Study on Global Ageing and Adult Health (SAGE) Wave 2, which is supported by the US National Institute on Ageing, Division of Behavioural and Social Research and National Governments. The core SAGE collects a comprehensive data on adults aged 18+, with an emphasis on populations aged 50+, from nationally representative samples in six countries: China, Ghana, India, Mexico, Russian Federation and South Africa. The SAGE Wave 2 for Ghana focused on the health and wellbeing as well as happiness level of older adults (50+) adults. SAGE surveys follow the standard procedures (i.e., sampling, questionnaire development, data collection, cleaning, coding and analysis) which allow cross-country comparison. The survey employs a strati ed two-stage sampling technique. The initial stage involves the selection of Primary Sampling Units (PSUs) by region and location (urban/rural) across the six countries. The second stage involves the systematic selection of enumeration areas (EAs), households (HH) with 50 + adults, and households (HH) with persons aged 18-49. For this study, only those 50 years and over (n = 781) who had complete information on the variables of interest were included [13]. Respondents were classi ed into four categories of functional age brackets: the "younger old" (50-64 years) "young old" (65-74 years); the "old-old" (75-84 years); and the "oldest old" (85 years and above) [45]. The data was deemed suitable for this study because it is nationally representative.

Dependent variable
Happiness In general, the dependent variable assessed if respondents lived happy life or were satis ed with life. This was measured by three variables including satisfaction with living condition (1 = very satis ed to 5 = very dissatis ed), overall quality of life (from 1 = very good to 5 = very bad), and level of happiness (from 1 = very happy to 5 = very unhappy). For the analysis of this study, "happy in life" was recoded into a binary variable, "very happy", "happy", and "moderately happy" were coded into 1 representing older adults who rated themselves as happy (1 = happy) while "unhappy" and "very unhappy" coded into 0 indication older adults who rated themselves as unhappy (0 = unhappy).

Independent variables
Social determinants of health: The independent variables as adopted from the Healthy People [2] were divided into ve categories including: (a) economic stability, (b) access to quality education, (c) access to quality healthcare, (d) neighbourhood and built environment, as well as (e) social and community context. Some of the variables measuring these indicators were renamed/recoded to re ect items in the SDOH model and to suit the analysis.
Economic stability denotes living out of poverty to afford basic things like healthy food, healthcare, housing, quality education. It was de ned in this study as having steady employment and earning enough to pay for food, housing, health care, and education can reduce poverty and improve health and well-being. It was measured in this study by earnings/income (yes/no), currently working, renamed employment (yes/no).
Access to quality education is where learners who are well-nourished, are ready to participate and learn, are supported by their families and communities; healthy, safe, and supportive environments. For the analysis, it was de ned as high-quality educational achievement and or opportunities for older adults in their entire life environment. it was measured by highest level of education (from 1 = Primary/JHS to 3 = Tertiary), and years educated (from 1 = < 10 years to 4 = > 30 years).
Access to quality healthcare is an indication that an individual gets the healthcare services needed where they live. In this study, it was considered as older adults' access to timely and high-quality healthcare services in the places where they have spent their entire lives. It was measured to include healthcare provider (from 1 = medical doctor to 7 = home health care worker), quality of healthcare (from 1 = very good to 5 = very bad), and satisfaction with quality of care (from 1 = very satis ed to 5 = very dissatis ed).
Neighbourhood and built environment (NBE) are the areas that promote the health and wellbeing of and individual. For this study, NBE denote improved health and safety in the place where older adults are born, grow, live, learn, work, play, worship, and aged. it comprised place of residence (rural/urban), always lived here in the community, renamed childhood residence (yes/no), as well as safe on the street, renamed safety (from 1 = completely safe to 5 = not safe).
Social and community context (SCC) refer to the relationships and interactions with family, friends, working colleagues and community members. In this study it is de ned as the social support that older adults need in the places where they were born, live, learn, work, play, worship, and aged/grow up. It was measured by community meetings renamed social integration (from 1 = never to 5 = daily), religious services renamed community engagement (from 1 = never to 5 = daily) and feel left out which was renamed social isolation (from 1 = never to 4 = often).
Access to healthy food included number of fruit servings per a day, and number of servings of vegetable per a day, renamed fruit and vegetable consumption (from 1 = < 2 servings to 4 = > 6 servings), hungry, no money to buy food (from 1 = every month to 5 = never).

Data analysis
Statistical analyses were performed with SPSS version 26 and Amos version 23. Data analysis was conducted by the used of the cross-tabulation, Con rmatory Factor Analysis (CFA) and the Structural Equation Modelling (SEM) techniques. First, the chi-square test was used to examine if the percentage distribution of the older adults who were happy in life were statistically signi cantly (p < 0.05) different by the selected background characteristics. SEM uses factor analysis to create an indicator score to measure a latent variable. Therefore, SEM was applied because the indicators used to measure the SDOH, and happiness were latent. The model was setup using a theory-based conceptual model (see Fig. 1) used for the study. At the CFA level, the model t for the latent variables was assessed, and were grouped into ve: economic stability, education access and quality, healthcare access and quality, neighbourhood and built environment, and social and community context according to the theory-based conceptual model for the analysis in this study. The CFA was therefore used to determine the measured variables that shared common variance and de ned a theoretically sound construct or latent variable. Measures that loaded unto one factor and effectively explained the variance were retained (See Fig. 2).
Also, the CFA was used to determine the statistically signi cant levels of the variables. The models were tted using the maximum likelihood estimation approach. At the nal stage of the model, modi cation indices were computed to improve the t of each latent variable. The goodness-of-t of the model was evaluated using the chi-square (χ²) test, the Non-normed Fit Index (NFI), Tucker-Lewis Index (TLI), Comparative Fit Index (CFI), and the Root Mean Square Error of Approximation (RMSEA). All the three relative t indices (NFI, TLI, and CFI) exceeded the 0,90 criterion [46]. The value of RMSEA was also lower than 0.08 [46]. All the variables measuring the latent variables were signi cant at 1% level of signi cance (P < 0.01) in the modi ed hypothesised model.

Results
Approximately 56 percent of older adults were females with 63.3 percent being relatively youngest (50-64 years). Sixty percent were married, 50.6 percent had attained Secondary/SHS education, 49.9 and 56.3 percent had 2 to 3 servings of fruits and vegetables respectively. Majority of older adults (76.8%) had lived in the same community/neighbourhood from childhood. Seventy one percent were working, and 68 percent were earning income. Almost all older adults (91.5%) were never hungry. Approximately 73 percent had walkable environment while 97.4 percent had access to modern healthcare. Sixty-ve percent were satis ed with quality of healthcare services, while 45.6 percent had good healthcare. Approximately 47 percent had never attended public meetings, 82.5 percent were never lonely, and 40.8 percent felt completely safe. Roughly 65 percent were satis ed with their living conditions, 52.8 percent had good overall quality of life whilst 72.5 percent were happy.  Table 2 shows the chi-square results of happiness in life and the background characteristics of the older adults. Overall, majority of the research participants (94.1%) were happy in life. Most of older adults who reported being happy were females (94.5%). Concerning age, old old (95.7%) were happier than younger old (93.5%). All older adults who were never married and separated/divorced/cohabiting were rather happier than those who were married. Older adults who were employed were happier (96.1%). Also, those who resided in urban area were much happier than those in rural setting (96.3% and 92.2%) respectively. Older adults who felt safer in the community reported being happy in life. Older adults who had higher education were much happier. Those with 2-3 and more than 6 servings of fruits and vegetables respectively, were happier. Older adults who were satis ed with their health and living conditions had better social integration, social engagement, and also had very good overall quality of life.

The Social Determinants and Happiness among Older Adults
As shown in Fig. 2, most of the indicators of social determinants of health loaded well on their latent variables. Economic stability was caused by income (Factor Score = 0.35, P < 0.01), and employment (Factor Score = 0.99, P < 0.001). Walkability (Factor Score = 0.73, P < 0.001), safety (Factor Score = 0.49, P < 0.001), and place of residence (Factor Score = 0.81, P < 0.001) were signi cantly related to neighbourhood and built environment. Access to quality education was highly measured by higher educational attainment (Factor Score = 0.97, P < 0.001), and years educated (Factor Score = 0.81, P < 0.001). Being hungry affected access to healthy food negatively (Factor Score=-0.49, P < 0.001). However, fruit consumption (Factor Score = 0.73, P < 0.001) and vegetable consumption (Factor Score = 0.70, P < 0.001) signi cantly caused access to healthy food. Community engagement (Factor Score = 0.48, P < 0.05), and social integration (Factor Score = 0.65, P < 0.001) positively affected social and community context, but stress negatively caused social and community context (Factor Score=-0.53, P < 0.001).
Satisfaction with health care (Factor Score = 0.20, P < 0.001), availability of health care provider (Factor Score = 0.28, P < 0.001) and healthcare coverage (Factor Score = 0.39, P < 0.001) were positively related to health care system. Happiness of older adults was strongly caused by level of happiness (Factor Score = 0.78, P < 0.001) and satisfaction with living condition (Factor Score = 0.88, P < 0.001).
The social determinants of health as were analysed in this study were directly and positively related to happiness of older adults. The results show that economic stability (β = 0.07), neighbourhood and built environment (β = 0.02, P < 0.001), access to quality education (β = 0.56, P < 0.01), access to healthy food (β = 0.48, P < 0.001) social and community context (β = 0.41, P < 0.05), and access to quality healthcare (β = 0.80, P < 0.001) had direct relationship with happiness of older adults.
Further, the ndings show that economic stability, social and community context covary (β = 0.45, P < 0.001) to affect happiness of older adults. Also, the results showed an indirect effect of healthcare system on happiness of older adults through access to healthy food (β = 0.54, P < 0.001). Also, healthy food mediated the effect of quality education on happiness of older adults (β = 0.84, P < 0.001) while access to quality education operated through neighbourhood and built environment (β = 0.92, P < 0.001) to in uence happiness of older adults.

Discussion
Enquiry into how SDOH (e.g. economic stability, access to quality education, health, neigbourhood, built environment as well as the social and community context of individuals) affect happiness of older adults has not received the requisite attention in Ghana. This study examined the relationships between the SDOH and happiness of older persons in Ghana. Direct and positive relationships are shown between the SDOH and happiness of older people. Economic stability (income and employment) of older persons was directly related to happiness, indicating the possibility of higher happiness of older persons, especially the youngest-old who are still capable of working and earning to afford their basic needs such as paying utility bills, and leisure activities. Our analysis supports results of Frey and Stutzer [9], Diener and Biswas-Diener [47], Blanch ower and Oswald [11] who found positive correlation between income and happiness.
In some instance, income has been identi ed to increase happiness [12]. Similarly, evidence shows a positive relationship between employment and happiness [13]. Indicating that economic stability has positive and signi cant relationship with the level of happiness of older persons.
We found that neighbourhood and built environment had direct and a positive effect on happiness of older adults. Although the socio-gerontological study by Lykken [48] argues that the ability to be happy in old age is determined by intrinsic but not external factors, ndings of our study possibly suggests that the external factors such as the SDOH affect happiness of older adults. Thus, the older adults who lived in better place of residence, either in the rural or urban areas and have walkable space in the neighbourhood could experience improved social relationships with neighbours, feel safer to move around freely in their communities and for leisure. Corroboratively, the experience of living in a poor neighbourhood has negative effect on happiness [25]. However, good relations with friends in the neighbourhoods positively enhance the happiness of older adults [23]. This is because the neighbourhood community represents the broader social environment that in uence the level of happiness of the individual [36,37]. The environmental factors that are directly connected to happiness include access to open, natural and green spaces, which are design features that allow for social interactions [24]. Therefore, it is not just the genetics that determine happiness of older adults but also, the neighbourhood and build environment play a direct and signi cant role in the level of happiness among older persons.
One of the important factors that in uence happiness is education [49]. Our results showed direct relationship between access to quality education and happiness of older adults, possibly indicating that older adults with higher educational attainment have higher self-con dence as well as the opportunity of getting a good job with higher income and savings which can positively make them feel happier than their colleagues who missed the educational and job opportunities. Consistent evidence shows that education has positive relationship with happiness [49][50][51]. Also, older persons who are highly educated, as compared to their counterparts with lower educational attainment, have higher self-con dence, decent jobs, and higher income which impact their level of happiness [27][28][29][30]. Brighouse and Swift [26] have shown that educational attainment increases level of happiness through independence and freedom of choice.
We found positive relationship between eating healthy food and happiness. This possibly explains the fact that older adults who have access to nutritious food with adequate (i.e., 3 to 5) servings of fruits and vegetables are happier than those who do not. The result of this study further con rms results of Blanch ower et al. [11] and Fararouei et al. [52] study that showed a linear relationship between happiness and the number of servings of fruits and vegetables consumed per day. Evidence shows that food that is enriched with fruits and vegetables is good for healthy living and therefore contribute to a feeling of happiness. Reasonably, low nutritional wellness has many health implications for older adults which also impact their happiness [53]. This suggestive that people who do not have access to healthy food rich in fruits and vegetables may have compromised health and happiness. Consistently, results from earlier studies show that healthy eating, speci cally food rich in fruit and vegetables reduces the chances of contracting food related diseases and enhances happiness of individuals [54][55][56]. Similarly, evidence show a positive relationship between food and happiness [57], indicating that eating a healthy food adds to one's happiness. Thus, there is a positive relationship between healthy eating and happiness indicating that nutritional behaviour in uences happiness and for that matter, older adults [58].
Consistent with earlier studies [36,37], our study revealed that social and community context had a direct and positive effect on happiness of older adults. Our observation is further supported by van der Have et al [31] who reported that the social and community context characteristics that are directly related to happiness include access to open social spaces for meetings and these include community centers, drinking spots for older adults, church places and game spots, which allow for social interaction among older adults. Plausible explanation could be that participation in social activities at the community level improves the sense of belonging, social contacts as well as social interaction and good relations in the neighbourhood and contribute to happiness of older adults. This corroborates with results of Davidson and Cotter [32] which showed that sense of community correlates signi cantly with happiness. Similarly, good relations in neighbourhoods have been identi ed to positively affect happiness of the individual [23]. Also, Helliwell and Putnam [33] showed that happiness is signi cantly related to spending time with friends and neighbours, civic participation, and trust in neighbourhoods and the local police. Meanwhile, older people are happier when they are socially engaged in the community where they live [38].
Besides, we realized that access to quality healthcare is positively related to happiness of older adults.
Happiness is an important factor that contributes signi cantly to the e ciency of the healthcare system [15]. A plausible reason for this could be that access to quality healthcare and satisfaction with health care provision as well as the availability of healthcare provider enhance happiness of older adults.
Further, Venkatapuram [18] also revealed a strong relationship between satisfaction with health and happiness. Therefore, the older adults have access to quality healthcare, the higher their happiness.
Venkatapuram [18] and Howell [19] concluded that happiness can be bene cial to health and longevity.
Also, good healthcare system has been recognised as a prerequisite to achieve happiness [16,17].
The study also revealed that food mediates the effects of healthcare access and quality on happiness.
This possibly explains that access to quality healthcare in a country helps to educate older people on health bene ts of eating nutritious food rich in fruits and vegetables which could lead to living a happy life. Also, ndings of this study show that food operate through access to quality education to impact happiness. A plausible explanation could be that higher educational attainment ensures having a good job, higher income and better understanding of nutrition with the means to afford nutritious food and therefore, improved happy life of older adults. Lastly, neighbourhood and build environment mediated the effect of quality education on happiness. This could be because higher educational attainment is in uenced by social integration and engagement. Our observation is further supported the nding of a study by Ariana [59] which showed a relationship between educational attainment and social integration.
However, Ariana's nding could not establish the mediating effect between quality education and happiness of older adults through neighbourhood and build environment.

Strengths and weaknesses
This study draws its conclusions from a nationally representative sample of aged people in Ghana. It also utilised multistage cluster sampling method in the selection of the respondents, whilst rigorous statistical analysis was conducted. The questionnaire and methods of data collection have also been validated. Moreover, the study included one more component (healthy food) to the indicators of SDOH by Healthy People and uncovers the direct and indirect effects of the social determinants of health on happiness of aged people that are important for policies and social interventions for older adults using SEM techniques. However, due to the cross-sectional nature of the survey, the study could not establish causality. Further, only variables with complete cases for this analysis were used in this study, which has the potential of producing some biased estimates. Results are generalisable to only older adults in Ghana. Lastly, the model should be tested for applicability in other countries.

Conclusions
In examining the social determinants of happiness of older adults, this study shows that the conditions in which older adults were born, live, learn, work, play, worship, and age (SDOH) positively impact their happiness in later life. Speci cally, economic stability and social and community context have signi cant linkages or intercorrelate to in uence happiness of older adults. Most importantly, effects of quality education and healthcare on happiness of older adults was mediated by access to healthy food. Indicating that eating healthy food plays a vital role for older adults who had access to quality education and healthcare. Lastly, neighbourhood and physical environment play important role between quality education and happiness of older adults in Ghana. Whatever, social policies and interventions targeting happiness among older adults should consider the social determinants of health, most importantly, the mediating effects of food on happiness through access quality education, and healthcare system, as well as the important role that neighbourhood and build environment could play between quality education and happiness of older adults. Declarations Not applicable

Abbreviations
Availability of data and material The dataset can be accessed by request at https://apps.who.int/healthinfo/systems/surveydata/index.php.

Ethics approval and consent for participation
The Wave 2 of the WHO-Ghana SAGE was approved by the Ethics Review Committee, World Health Organisation, Geneva, Switzerland, and the Ghana Health Service. A written informed consent was given to individual respondents. The authors of this manuscript were not directly involved in the data collection processes but rather obtained access by requesting for the data. The dataset can be accessed at https://apps.who.int/healthinfo/systems/surveydata/ index.php/catalog/sage.

Consent for publication
Not applicable. Figure 1 Theoretical model of Social Determinants of Health and Happiness See image above for gure legend.