How Social Participation Benefits the Chronically Ill: Self-Management as a Mediating Pathway

Abstract

Objective: Few studies have tested the purported causal mechanisms between social participation and health, especially among those in poor health. This study aimed to determine whether self-management of chronic disease operates as a pathway through which social participation affects health-related quality of life. Method: I utilized causal mediation analysis among 600 low-income older Singaporeans living with chronic conditions, to test whether self-management mediates the association between social participation and health-related quality of life. Results: Results show that self-management fully mediated the positive effect of informal social participation on health-related quality of life. Formal social participation was found to have a negative direct effect on health-related quality of life. Discussion: These findings reiterate the primacy of family and friends for older adults, but highlight that a better understanding of formal engagement with the low-income population is still needed.

Keywords

chronic conditions Singapore mediation health-related quality of life social participation

Introduction

Social participation is a key modifiable determinant of one’s health and well-being, and a vital component of successful aging (Adams, Leibbrandt, & Moon, 2011; Douglas, Georgiou, & Westbrook, 2016). Thus far, researchers have focused on establishing the causes and consequences of two main types of social participation for older adults—formal and informal. Although formal social participation usually entails involvement in formal groups and organizations in the community, informal social participation refers to the amount of contact and time that one spends with family and friends (van Ingen, 2008). Determinants of social participation such as marital status, income, and/or physical limitations (e.g., Donnelly & Hinterlong, 2010; Pinto & Neri, 2017) tend to differ between formal and informal social participation (van Groenou & Deeg, 2010), but both types of social participation seem to have an overall positive association with health outcomes. Many previous studies have found formal/informal social participation to be protective against negative health outcomes such as mortality and morbidity (e.g., Agahi & Parker, 2008; Tomioka, Kurumatani, & Hosoi, 2017), providing substantial support for countries aiming to increase social participation as a strategy to keep older adults healthy and sustain them as productive members of society (Raymond et al., 2013).

Social participation has been well studied across a range of geographical contexts and health outcomes, but two important limitations in the extant literature remain. First, less is known about whether and how social participation may aid those already in poor health, even though individuals are likely to live a greater proportion of their lives with a chronic condition compared with past generations (Crimmins, 2015). Galenkamp and Deeg (2016) have suitably pointed out that

an increase in the prevalence of chronic diseases, together with the rising share of older people in the population, asks for a shift of focus from the “successful” older people with limited disease or disability to those older individuals who live with multiple chronic diseases. (p. 87)

Yet while attempts to establish the determinants of social participation for older adults living with chronic conditions are suitably made (Galenkamp & Deeg, 2016), they do not examine whether and how social participation may affect the well-being of this subpopulation through different pathways than it does among the general population. Second, even though researchers have suggested a wide range of mechanisms through which social participation may operate to influence health outcomes, existing studies seldom test these pathways empirically. Rather, the majority of research has sought to confirm the salubrious effects of social participation on health instead, neglecting to explicitly test and/or clarify the mechanisms through which this association occurs (Thoits, 2011). It, therefore, remains unclear how social participation promotes well-being, even though it is commonly accepted that it does.

One especially pertinent outcome for the chronically ill is health-related quality of life, because chronic conditions have long-term consequences and are often intrusive to everyday life (Schulman-Green et al., 2012). Health-related quality of life is a multidimensional concept that captures aspects of quality of life that are relevant to health, accommodating both subjective and objective perspectives across multiple domains of health (e.g., physical, psychological, etc.; Testa & Simonson, 1996). Health promotion efforts are thus often aimed at maximizing one’s quality of life, and its attendant measures have been used widely to compare and evaluate the effectiveness of health interventions for economic and medical decision making (Kind, Lafata, Matuszewski, & Raisch, 2009). In this context, research aimed at understanding pathways to enhance health-related quality of life can serve to inform future strategies to improve the well-being of older adults living with a chronic condition.

Social participation plays a substantial role in shaping older adults’ quality of life (Gabriel & Bowling, 2004). Participating in activities, maintaining social roles, and developing harmonious relationships help older adults perceive their lives more optimistically (Levasseur, St-Cyr Tribble, & Desrosiers, 2009), and provide a sense of belonging to the community (Aw et al., 2017). Researchers have typically found statistically significant associations between social participation and health-related quality of life, even though the strength of this association appears to vary across studies. While some studies have found the association to be relatively weak (e.g., Levasseur, Desrosiers, & Noreau, 2004; Takada, Sashika, Wakabayashi, & Hirayasu, 2016), others have established a strong association across various populations (e.g., Benka et al., 2016; He et al., 2016; Li, Lin, & Chen, 2011). This variation in effect size may possibly be addressed through a better understanding of its causal pathways, but these have remained largely opaque due to a lack of empirical work on mediating mechanisms.

Self-Management as Mediator

It is thus noteworthy that social participation may be instrumental not only in promoting quality of life by preventing the onset of chronic diseases but also in improving the self-management of these conditions for those already living with a chronic disease. Self-management has been defined as the “ability of the individual, in conjunction with family, community, and healthcare professionals, to manage symptoms, treatments, lifestyle changes, and psychosocial, cultural, and spiritual consequences of chronic diseases” (Richard & Shea, 2011, p. 257). Broadly, it is the way through which older adults manage their resources in daily life to maintain their well-being and remain independent for as long as possible, thus enabling the individual to “age successfully” (Steverink, Lindenberg, & Slaets, 2005). First, self-management training is a well-studied health intervention used by medical professionals to improve patients’ quality of life, with multiple studies supporting its efficacy (Bodenheimer, Lorig, Holman, & Grumbach, 2002; Cochran & Conn, 2008; Jonkman, Schuurmans, Groenwold, Hoes, & Trappenburg, 2016). Systematic reviews of the literature suggest that the central benefit of self-management training is that it enhances participants’ problem-solving skills, which in turn improves their psychosocial and/or mental well-being (Barlow, Wright, Sheasby, Turner, & Hainsworth, 2002; Jonkman et al., 2016). Other salubrious effects include reduced pain and disability (Grady & Gough, 2014), resulting in a better overall health-related quality of life for the individual. In the bigger picture, self-management support constitutes one of six essential elements in the Chronic Care Model, which focuses on improving disease outcomes through system-level changes within the primary health care system (Grady & Gough, 2014). Studies have found that providing better self-management support to patients tends to reduce emergency department visits and length of hospital stays (Bodenheimer, Wagner, & Grumbach, 2002), both of which may reflect improved health-related quality of life (Dominick, Ahern, Gold, & Heller, 2002).

There are several reasons why self-management may function as a major pathway through which social participation improves overall well-being. First, self-management does not occur solely at the level of the individual—its success is often dependent on influence from family and friends (Gallant, 2003). Social participation has been thought to facilitate the exchange of health information and promote better health behavior through social influence and shared norms (i.e., informal social control; Berkman, Glass, Brissette, & Seeman, 2000). In the case of chronic illness, engaging with others in the context of social activity proffers opportunities for friends and family to influence the individual (through reminders, persuasion, or encouragement etc.) into adopting healthy behavior and better management of their chronic condition(s). Furthermore, carrying out activities successfully together with others may also foster a greater sense of control and/or self-efficacy over one’s own life (Thoits, 2011), in turn promoting more active management of one’s condition through exercise and diet (Clark & Dodge, 1999), as well as improved mental health (Mikula et al., 2017). Yet while social participation and self-management have both been independently accepted as important predictors of the health and well-being of older adults, no study to my knowledge has sought to establish self-management as a possible mediator of social participation. Using data from low-income older adults living in Singapore, this study thus builds upon the extant literature on social participation and health and seeks to investigate whether (and to what extent) self-management of chronic conditions mediates the relationship between social participation and health-related quality of life.

The Singapore Context

Singapore is a multi-ethnic city-state in Southeast Asia, with a population of approximately 5.6 million and land area of 719 km². Yet despite economic prosperity and having built up a strong public infrastructure over the past few decades, Singapore has become one of the most rapidly aging countries in the world, due to its high life expectancy and subreplacement fertility rate (Singapore National Population and Talent Division, 2015). Projected increases in the prevalence of chronic diseases such as diabetes (Phan et al., 2014) have recently become a source of anxiety for the state, due to potential spikes in health care expenditure and losses in productivity. The Singapore government has recently responded by investing in strategies aimed at helping older adults age actively in the community, in the hope that such engagement will prevent and/or delay the onset of chronic illness. However, a history of high reliance on family members for care (Mehta, 2006) and social life means that social participation for older adults continues to be predominantly structured around family interactions (i.e., informal social participation; Aw et al., 2017), rather than participation in formally organized events or activities (i.e., formal social participation). Furthermore, the majority of older adults in Singapore are not English-literate and are likely to face issues managing their illnesses even in basic tasks such as taking prescribed medication correctly (R. Malhotra et al., 2019). As a result, self-management is likely to be a key factor in determining health-related quality of life among older adults. The current study focuses specifically on those from a low-income background, and thus who are more likely to live alone (Gubhaju, Østbye, & Chan, 2018). This subpopulation of older adults is also likely to reap a greater benefit from social participation compared with those with a higher socioeconomic status (Niedzwiedz et al., 2016).

Design and Method

Study Design and Participants

For this study, data from a 2016 survey conducted among low-income, community-dwelling older adults (aged 60 and above) in Singapore were used. Respondents were occupants of public rental flats in one Housing and Development Board (HDB) Estate located near the central region of Singapore, which comprises highly subsidized public housing apartments available exclusively to low-income families earning less than 1,500 Singapore dollars (SGD) a month. Out of 1,908 unique household addresses identified and eligible for the study, 528 were uncontactable. Of the remaining, 928 (67.2%) eligible participants agreed to participate in face-to-face interviews with a structured questionnaire. Respondents free from chronic conditions (including angina/myocardial infarction, any form of heart disease, cancer, cerebrovascular disease, hypertension, diabetes, respiratory illness, digestive illness, ailments of the liver or gall bladder, chronic back pain, osteoporosis, cataract and glaucoma) and/or those with any missing information on the variables of interest were excluded, resulting in a final analytic sample of 600 older adults. In line with past research, the following hypotheses were tested:

Hypothesis 1: Both formal and informal social participation are associated with health-related quality of life.

Hypothesis 2: The association between formal and informal social participation and health-related quality of life is mediated by self-management.

Outcome Variable

The EQ-5D-5L was used to measure health-related quality of life (The EuroQol Group, 1990). Having been validated in many countries including Singapore, the EQ-5D-5L consists of two components: (a) the descriptive system assessing five domains (mobility, self-care, usual activities, pain/discomfort, and anxiety/depression) from which an index score can be calculated; and (b) the EQ visual analog scale (EQ VAS). For this analysis, index scores from the EQ-5D health states were computed using Singapore preference weights (Luo, Wang, Thumboo, Lim, & Vrijhoef, 2014).

The range of possible Singapore utility-based EQ-5D index scores is −0.769 to 1.00, where a higher score represents a better state of health (see Abdin, Subramaniam, Vaingankar, Luo, & Chong, 2015). Possible scores for the EQ VAS range from values of 0 to 100, where 0 represents the worst health you can imagine and 100 being the best health you can imagine.

Mediator Variable

The Partners in Health (PIH) scale—a generic self-rated measure of chronic disease self-management—was used to measure self-management. This 12-item scale was recently validated in the Asian context (Chiu, Tam, Siu, Chau, & Battersby, 2017), and is made up of three subscales measuring an individual’s knowledge, coping, and symptom management/adherence in relation to his or her health condition. Examples of items from the scale include “I am able to deal with health professionals to get the services I need” and “I manage the effect of my health condition(s) on my social life.” Each item on the scale was scored from 0 to 8, and possible scores for the PIH scale, therefore, range from 0 to 96. I standardize this measure before including it into the mediation model. The inter-item reliability for this scale in the sample was satisfactory (α = .84).

Predictor (or “Treatment”) Variable

As highlighted earlier, social participation is commonly categorized as formal or informal (Levasseur, Richard, Gauvin, & Raymond, 2010). For formal social participation, respondents were asked how often (in the past 6 months) they had (a) attended a community event organized by grassroots associations (e.g., Resident’s Committee, Community Development Council etc.), (b) took part in activities at Senior Activity Centers, or (c) attended a place of worship (e.g., church, mosque etc.). Informal social participation was assessed by asking respondents how often (in the past 6 months) they had gone out with family members or friends. This variable was dichotomized using the cut-off point of at least once a week in line with previous research establishing a link between social participation and health among Singaporeans and elsewhere (see Ang, 2016; Ng, Broekman, Niti, Gwee, & Kua, 2009; Sørensen, Waldorff, & Waldemar, 2008). The causal mediation framework (explained in further detail below) emulates the experimental setting by splitting the observations into two groups (i.e., treatment and control), and then weighting for self-selection into the groups based on observed covariates. In this case, those with formal/informal social participation levels of at least once a week (coded as 1) were considered the treatment group, while the rest became part of the control group (coded as 0).

Covariates

Covariates used in this analysis include sociodemographic variables such as age, gender (male/female), ethnicity (Chinese/nonChinese), education (no formal education/primary/secondary and above), marital status (not married/married), number of living children, and work status (not working/working). Because the severity of one’s health condition may confound the relationship between social participation and quality of life, several measures of health are also included in the analysis. These are the number of chronic conditions diagnosed by a medical professional (1-2/3+), including the presence of any limitation (no limitation/at least one limitation) in activities of daily living (ADL). ADLs assessed included taking a bath, dressing up, eating, standing up and sitting down from a bed/chair, walking around the house, going out of the house, and using the toilet.

Analytical Strategy

To estimate the effects of social participation on health-related quality of life and how much of this relationship is mediated by self-management, the causal mediation strategy proposed by Imai, Keele, and Tingley (2010) was adopted and executed using the mediation package (Hicks & Tingley, 2011) in STATA 15. The causal mediation strategy employs the counterfactual (or quasi-experimental) framework to identify mediating variables between the predictor (or treatment) and outcome variables, and allow for a causal interpretation of the observed effects even with cross-sectional data. Briefly, the counterfactual framework states that true causal effects can only be estimated after observing realized outcomes from the same individuals under different possible conditions (see Höfler, 2005 for further explanation). Although these effects can never be empirically observed (because only one outcome can be observed from each individual), statistical models adopting this framework attempt to approximate true causal effects from observational data through procedures such as propensity score matching and inverse proportionality of treatment weighting. A total of four separate sequential models using ordinary least squares regression were estimated by type of social participation and health-related quality of life outcome. Furthermore, 5,000 simulations for each model were computed to approximate parameter uncertainty and present the 95% confidence interval (CI) derived from the resulting distribution (Hicks & Tingley, 2011).

Causal interpretation of the model is dependent on an identification assumption termed sequential ignorability, consisting of two parts. First, treatment assignment is expected to be unconfounded (i.e., ignorable). To achieve unconfoundedness with the current observational data, propensity scores of engaging in each type of social participation were estimated using the covariates, and inverse proportionality of treatment weights were constructed and included in the model estimation. This quasi-experimental method approximates random assignment into treatment and control groups by adjusting for the likelihood of an individual self-selecting into the treatment group. In the context of this study, this means that the effects of formal/informal social participation can be interpreted as “treatment effects” as in the experimental context, as if one group was randomly assigned to participate at least once a week while the other served as a control group. The second part of the sequential ignorability assumption builds on the first and further assumes that the observed mediator is also ignorable. To address this, I conduct sensitivity analyses recommended by Imai et al. (2010), also using the mediation package. These sensitivity analyses address the fact that unobserved pretreatment variables may confound the relationship of interest, based on the error terms of the mediator and the outcome. Results showed that for the average causal-mediated effect to equal zero, the two error terms for each model would need to have a correlation of at least .24 (details can be found in Imai, Keele, & Yamamoto, 2010).

Results

Table 1 describes the characteristics of the analytic sample. In all, 32% of the sample participated in formal social activities at least once a week, while 25% participated in informal social activities at least once a week. Respondents reported lower EQ-5D index mean scores than the general Singapore population in a comparable age group (Abdin et al., 2015). EQ VAS scores were also lower than previously observed among diabetes patients (Wang, Tan, Tay, Thumboo, & Luo, 2015). This was anticipated given that in addition to having at least one chronic condition, the sample also comprised low-income older adults. The mean score for the PIH scale was 76.3. The majority of respondents were male (58.3%), of Chinese descent (84.3%), not married (69.8%), not working (66.3%), and had no formal education (47.8%). In terms of health, most respondents had three or more chronic conditions (60.5%). Only a minority reported having at least one ADL (22.2%), but this figure is noticeably higher than those found in previous studies of the general Singapore older adult population (C. Malhotra, Chan, Malhotra, & Østbye, 2011).

Table 1.

Characteristics of the Sample (n = 600).

Variables	% or M (SD)
Formal social participation
Less than once a week	67.8
At least once a week	32.2
Informal social participation
Less than once a week	75.0
At least once a week	25.0
EQ VAS	69.5 (19.8)
EQ-5D index score	0.8 (0.3)
PIH scale score	76.3 (14.0)
Age	70.8 (7.4)
Gender
Male	58.3
Female	41.7
Ethnicity
Chinese	84.3
Non-Chinese	15.7
Education
No formal education	47.8
Primary	35.7
Secondary and above	16.5
Marital status
Nonmarried	69.8
Married	31.2
No. of living children	1.6 (1.9)
Work status
Not working	66.3
Working	33.7
Number of chronic conditions
1-2	39.5
3+	60.5
ADL limitations
No difficulty	77.8
At least one difficulty	22.2

Note. EQ VAS = EQ visual analog scale; EQ-5D = EuroQol-5D; PIH = Partners in Health; ADL = activities of daily living.

Bivariate correlations of the key variables were estimated and are displayed in Table 2. These pairwise correlations provide insight as to what kinds of relationships to expect, before controlling for covariates. Statistically significant correlations were found between the PIH scale score and health-related quality-of-life measures (r = .32 for the EQ-5D index score and r = .31 for the EQ VAS). Informal social participation was also found to be significantly correlated with the PIH scale score (r = .17) and the EQ VAS (r = .10). These correlations imply that the association between informal social participation and health-related quality of life may be mediated by self-management. No significant correlations between formal social participation and the other variables of interest were observed, suggesting that formal social participation may not be associated with self-management or health-related quality of life.

Table 2.

Pairwise Correlations Between the Variables of Interest.

	(1)	(2)	(3)	(4)	(5)
1. Formal social participation	1.000
2. Informal social participation	0.006	1.000
3. PIH scale score	0.077	0.173*	1.000
4. EQ-5D index score	−0.047	0.076	0.319*	1.000
5. EQ VAS	0.050	0.097*	0.309*	0.471*	1.000

Note. PIH = Partners in Health; EQ-5D = EuroQol-5D; EQ VAS = EQ visual analog scale.

*p < .05.

Results from the full model are shown in Table 3. After controlling for covariates, the mediation model for informal social participation showed a significant total effect on both the EQ-5D index score (β = 0.038) and the EQ VAS (β = 3.613). Approximately 67% (β = 0.026) and 45% (β = 1.667) of the total effects were mediated by the PIH scale score, respectively. The remaining direct effect of informal social participation on health-related quality of life was not statistically distinguishable from zero, suggesting a complete mediation by self-management. The direct and total effects of formal social participation on the EQ-5D index score were significant. Both effects were negative, indicating that net of self-management, formal social participation leads to a decrease in health-related quality of life.

Table 3.

Results From Models Estimating the Mediation of the Social Participation−Quality of Life Association by Self-Management.

	(1)EQ-5D index scoremodel-averaged coefficient[95% CI]	(2)EQ VASmodel-averaged coefficient[95% CI]
Formal social participation
Average causal-mediated effect	0.007[−0.003, 0.018]	0.573[−0.219, 1.488]
Direct effect	−0.043 [−0.077, −0.010]	1.164[−2.167, 4.450]
Total effect	−0.036 [−0.072, −0.001]	1.737[−1.683, 5.15]
Informal social participation
Average causal-mediated effect	0.026 [0.010, 0.047]	1.667 [0.699, 2.903]
Direct effect	0.011[−0.028, 0.050]	1.944[−1.611, 5.449]
Total effect	0.038 [0.0003, 0.074]	3.613 [0.062, 7.192]

Note. Models control for age, gender, ethnicity, education, marital status, work status, number of chronic conditions, and the presence of any ADL limitation. EQ-5D = EuroQol-5D; EQ VAS = EQ visual analog scale; CI = confidence interval; ADL = activities of daily living; Bold-faced values are p < 0.05.

To juxtapose the effects of social participation by type, path diagrams of the estimated models are presented in Figure 1, focusing on the EQ-5D index score as the outcome. The figure illustrates that self-management “suppresses” (see MacKinnon, Krull, & Lockwood, 2000) the direct effect of formal social participation on the EQ-5D index score. By contrast, the total effect of informal social participation was mediated by self-management. The size of these effects fall within the range of a “minimum clinically important difference (MCID)” for the EQ-5D (MCIDs are scores that indicate the smallest meaningful change for a patient) found in previous studies (Coretti, Ruggeri, & McNamee, 2014), indicating its substantive importance.

Figure 1.

Path diagram between social participation, self-management, and the EQ-5D index score.

Discussion and Implications

The purpose of this study was to determine whether self-management mediates the association between social participation and health-related quality of life, and to what extent. Utilizing the counterfactual framework for causal mediation analysis, evidence supporting the hypothesis that self-management acts as an important pathway and mediator through which social participation affects health-related quality of life was found. This relationship, however, differs by the type of social participation (i.e., formal or informal) considered.

Self-management mediated a large proportion of the total effect for informal social participation, such that the remaining direct effect on health-related quality of life was found to be statistically nonsignificant (i.e., complete mediation). First, this provides support to the theory that increased contact with friends and family promotes well-being primarily through social control mechanisms on health behavior (Thoits, 2011). For individuals with a chronic illness, this translates into better self-management of one’s condition. The findings build upon existing studies suggesting that family involvement in health promotion and intervention are likely to be beneficial for those living with one or more chronic conditions (Martire, Lustig, Schulz, Miller, & Helgeson, 2004). Such an approach may be especially efficacious in countries with a strong emphasis on the family (e.g., Singapore, Hong Kong), where older adults often wish to avoid being a burden to other family members (Cheng, Chan, & Chan, 2008) and may feel a greater propensity to take action toward this end (e.g., by better managing their health). Second, the findings provide insight into how higher rates of hospital readmission and utilization among the low-income population (Low et al., 2016) may possibly be addressed by the health system. Although support from family and friends has been given relatively little attention in the literature on the Chronic Care Model thus far (Davy et al., 2015), findings here suggest that health care providers utilizing the Chronic Care Model to improve care delivery should more carefully consider (and adapt to) the role of informal support from family and friends in promoting self-management and overall well-being (Grady & Gough, 2014; Rosland & Piette, 2010). Finding ways to encourage informal social participation among those with chronic illnesses can improve self-management, which may, in turn, reduce hospital readmission/utilization among the low-income population. Given the strong emphasis on family responsibility in the Singapore context (Mehta, 2006), it is highly likely that self-management is largely shaped through informal social participation, rather than through trainings in formal settings. This is fairly evident in the findings on formal social participation (discussed below).

For formal social participation, self-management appeared to function as an inconsistent mediator (also known as a suppressor; MacKinnon et al., 2000) with the EQ-5D index score (but not the EQ VAS) as an outcome. This suggests that for the formal social participation, its direct (and negative) effect on health-related quality of life appears to be masked by its indirect (and positive) effect through self-management. Only partial evidence for inconsistent mediation was found, however, as the latter (indirect) effect did not achieve statistical significance. The direct negative effect of formal social participation on health-related quality of life (although only on the EQ-5D index score but not the EQ VAS) contradicts the findings of other studies conducted in Asia, where most have found positive effects instead (e.g., He et al., 2016; Li et al., 2011). One possible explanation for this is that given strong ideals of meritocracy in Singapore, low-income individuals are often faced with a challenging institutional environment where they are met with bureaucratic complications and a general lack of empathy toward their plight (Suen & Thang, 2018). Even though social assistance for various needs is available, Suen and Thang (2018) noted that for individuals from low-income groups, “the experience of obtaining social support from formal sources was described [by potential recipients] as being almost tantamount to a violation of their dignity and sense of self-worth.” (p. 12) The general frustration and/or dissatisfaction toward formal organizations may extend to social activities that occur within formal contexts, negating any potential benefits that may be obtained from participating in them. Such an interpretation aligns with Abramson’s (2015) observation that the meaning and efficacy of social participation is often contingent on the organizational context through which it occurs. Therefore, even though prior studies in other countries (e.g., Brody et al., 2002; Wattana, Srisuphan, Pothiban, & Upchurch, 2007) have demonstrated that self-management training can help to improve health outcomes, it is unclear whether conducting this training in a formal setting will be effective within the Singapore context. More attention to how low-income groups engage in formal settings is certainly needed.

Main strengths of this study include the use of a relatively large sample (n = 600) of community-dwelling older adults with at least one chronic condition, as well as the use of a counterfactual framework that is able to support a causal interpretation of mediation mechanisms. In the past, similar studies have typically relied on a sample of a specific disease population and/or a much smaller set of observations (e.g., Benka et al., 2016; Li et al., 2011). A majority of these studies also fail to address reverse causation in their analyses, a notable shortcoming given that social participation and health share a reciprocal relationship (Griffith et al., 2017). This study has attempted to address issues of causality through a quasi-experimental design, but the use of longitudinal data in future research will better support the current findings. Also, while a key limitation of the current study is that data were not drawn from a random sample of all low-income older adults in Singapore, the housing estate under consideration is likely to represent a typical case of low-income elderly in Singapore. As the public rental scheme is administered by the HDB with its strict eligibility requirements, the sociodemographic makeup of public rental flat estates is likely to be similar across regions, especially because housing policy remains a fundamental way through which the Singapore government manages social integration (Sim, Yu, & Han, 2003). It is improbable that there will be substantial differences in accessibility to health care services by geographical area, given Singapore’s small size and efficient health care system. Furthermore, the social participation measures here are not exhaustive, and should not be treated as such. For instance, other forms of informal contact such as phone calls or home visits were not assessed. Those considered in this analyses, however, reflect an important set of social activities in the Singapore cultural context (Aw et al., 2017) that have been used in various national surveys linking social participation to health (Ang, 2016), and are likely to be satisfactorily indicative of total social participation. Finally, certain variables which may lie on the causal pathway such as personal mastery/self-efficacy were not measured in the survey, and thus could not be included in the models. Future research should seek to clarify how these constructs may influence the relationship between social participation, self-management, and health-related quality of life.

In sum, among low-income older adults with at least one chronic condition, informal social participation (but not formal social participation) was positively related to health-related quality of life. Self-management mediated the positive effect of informal social participation on health-related quality of life, but “suppressed” the negative direct effect of formal social participation instead. These results are in line with a prior study finding protective effects for informal social participation against 4-year mortality (Ang, 2016), reflecting the central role of family and friends in determining outcomes for health and well-being. Findings show that this positive effect occurs through self-management—an important pathway to improved well-being among those with chronic illnesses. However, strategies to engage older adults in the community would benefit from further research on causal mechanisms. Insights gleaned from such studies are likely to highlight how best to optimize the benefits from formal social participation (and reduce any of its possible negatives), especially where chronic care is concerned. This is pertinent for members from low-income groups, who are more likely to face both estrangement from family members (Aw et al., 2017) and hostility from formal organizations (Suen & Thang, 2018).

Footnotes

Declaration of Conflicting Interests

The author declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: The author acknowledges use of the services and facilities of the Population Studies Center at the University of Michigan, funded by NICHD Center Grant R24HD041028. The data utilized for this article are from a community survey funded by Singapore Health Services Pte Ltd and Duke-NUS Medical School.

ORCID iD

Shannon Ang

References

Abdin

Subramaniam

Vaingankar

J. A.

Luo

Chong

S. A.

(2015). Population norms for the EQ-5D index scores using Singapore preference weights. Quality of Life Research, 24, 1545-1553. doi:10.1007/s11136-014-0859-5

Abramson

C. M.

(2015). The end game: How inequality shapes our final years. Boston, MA: Harvard University Press.

Adams

K. B.

Leibbrandt

Moon

(2011). A critical review of the literature on social and leisure activity and wellbeing in later life. Ageing & Society, 31, 683-712. doi:10.1017/S0144686X10001091

Agahi

Parker

M. G.

(2008). Leisure activities and mortality: Does gender matter? Journal of Aging and Health, 20, 855-871. doi:10.1177/0898264308324631

Ang

(2016). Social participation and mortality among older adults in Singapore: Does ethnicity explain gender differences? The Journals of Gerontology: Series B: Psychological Sciences and Social Sciences, 73(8), 1470-1479. doi:10.1093/geronb/gbw078

Koh

Y. J.

Wong

M. L.

Vrijhoef

H. J. M.

Harding

S. C.

. . . Hildon

Z. J. L.

(2017). Explaining the continuum of social participation among older adults in Singapore: From “closed doors” to active ageing in multi-ethnic community settings. Journal of Aging Studies, 42(Suppl. C), 46-55. doi:10.1016/j.jaging.2017.07.002

Barlow

Wright

Sheasby

Turner

Hainsworth

(2002). Self-management approaches for people with chronic conditions: A review. Patient Education and Counseling, 48, 177-187. doi:10.1016/S0738-3991(02)00032-0

Benka

Nagyova

Rosenberger

Macejova

Lazurova

van der Klink

J. L. L.

. . . van Dijk

J. P.

(2016). Social participation and health related quality of life in early and established rheumatoid arthritis patients. Journal of Developmental and Physical Disabilities, 28, 381-392. doi:10.1007/s10882-016-9475-3

Berkman

L. F.

Glass

Brissette

Seeman

T. E.

(2000). From social integration to health: Durkheim in the new millennium. Social Science & Medicine, 51, 843-857.

10.

Bodenheimer

Lorig

Holman

Grumbach

(2002). Patient self-management of chronic disease in primary care. Journal of the American Medical Association, 288, 2469-2475. doi:10.1001/jama.288.19.2469

11.

Bodenheimer

Wagner

E. H.

Grumbach

(2002). Improving primary care for patients with chronic illness: The chronic care model, part 2. Journal of the American Medical Association, 288, 1909-1914. doi:10.1001/jama.288.15.1909

12.

Brody

B. L.

Roch-Levecq

Gamst

A. C.

Maclean

Kaplan

R. M.

Brown

S. I.

(2002). Self-management of age-related macular degeneration and quality of life: A randomized controlled trial. Archives of Ophthalmology, 120, 1477-1483. doi:10.1001/archopht.120.11.1477

13.

Cheng

S.-T.

Chan

A. C. M.

(2008). Older people’s realisation of generativity in a changing society: The case of Hong Kong. Ageing & Society, 28, 609-627. doi:10.1017/S0144686X07006903

14.

Chiu

T. M. L.

Tam

K. T. W.

Siu

C. F.

Chau

P. W. P.

Battersby

(2017). Validation study of a Chinese version of partners in health in Hong Kong (c-pih hk). Quality of Life Research, 26, 199-203. doi:10.1007/s11136-016-1315-5

15.

Clark

N. M.

Dodge

J. A.

(1999). Exploring self-efficacy as a predictor of disease management. Health Education & Behavior, 26, 72-89. doi:10.1177/109019819902600107

16.

Cochran

Conn

V. S.

(2008). Meta-analysis of quality of life outcomes following diabetes self-management training. The Diabetes Educator, 34, 815-823. doi:10.1177/0145721708323640

17.

Coretti

Ruggeri

McNamee

(2014). The minimum clinically important difference for EQ-5D index: A critical review. Expert Review of Pharmacoeconomics & Outcomes Research, 14, 221-233. doi:10.1586/14737167.2014.894462

18.

Crimmins

E. M.

(2015). Lifespan and healthspan: Past, present, and promise. The Gerontologist, 55, 901-911. doi:10.1093/geront/gnv130

19.

Davy

Bleasel

Liu

Tchan

Ponniah

Brown

(2015). Effectiveness of chronic care models: Opportunities for improving healthcare practice and health outcomes: A systematic review. BMC Health Services Research, 15(1), Article 194. doi:10.1186/s12913-015-0854-8

20.

Dominick

K. L.

Ahern

F. M.

Gold

C. H.

Heller

D. A.

(2002). Relationship of health-related quality of life to health care utilization and mortality among older adults. Aging Clinical and Experimental Research, 14, 499-508. doi:10.1007/BF03327351

21.

Donnelly

E. A.

Hinterlong

J. E.

(2010). Changes in social participation and volunteer activity among recently widowed older adults. The Gerontologist, 50, 158-169. doi:10.1093/geront/gnp103

22.

Douglas

Georgiou

Westbrook

(2016). Social participation as an indicator of successful aging: An overview of concepts and their associations with health. Australian Health Review, 41, 455-462. doi:10.1071/AH16038

23.

The EuroQol Group. (1990). EuroQol—A new facility for the measurement of health-related quality of life. Health Policy, 16, 199-208. doi:10.1016/0168-8510(90)90421-9

24.

Gabriel

Bowling

A. N. N.

(2004). Quality of life from the perspectives of older people. Ageing & Society, 24, 675-691. doi:10.1017/S0144686X03001582

25.

Galenkamp

Deeg

D. J. H.

(2016). Increasing social participation of older people: Are there different barriers for those in poor health? Introduction to the special section. European Journal of Ageing, 13, 87-90. doi:10.1007/s10433-016-0379-y

26.

Gallant

M. P.

(2003). The influence of social support on chronic illness self-management: A review and directions for research. Health Education & Behavior, 30, 170-195. doi:10.1177/1090198102251030

27.

Grady

P. A.

Gough

L. L.

(2014). Self-management: A comprehensive approach to management of chronic conditions. American Journal of Public Health, 104(8), e25-e31. doi:10.2105/AJPH.2014.302041

28.

Griffith

L. E.

Raina

Levasseur

Sohel

Payette

Tuokko

. . . Patterson

(2017). Functional disability and social participation restriction associated with chronic conditions in middle-aged and older adults. Journal of Epidemiology and Community Health, 71, 381-389.

29.

Gubhaju

Østbye

Chan

(2018). Living arrangements of community-dwelling older Singaporeans: Predictors and consequences. Ageing & Society, 38(6), 1174-1198. doi:10.1017/S0144686X16001495

30.

Cui

Liang

Zhong

. . . Huang

(2016). Social participation, willingness and quality of life: A population-based study among older adults in rural areas of china. Geriatrics & Gerontology International, 17, 1593-1602. doi:10.1111/ggi.12939

31.

Hicks

Tingley

(2011). Causal mediation analysis. Stata Journal, 11, 605-619.

32.

Höfler

(2005). Causal inference based on counterfactuals. BMC Medical Research Methodology, 5(1), Article 28. doi:10.1186/1471-2288-5-28

33.

Imai

Keele

Tingley

(2010). A general approach to causal mediation analysis. Psychological Methods, 15, 309-334. doi:10.1037/a0020761

34.

Imai

Keele

Yamamoto

(2010). Identification, inference and sensitivity analysis for causal mediation effects. Statistical Science, 25, 51-71. doi:10.1214/10-STS321

35.

Jonkman

N. H.

Schuurmans

M. J.

Groenwold

R. H. H.

Hoes

A. W.

Trappenburg

J. C. A.

(2016). Identifying components of self-management interventions that improve health-related quality of life in chronically ill patients: Systematic review and meta-regression analysis. Patient Education and Counseling, 99, 1087-1098. doi:10.1016/j.pec.2016.01.022

36.

Kind

Lafata

J. E.

Matuszewski

Raisch

(2009). The use of QALYs in clinical and patient decision-making: Issues and prospects. Value in Health, 12, S27-S30. doi:10.1111/j.1524-4733.2009.00519.x

37.

Levasseur

Desrosiers

Noreau

(2004). Is social participation associated with quality of life of older adults with physical disabilities? Disability and Rehabilitation, 26, 1206-1213. doi:10.1080/09638280412331270371

38.

Levasseur

Richard

Gauvin

Raymond

. (2010). Inventory and analysis of definitions of social participation found in the aging literature: Proposed taxonomy of social activities. Social Science & Medicine, 71, 2141-2149. doi:10.1016/j.socscimed.2010.09.041

39.

Levasseur

St-Cyr Tribble

Desrosiers

(2009). Meaning of quality of life for older adults: Importance of human functioning components. Archives of Gerontology and Geriatrics, 49(2), e91-e100. doi:10.1016/j.archger.2008.08.013

40.

Y.-P.

Lin

S.-I.

Chen

C.-H.

(2011). Gender differences in the relationship of social activity and quality of life in community-dwelling Taiwanese elders. Journal of Women & Aging, 23, 305-320. doi:10.1080/08952841.2011.611052

41.

Low

L. L.

Wah

M. J.

Tan

S. Y.

Liu

Lee

K. H.

(2016). Housing as a social determinant of health in Singapore and its association with readmission risk and increased utilization of hospital services. Frontiers in Public Health, 4, Article 109. doi:10.3389/fpubh.2016.00109

42.

Luo

Wang

Thumboo

Lim

Y.-W.

Vrijhoef

H. J. M.

(2014). Valuation of EQ-5D-3L health states in Singapore: Modeling of time trade-off values for 80 empirically observed health states. Pharmacoeconomics, 32, 495-507. doi:10.1007/s40273-014-0142-1

43.

MacKinnon

D. P.

Krull

J. L.

Lockwood

C. M.

(2000). Equivalence of the mediation, confounding and suppression effect. Prevention Science, 1, 173-181. doi:10.1023/A:1026595011371

44.

Malhotra

Chan

Malhotra

Østbye

(2011). Fifteen dimensions of health among community-dwelling older Singaporeans. Current Gerontology and Geriatrics Research, Article 128581.

45.

Malhotra

Bautista

M. A. C.

Tan

N. C.

Tang

W. E.

Tay

Tan

A. S. L.

. . . Vaillancourt

(2019). Bilingual text with or without pictograms improves elderly Singaporeans’ understanding of prescription medication labels. The Gerontologist, 59(2), 378-390. doi:10.1093/geront/gnx169

46.

Martire

L. M.

Lustig

A. P.

Schulz

Miller

G. E.

Helgeson

V. S.

(2004). Is it beneficial to involve a family member? A meta-analysis of psychosocial interventions for chronic illness. Health Psychology, 23, 599-611. doi:10.1037/0278-6133.23.6.599

47.

Mehta

K. K.

(2006). A critical review of Singapore’s policies aimed at supporting families caring for older members. Journal of Aging & Social Policy, 18(3-4), 43-57. doi:10.1300/J031v18n03_04

48.

Mikula

Nagyova

Krokavcova

Vitkova

Rosenberger

Szilasiova

. . . van Dijk

J. P.

(2017). Self-esteem, social participation, and quality of life in patients with multiple sclerosis. Journal of Health Psychology, 22, 984-992. doi:10.1177/1359105315621778

49.

T. P.

Broekman

B. F. P.

Niti

Gwee

Kua

E. H.

(2009). Determinants of successful aging using a multidimensional definition among Chinese elderly in Singapore. The American Journal of Geriatric Psychiatry, 17, 407-416. doi:10.1097/JGP.0b013e31819a808e

50.

Niedzwiedz

C. L.

Richardson

E. A.

Tunstall

Shortt

N. K.

Mitchell

R. J.

Pearce

J. R.

(2016). The relationship between wealth and loneliness among older people across Europe: Is social participation protective? Preventive Medicine, 91, 24-31. doi:10.1016/j.ypmed.2016.07.016

51.

Phan

T. P.

Alkema

Tai

E. S.

Tan

K. H. X.

Yang

Lim

W.-Y.

. . . Cook

A. R.

(2014). Forecasting the burden of type 2 diabetes in Singapore using a demographic epidemiological model of Singapore. BMJ Open Diabetes Research & Care, 2(1), Article e000012.

52.

Pinto

J. M.

Neri

A. L.

(2017). Trajectories of social participation in old age: A systematic literature review. Revista Brasileira de Geriatria e Gerontologia, 20, 259-272.

53.

Raymond

Sevigny

Tourigny

Vezina

Verreault

Guilbert

A. C.

(2013). On the track of evaluated programmes targeting the social participation of seniors: A typology proposal. Ageing & Society, 33, 267-296. doi:10.1017/S0144686X11001152

54.

Richard

A. A.

Shea

(2011). Delineation of self-care and associated concepts. Journal of Nursing Scholarship, 43, 255-264. doi:10.1111/j.1547-5069.2011.01404.x

55.

Rosland

A.-M.

Piette

J. D.

(2010). Emerging models for mobilizing family support for chronic disease management: A structured review. Chronic Illness, 6, 7-21. doi:10.1177/1742395309352254

56.

Schulman-Green

Jaser

Martin

Alonzo

Grey

McCorkle

. . . Whittemore

(2012). Processes of self-management in chronic illness. Journal of Nursing Scholarship, 44, 136-144. doi:10.1111/j.1547-5069.2012.01444.x

57.

Sim

L. L.

S. M.

Han

S. S.

(2003). Public housing and ethnic integration in Singapore. Habitat International, 27, 293-307. doi:10.1016/S0197-3975(02)00050-4

58.

Singapore National Population and Talent Division. (2015). Population in brief 2015. Singapore: National Population and Talent Division, Prime Minister’s Office.

59.

Sørensen

L. V.

Waldorff

F. B.

Waldemar

(2008). Social participation in home-living patients with mild Alzheimer’s disease. Archives of Gerontology and Geriatrics, 47, 291-301. doi:10.1016/j.archger.2007.08.011

60.

Steverink

Lindenberg

Slaets

J. P. J.

(2005). How to understand and improve older people’s self-management of wellbeing. European Journal of Ageing, 2, 235-244. doi:10.1007/s10433-005-0012-y

61.

Suen

Thang

L. L.

(2018). Contextual challenges and the mosaic of support: Understanding the vulnerabilities of low-income informal caregivers of dependent elders in Singapore. Journal of Cross-Cultural Gerontology, 33(2), 163-181. doi:10.1007/s10823-017-9334-4

62.

Takada

Sashika

Wakabayashi

Hirayasu

(2016). Social participation and quality-of-life of patients with traumatic brain injury living in the community: A mixed methods study. Brain Injury, 30, 1590-1598. doi:10.1080/02699052.2016.1199901

63.

Testa

M. A.

Simonson

D. C.

(1996). Assessment of quality-of-life outcomes. The New England Journal of Medicine, 334, 835-840. doi:10.1056/NEJM199603283341306

64.

Thoits

P. A.

(2011). Mechanisms linking social ties and support to physical and mental health. Journal of Health and Social Behavior, 52, 145-161. doi:10.1177/0022146510395592

65.

Tomioka

Kurumatani

Hosoi

(2017). Association between social participation and 3-year change in instrumental activities of daily living in community-dwelling elderly adults. Journal of the American Geriatrics Society, 65, 107-113. doi:10.1111/jgs.14447

66.

van Groenou

M. B.

Deeg

D. J. H.

(2010). Formal and informal social participation of the “young-old” in the Netherlands in 1992 and 2002. Ageing & Society, 30, 445-465. doi:10.1017/S0144686X09990638

67.

van Ingen

. (2008). Social participation revisited. Acta Sociologica, 51, 103-121. doi:10.1177/0001699308090038

68.

Wang

Tan

N.-C.

Tay

E.-G.

Thumboo

Luo

(2015). Cross-cultural measurement equivalence of the 5-level EQ-5D (EQ-5D-5L) in patients with type 2 diabetes mellitus in Singapore. Health and Quality of Life Outcomes, 13(1), Article 103. doi:10.1186/s12955-015-0297-2

69.

Wattana

Srisuphan

Pothiban

Upchurch

S. L.

(2007). Effects of a diabetes self-management program on glycemic control, coronary heart disease risk, and quality of life among Thai patients with type 2 diabetes. Nursing & Health Sciences, 9, 135-141. doi:10.1111/j.1442-2018.2007.00315.x