Older age is a transitional period when people experience changes not only in physical health, but also in social roles (for example, retirement, children are grown) that can influence opportunities for social participation.1 Epidemiological studies suggest that social activities may be particularly important for older adults.2 The possible health benefits include reduced risk of mortality,3,4 disability5-7 and depression,8,9 and better cognitive health,10-12 self-rated health13,14 and health-related behaviours.2 Thus, social participation has been integrated into research and policy frameworks of aging. For instance, social engagement—involvement in meaningful activities and maintaining close relationships—is a component of successful aging.15
The relationships between social participation and health are not well understood, and may operate through multiple pathways.2,16 For example, the physiological impact of social isolation is hypothesized to influence the neuroendocrine and immune systems.16 As well, social ties may encourage individuals to engage in health-promoting behaviours such as physical activity and seeking medical care, or to refrain from damaging ones such as smoking.2,16
The psychological effects of social connectedness may include feelings of self-efficacy, a sense of meaning and purpose, and better mental health.2,16 In particular, interactions that provide social support are thought to be facilitators of health, in which case, perceived social support would act as a mediating factor between social participation and health and well-being.9 Moreover, an individual’s perceptions of the availability of social support are thought to be more important than received support, which is confounded with need.17
This study examines the relationship between the number of social activities in which seniors “frequently” participate and three measures of health and well-being: self-perceived health, loneliness, and life dissatisfaction. The extent to which social support mediates the effect of social participation on health and well-being is considered. In addition, for the first time in a nationally representative Canadian study, reported barriers to greater social participation are examined.
The data for this analysis are from the 2008/2009 Canadian Community Health Survey (CCHS)―Healthy Aging. This cross-sectional survey collected information about factors, influences and processes that contribute to healthy aging from people aged 45 or older living in private dwellings in the ten provinces. The sampling frame excluded full-time members of the Canadian Forces and residents of the three territories, Indian reserves, Crown lands, institutions, and some remote areas. The survey was conducted from December 1, 2008 through November 30, 2009, using computer-assisted personal interviewing. Response rates were 80.8% (household level), 92.1% (person level), and 74.4% (combined), for a final sample of 30,865 respondents. This analysis uses a sample of 16,369 seniors (65 or older), representing 4.4 million people.
Frequent social participation
Respondents were asked how often in the past 12 months (at least once a day, at least once a week, at least once a month, at least once a year, never) they participated in eight different activities. Frequent participation was classified as at least weekly for:
Frequent participation was classified as at least monthly for activities typically done less often:
Health and well-being
CCHS―Healthy Aging respondents were asked: “In general, would you say your health is ...” The response options were dichotomized to reflect positive (excellent/very good/good) versus poor (fair/poor) self-perceived health.
The Three-Item Loneliness Scale18 measures an individual’s loneliness. On a three-point Likert scale (hardly ever, some of the time, often), respondents answered the questions: “How often do you feel:
Higher scores indicated greater loneliness; the distribution was skewed toward lower scores. Scores were dichotomized to classify respondents in the top quintile of the frequency distribution as experiencing loneliness. Those classified as experiencing loneliness responded “some of the time” to two or more questions, or “often” to one or more questions.
Respondents were asked, “How do you feel about your life as a whole right now?” and answered based on a scale where 0 meant “very dissatisfied” and 10 meant “very satisfied.” Respondents in the bottom quintile of the frequency distribution (a score of 6 or less) were classified as having life dissatisfaction.
In this study, the three measures of health and well-being were mildly to moderately correlated with each other. The Pearson correlation coefficients were -0.17 for self-perceived health and loneliness; -0.36 for self-perceived health and life dissatisfaction; and 0.23 for loneliness and life dissatisfaction. Despite some overlap in the three variables, they are treated as individual constructs in this analysis.
Three age groups were defined: 65 to 74, 75 to 84, and 85 or older. In logistic regression models, age was measured as a continuous variable and contained values of 65 or more.
Household income quintiles were defined: lowest, low-middle, middle, high-middle and highest.
Highest level of education was categorized as: less than secondary graduation, secondary graduation, some postsecondary, and postsecondary graduation.
Retirement status, based on Statistics Canada’s standard definition of retirement (http://www.statcan.gc.ca/concepts/definitions/retirement-retraite-eng.htm),was categorized as completely retired and not completely retired. To be considered completely retired, the respondent could not be in the labour force and had to have received income from “retirement-like sources” during the past 12 months.19 Respondents older than 75 were excluded from the labour force module of the CCHS—Healthy Aging, and so were considered to be completely retired for this analysis. Retirement-like income sources included dividends and interest; benefits from the Canada or Quebec Pension Plan; job-related retirement pensions, superannuation and annuities; RRSP or RRIF; and Old Age Security and Guaranteed Income Supplement.
The Health Utilities Index (HUI) Mark III assesses functional health status in eight domains: vision, hearing, speech, ambulation, dexterity, emotion, cognition, and pain and discomfort.20,21 Overall scores were categorized into four levels of disability: none (1.00), mild (0.89 to 0.99), moderate (0.70 to 0.88), and severe (less than 0.70)
The number of behavioural risk factors was based on whether respondents reported heavy drinking (five or more drinks on one or more occasions monthly); were smokers (daily smoker or had quit less than 15 years ago), and were physically inactive (a score below the mean on the Physical Activity Scale for the Elderly).22
Social support was measured based on the Medical Outcomes Study (MOS) Social Support Survey.17 This is a measure of perceived rather than actual social support received. All questionnaire items measuring social support use a standard preamble: “How often is each of the following kinds of support available to you if you need it?” Each item was scored according to the frequency with which support was available: none of the time (score 0), a little of the time (1), some of the time (2), most of the time (3), and all of the time (4).
For each dimension of social support, a variable was derived based on the summed scores of responses to the individual items. The frequency distribution of responses for each dimension was skewed toward higher scores. For ease of interpretation in univariate and bivariate analysis, each variable was dichotomized so that respondents with scores in the lowest tercile of the frequency distribution were considered to have low social support (12 or less for positive interaction; 12 or less for tangible support; 25 or less for emotional or informational support; 10 or less for affection). Social support scores were used in their continuous forms (based on their summed scores) in multiple logistic regression models.
Frequencies and cross-tabulations weighted to be representative of the population aged 65 or older who resided in the provinces in 2008/2009 were produced to estimate the prevalence of social participation and barriers to social participation in the household population, and to examine characteristics associated with the health and well-being outcomes (Appendix Table A).
Logistic regression models were used to assess associations between the number of social activities in which a person frequently participated and measures of health and well-being. All analyses combined men and women in the same sample. An initial series of models controlled for the number of social activities in which respondents frequently participated, and age and sex. A second series added socio-demographic and health covariates that might also be associated with health and well-being: household income, education, retirement status, disability, and behavioural risk factors. To assess the mediating role of social support, the final models included the four social support dimensions. Because of the potential for multicolinearity, each social support variable was entered singly into the fully controlled models. This study presents only the results of the final models.
To account for the complex design of the CCHS, standard errors, coefficients of variation and confidence intervals were estimated with the bootstrap technique.23,24 The statistical significance level was set at <0.05.
On the whole, Canadian seniors tended to report positive health and well-being—more than three-quarters perceived their health to be good, very good or excellent; less than one in five was classified as lonely or dissatisfied with life (Table 1). Younger seniors (65 to 74) were more likely than older seniors to have positive self-perceived health, and less likely to be lonely or to report life dissatisfaction. Women were more likely than men to be lonely.
Seniors with higher levels of household income and education, and who were not completely retired, were more likely to report positive self-perceived health, and less likely to be lonely or dissatisfied with life than were those in lower-income households, with less education, and who were retired.
The more severe the disability and the greater the number of behavioural risk factors, the less likely were seniors to report positive self-perceived health, and the more likely they were to be lonely or dissatisfied with life.
Seniors with low social support were less likely than were those with high social support to report positive self-perceived health, and more likely to be lonely and dissatisfied with life.
The majority of seniors (80%) were frequent participants in at least one social activity (Appendix Table A). As the number of activities increased, their likelihood of reporting positive self-perceived health rose, and their likelihood of reporting loneliness or life dissatisfaction decreased (Table 1).
Activities with family or friends were the most common, with just over half of senior men and women participating frequently in this type of social activity (Figure 1). Women were more likely than men to be frequent participants in family and friend, church, educational and “other” activities, while men were more likely to be frequent participants in sports. With the exception of church and “other” activities, participation in most types of social activities was lower at older ages (Figure 2).
The number of social activities in which individuals frequently participated was strongly and significantly related to each of the health and well-being outcomes, independent of age and sex (data not shown). Even when socio-demographic and health characteristics were taken into account, the relationships between social participation and each health and well-being measure persisted, although they were attenuated (data not shown).
In the full models, which also controlled individually for the four dimensions of social support, the relationships between social participation and health and well-being were further attenuated, but remained significant in all but one instance (Table 2). Social participation was not significantly associated with loneliness when positive social interaction was controlled. In all other instances, a gradient in odds ratios was apparent with each increase in the number of activities, although to varying degrees for each outcome. Except for the association between affection and positive self-perceived health, each dimension of social support was, itself, significantly associated with each health and well-being outcome.
Table 2 Adjusted odds ratios relating number of frequent social activities and social support dimensions to positive self-rated health, loneliness and life dissatisfaction, household population aged 65 or older, Canada, excluding territories, 2008/2009
In multivariate analysis that controlled for age, sex and socio-demographic and health characteristics (but not social support), some factors that were significant in the bivariate analysis, such as income, disability and health behaviours (Table 1), remained significantly associated with each outcome, but education and retirement status were significantly associated only with self-perceived health (data not shown). When social support dimensions were added, the results for socio-demographic and health covariates were similar, except that the association between household income and loneliness was significant only for the highest income category.
Nearly one in four seniors (24%) reported that they would have liked to have participated in more social, recreational or group activities in the past year. Younger seniors and women were more likely to have felt this way (Table 3).
Table 3 Percentage reporting desire to have participated in more social activities in past 12 months, by age group and factors preventing particpation, household population aged 65 or older, Canada excluding territories, 2008/2009
The most commonly mentioned obstacle to participating in more activities was a health limitation (33% of men, 35% of women). Being too busy was also a leading reason, but more so among men (28%) than women (16%). Personal or family responsibilities prevented about 1 in 10 seniors from participating in more activities. Women were more likely than men to report not wanting to go alone to an activity (17% versus 9%) or transportation problems (11% versus 4%).
Social participation may not be entirely dependent upon personal choice—external factors can play a role. For example, the cost and the availability of activities in the area or at a suitable time or location can influence participation. Such barriers were reported by 4% to 9% of Canadian seniors.
The results of this study support other research,2,14 in that social relationships were shown to be significantly associated with health and well-being independent of socio-demographic and health factors. In earlier research, each outcome examined in this study—self-perceived health,25 loneliness,26 and life dissatisfaction27—has been linked to poor health and mortality.
While the optimal amount of social participation depends on the individual, a gradient, or dose-response relationship, appears to exist. The greater the number of frequent social activities, the higher the odds of positive self-perceived health, and the lower the odds of loneliness and life dissatisfaction. This is consistent with research that has found seniors with a wider range of social ties have better well-being.28
The associations between social participation and health and well-being were attenuated, but persisted, when socio-demographic and health factors were controlled. When social support was included in the models, the associations were further attenuated, but generally remained. However, when positive social interaction was taken into account, the association between frequent social participation and loneliness was no longer significant.
The elements in the positive social interaction dimension of social support (has someone to have a good time with, get together with for relaxation, do things with to get his/her mind off things, or do something enjoyable with) seem closely aligned with social participation. However, social participation and the positive social interaction dimension of social support were only mildly correlated (Pearson correlation 0.20), indicating that multicollinearity does not account for the finding. This indicates that the measure of interaction used here (number activities in which one frequently participates) and the perceived availability of positive social interaction are not interchangeable concepts.
Some research suggests that it is the quality, not the size, of social networks that matters for the relationship with health and well-being.2,9,29-32 In this study, dimensions of perceived social support were used to approximate the quality of social interactions. The fact that social support partially or completely mediated the associations, and that individual dimensions of social support were, themselves, independently associated with measures of health and well-being, corroborates this hypothesis.
The strong associations between social participation and health and well-being emphasize the importance of addressing the barriers faced by the nearly one-quarter of seniors who reported a desire to participate in more social activities.
Because this is a cross-sectional study, the possibility of reverse causality cannot be ruled out; that is, people in poor health may be unable to maintain social participation, and those who participate frequently may be in better health. Nonetheless, the relationship between social participation and health and well-being persisted even after accounting for functional health status. Some longitudinal studies have found similar results.33,34
People who are not healthy may still benefit from social participation, perhaps more so. However, tests for interaction effects between level of disability and social participation in models of health and well-being outcomes were not significant in this analysis (data not shown). It is also likely that there are reciprocal effects between social participation and health and well-being,2 such that better health allows for greater social participation, which, in turn, improves or maintains health, allowing for the maintenance or increase in the level of social participation.
Seniors in care institutions were excluded from the survey. However, results were similar in a study of institutionalized seniors.35 Specifically, seniors who participated in social and recreational activities were more likely than those who did not participate in such activities to report positive self-perceived health.
The types of social activities about which respondents to the survey were asked did not include the internet or social media. Seniors’ use of the internet tends to be for communication,36 and has been associated with lower levels of loneliness.37 Exclusion of online activity from this study may underestimate seniors’ social participation, and also, the extent of associations between social participation and health and well-being.
According to the 2008/2009 Canadian Community Health Survey—Healthy Aging, four-fifths of seniors were frequent participants in social activities. The results of this analysis highlight the importance of frequent social participation to maintaining quality of life. Of particular relevance to policy and program development are the reported barriers to seniors’ social participation. In addition to the frequency of social participation, future research could focus on seniors’ satisfaction with social participation and longitudinal associations with health and well-being.
Statistics Canada thanks all participants for their input and advice during the development of the Canadian Community Health Survey―Healthy Aging. The survey content was developed by the Health Statistics Division at Statistics Canada in consultation with Health Canada, the Public Health Agency of Canada, and experts conducting the Canadian Longitudinal Study on Aging (CLSA), a major strategic initiative of the Canadian Institute for Health Research. Consultations included stakeholders from Human Resources and Social Development Canada and provincial and territorial health ministries. The addition of 5,000 respondents aged 45 to 54 was funded by the CLSA.