| HOME | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|---|
| ||||||||||||||||||||||||||||||||
RESEARCH ARTICLE |
Department of Human Development and Family Studies, University of Wisconsin-Madison.
 |
Abstract |
|---|
 TOP Abstract MethodsResultsDiscussionReferences |
|---|
Methods. Longitudinal data came from 4,646 respondents aged 35 to 92 in the National Survey of Families and Households (1987–1993) who reported having no functional limitations at Time 1.
Results. Multivariate models controlling for sociodemographic factors, as well as psychological well-being at Time 1, indicated that developing functional limitations over a 5-year period was associated with greater increases in depressive symptoms and lower levels of personal growth. Increases in depressive symptoms, however, were less severe among men who were continuously involved in recreational groups than among men who were not continuously involved in recreational groups. Additionally, the association between developing functional limitations and lower levels of personal growth did not hold for men or women who continuously participated in religious groups.
Discussion. Findings suggest that continuous participation in certain types of voluntary groups can moderate the problematic effects of developing functional limitations on psychological well-being.
FUNCTIONAL limitations—or having health conditions that interfere with one's ability to complete daily tasks of living—have profound implications for the quality of individuals' lives, as well as for social systems that must provide care for people with disabilities (Lawton, 1991
). Given the challenges that functional health impairments pose for both individuals and society, much research has focused on identifying the causes of functional limitations so as to inform efforts to lower rates of impairment (e.g., Kono, Kai, Sakato, & Rubenstein, 2004; McCusker, Kakuma, & Abrahamowicz, 2002). Although promoting functional health at the individual and population levels is important, the fact remains that an increasing number of adults in the United States have functional limitations (American Association of Retired Persons, 1998). Advancing the longevity of morbidity hypothesis, scholars have suggested that it is becoming increasingly common for people with functional limitations to survive more years living with disability (Zarit & Zarit, 1998). As a rising number of adults acquire and maintain functional limitations, scholarly efforts to more fully and systematically understand the consequences of living with functional limitations become increasingly critical.
The aim of this study was in part to address this need by investigating variation in individuals' psychological well-being following their developing functional limitations. Guided by continuity theory and using prospective, longitudinal data from the National Survey of Families and Households (1987–1993), this study examined continuous participation in voluntary groups (recreational, religious, and civic) as a protective factor against the deleterious psychological consequences of having developed functional limitations.
Theoretical Background
Continuity theory, a well-developed social gerontological theory, provides insight into how ongoing voluntary group participation might help individuals to avoid the negative psychological effects of developing functional limitations. Continuity theory advances the assumption that as people age, they are likely to experience significant life changes, such as widowhood and retirement (Atchley, 1989, 1993). The theory posits that when facing such changes, individuals are predisposed to think of themselves and their lives as having coherence and consistency. Continuity theory postulates that people strive to achieve a sense of stability both in terms of internal continuity (i.e., maintaining a coherent sense of who one is) and external continuity (i.e., maintaining a coherent sense of one's social world). Although continuity theory recognizes that striving towards continuity might not be adaptive for well-being under all circumstances, the theory posits that continuity is usually an effective means by which people can develop and adapt as a result of change (Atchley, 1989).
Continuity theory explicates mental and behavioral strategies that can foster an individual's sense of continuity. For example, interpreting new information about one's self in a way that validates one's existing self-concept can help a person maintain a sense of internal continuity, and soliciting the company of familiar persons can enhance a person's sense of external continuity. Continuity theorists have discussed a variety of life domains in which people might employ continuity-maintaining strategies, including the domain of activities (Atchley, 1993). Researchers have identified participation in voluntary groups, in particular, as a promising source of activity-related continuity in that activities occurring within overt social environments—such as within the context of a stable group infrastructure or within well-defined role relationships can more readily provide individuals with a sense of stability over time (Atchley, 1989).
In sum, continuity theory suggests that as people experience significant life changes, individuals engage in mental and behavioral strategies that help them to maintain a sense of stability in terms of their self and social worlds. As such, continuity theory suggests that as adults make the transition to a state of impaired functional health (i.e., a significant life change), continuous participation in voluntary groups (i.e., an activity-related source of continuity) can help them to maintain higher levels of well-being (i.e., adapt more beneficially to this change).
Empirical Background
A substantial body of evidence suggests that loss of physical health is a risk factor for individuals' poor psychological well-being (Okun, 1984). Moreover, several studies drawing on longitudinal data from large population samples have indicated that, overall, individuals' greater levels of physical disability are associated with their poorer mental health over time (Kunzmann, Little, & Smith, 2000; Taylor & Lynch, 2004). Nevertheless, findings from other studies have indicated that there is substantial variation in psychological well-being among people with physical health problems (Brief, Butcher, & George, 1993) and that individuals' perceptions of their health-related quality of life and overall well-being do not entirely overlap (Spiro & Bosse, 2000). These findings suggest that there are likely factors that can protect aspects of adults' well-being when they are faced with the challenges of impaired functional health.
Scholars have long discussed that social integration and productive activities can promote optimal states of well-being, particularly among individuals facing age-related changes (Rowe & Kahn, 1997). Building on this idea, researchers have focused increasingly on the physical and mental health effects of participation in voluntary groups, or groups that are (a) outside of one's private network of friends and families, (b) not mandated by the government or the market, and (c) formed as result of shared interests among a collective of individuals. Previous empirical studies have documented linkages between voluntary group participation and higher levels of well-being. For example, previous research has indicated that formal religious participation is associated with better physical and mental health (see Ellison & Levin, 1998, for a review), participation in activities within formal social groups is predictive of adults' greater happiness over time (Menec, 2003), and formal volunteering is linked with fewer depressive symptoms and better self-reported health and functioning (Morrow-Howell, Hinterlong, Rozario, & Tang, 2003).
Although this growing body of research demonstrates the potential benefits of voluntary group participation for individuals' well-being, understanding of the linkages between voluntary group participation and enhanced well-being is far from complete. Although scholars have posited that voluntary group participation is particularly salient for the well-being of more vulnerable adults (e.g., Silverstein & Parker, 2002), there has been little investigation of whether voluntary group participation can protect individuals who are specifically at risk for poor mental health. Additionally, few studies regarding the associations between voluntary group participation and psychological well-being have simultaneously examined participation in multiple types of voluntary groups, as well as multiple dimensions of psychological well-being. Given previous findings that different psychosocial processes might lead to participation in distinct types of voluntary groups (Janoski & Wilson, 1995), and that the effects of voluntary group participation might be particularly salient in terms of certain aspects of psychological well-being (Greenfield & Marks, 2004), it is important for additional research to simultaneously examine a range of types of voluntary group participation as well as multiple mental health outcomes. Also, very few studies on the linkages between voluntary group participation and psychological well-being have considered the dynamic nature of such participation (i.e., that participants in voluntary groups differ with respect to how long they have been involved in such groups). Continuity theory suggests conditions under which more long-term participation in voluntary groups might be particularly advantageous for individuals' psychological well-being. Finally, although scholars have discussed the extent to which voluntary group participation differs for men and women (e.g., Cutler & Hendricks, 2000), few studies have explicitly examined whether the potential psychological consequences of participation in diverse types of voluntary groups vary by gender.
This study aimed to address these gaps in the literature on voluntary group participation and psychological well-being, as well as gaps in research on variation in the effects of functional limitations on psychological well-being, by examining continuous participation in voluntary groups as a protective factor against the negative psychological effects of having developed functional limitations for men and women.
Hypotheses
Guided by continuity theory and previous empirical work, we formulated a conceptual model that suggests the risk-buffering effect of continuous participation in voluntary groups for adults' psychological well-being in the face of developing functional limitations (Figure 1). We derived three hypotheses (H) from this model and formulated two research questions (RQ).
|
|
METHODS |
|---|
|
TOPAbstractMethodsResultsDiscussionReferences |
|---|
; Sweet, Bumpass, & Call, 1988). At Time 1 (T1; 1987–1988), the study included interviews with 13,007 persons (response rate = 75%), and at Time 2 (T2; 1992–1993) the study included follow-up interviews with 10,007 respondents (response rate = 82%). The current study's subsample included all respondents who completed interviews at T1 and T2 (to allow for longitudinal analyses), who were at least 35 years old at T1 (to focus analyses on midlife and older adults), and who reported no functional limitations at T1 (to allow for a prospective design), yielding a sample size of 4,646 respondents.
Measures
Depressive symptoms
At T1 and T2 the study included a 12-item version of the Center for Epidemiologic Studies–Depression scale (Radloff, 1977). This index asked respondents how many days during the past week they had experienced various depressive symptoms, such as feeling that everything they did was an effort and talking less than usual. If respondents answered at least 6 of the 12 items, they received an average-item score, with higher scores indicating more depressive symptoms. To partly, although not entirely, correct for a positive skew in the distribution of respondents' scores on this scale, we logged respondents' scores after adding a constant of 1. Cronbach's alpha for depressive symptoms was.93 at both T1 and T2. Table 1 provides correlations and descriptives for all analytic variables.
|
), we deemed it valuable to explore this aspect of psychological well-being in our study. At T2, the study included Ryff's three-item Personal Growth index. Ryff created this index for large survey use as an additive measure designed to represent the conceptual breadth of personal growth, which she found in factor analyzing her 20-item scale (Ryff & Keyes, 1995). This index asked respondents to report the degree to which they agreed or disagreed with statements indicating feelings of personal growth on a six-point continuum (1 = strongly disagree), including "For me, life has been a continuous process of learning, changing, and growth," "I gave up trying to make big improvements or changes in my life a long time ago," and "I think it is important to have new experiences that challenge how I think about myself and the world." Items were coded and summed such that higher scores indicated more personal growth. This additive index correlates highly (r >.70) with its parent 20-item, highly reliable scale (Ryff & Keyes, 1995).
Developed functional limitations
Respondents reported on functional limitations at T1 and T2 by responding to a set of items asking: "Do you have a physical or mental condition that limits your ability to: (a) care for personal needs, such as dressing, eating or going to the bathroom; (b) move about inside the house; (c) do day-to-day household tasks; (d) climb a flight of stairs; (e) walk six blocks?" To create this study's prospective, longitudinal design, we included in the sample only respondents who reported "no" to all of these items at T1. To assess whether respondents transitioned into a state of impaired functional health between T1 and T2, we coded as 1 respondents who reported that a health condition at all limited their ability to engage in any of the activities at T2. We coded all other respondents as 0 on this variable.
Continuous participation in voluntary groups
Because the focus of this study was on voluntary group participation that is specifically continuous over time, we created dichotomous measures to assess whether respondents reported participation in voluntary groups at both T1 and T2 (1 = continuous participation; 0 = noncontinuous participation). More specifically, on the measure of continuous participation in recreational groups, we coded as 1 respondents who reported at least monthly participation in sports, hobby or garden groups, or discussion groups at both T1 and T2. On the measure of continuous participation in religious groups, we coded as 1 respondents who reported at least monthly participation in church-affiliated groups (with the measure specifying that the respondents were not to consider participation in religious services for this item) at both T1 and T2. On the measure of continuous participation in civic groups, we coded as 1 respondents who reported at least monthly participation in service clubs, fraternal groups, or political groups at both T1 and T2.
Control variables
Previous studies have demonstrated that several sociodemographic factors—such as race/ethnicity, gender, age, education, income, parental status, marital status, and employment status—are associated with physical health, voluntary group participation, and/or psychological well-being (e.g., Mroczek & Kolarz, 1998; Ryff, 1995). This study statistically controlled for these factors in all models to reduce confounding that might have occurred between these variables and variables of primary substantive interest. We created dichotomous variables for gender, employment status, marital status, and whether respondents had a child younger than 19 years old in the household at T2. We created multicategorical variables to indicate respondents' educational attainment at T2 and race/ethnicity at T1, and we created continuous variables for respondents' age at T1 and household income at T2.
Analytic Sequence
We first examined several sets of bivariate correlations to provide evidence for the utility of multivariate models (see Table 1). The modest size of the correlation between the dependent variables (r = –.20), as well as among the types of continuous voluntary group participation (.05 < r 
.14), indicated that these variables were not empirically redundant with each other. We also found that the correlations between the dichotomous measure of functional limitations and the dichotomous measures of continuous participation in voluntary groups were small (|r| –.09). Furthermore, we estimated binomial logit models that regressed the log odds of continuous participation in each type of voluntary groups on all control variables, as well as on a continuous measure of the number of functional limitations acquired between T1 and T2 (ranging from 0 to 5). The coefficient for functional limitations did not achieve statistical significance at a robust level (p .05) in any of these models, further suggesting that continuous participation in voluntary groups was not significantly associated with the severity of functional decline.
In additional preliminary analyses, we estimated models that included three-way interaction terms indicating the product among developing functional limitations, continuous participation in recreational or religious groups, and a dichotomous variable indicating whether respondents were at least 47 years old (the median age for the sample at T1). (We did not estimate models with three-way interaction terms involving continuous participation in civic groups because only 18 respondents developed functional limitations between T1 and T2 and also reported continuous participation in civic groups.) None of the interaction terms achieved statistical significance at a robust level (p .05, two-tailed). Therefore, we subsequently analyzed data from respondents of all ages together.
We estimated ordinary least squares regression models to address our hypotheses and research questions. To provide evidence for the associations between the independent variables and changes in respondents' psychological well-being, we regressed the well-being outcomes at T2 on the control variables, as well as on respondents' scores on depressive symptoms at T1. (Because the National Survey of Families and Households did not elicit respondents' reports of personal growth at T1, all models estimated for respondents' personal growth included respondents' depressive symptoms at T1 as a baseline well-being control.) Because results based on analyses with the weighted and unweighted data were comparable, we report results based on the unweighted data because these analyses provide estimates with more reliable standard errors (Winship & Radbill, 1994).
To explore RQ1, we first estimated models that included two-way interaction terms between gender and continuous participation in each type of voluntary group (e.g., Gender x Continuous participation in recreational groups). To explore RQ2, we estimated models that included three-way interaction terms between gender, continuous participation in religious or recreational groups, and functional limitations (e.g., Gender x Continuous participation in recreational groups x Functional limitations). To examine evidence for H1, we estimated multivariate models by regressing the well-being outcomes on the dichotomous variable indicating whether respondents had developed functional limitations between T1 and T2. To examine evidence for H2, we added the block of three dichotomous variables indicating continuous participation in each of the voluntary groups to the previous set of models, plus any two-way interaction terms including gender that had achieved statistical significance (p .05, two-tailed) in previous analyses. To examine evidence for H3, we separately added two-way interaction terms (i.e., the product between each type of continuous voluntary group participation and whether respondents had developed functional limitations between T1 and T2) to a model containing all main effect variables in addition to relevant three-way interaction terms that had achieved statistical significance in previous analyses.
|
RESULTS |
|---|
|
TOPAbstractMethodsResultsDiscussionReferences |
|---|
.05; Table 2, Model 2). We used estimates from this model to graph predicted scores for men and women who differed as to whether they continuously participated in civic groups across the 5-year period. The baseline model included respondents at the mean level on continuous variables and in the zero categories for dichotomous variables. As Figure 2 demonstrates, continuous participation in civic groups was associated with lower levels of depressive symptoms for men but not for women.
|
|
.05; Table 2, Model 3). This three-way interaction term indicates that the influence of continuous participation in recreational groups on the association between functional limitations and depressive symptoms differed for men and women. We further interpret this result when describing evidence for H3.
Developing Functional Limitations and Psychological Well-Being
H1 predicted that respondents who developed functional limitations over a 5-year period would report poorer psychological well-being than respondents who did not develop functional limitations. Results from models that evaluated this hypothesis (refer to Tables 2 and 3, Model 1) indicated that respondents who developed functional limitations reported greater increases in depressive symptoms (b =.24, p .001) and lower levels of personal growth (b = –.44, p .001) in contrast to respondents who did not develop functional limitations. Pending interaction results with respect to H3, these analyses provided evidence in support of H1.
|
.01); however, continuous participation in recreational groups was not associated with differences in depressive symptoms, and continuous participation in civic groups was associated with differences in depressive symptoms among men but not among women. Furthermore, continuous participation in recreational groups was associated with higher levels of personal growth at T2 (b =.42, p .001), as was continuous participation in religious groups (b =.44, p .001), but continuous participation in civic groups was not associated with levels of personal growth at T2. These analyses provided partial support for H2.
Continuous Participation in Voluntary Groups as a Protective Factor
H3 predicted that respondents who developed functional limitations and who continuously participated in voluntary groups would report better psychological well-being than respondents who also developed functional limitations but who did not continuously participate in voluntary groups. In addition to further interpreting the association between the three-way interaction among gender, continuous participation in recreational groups, and functional limitations and depressive symptoms reported previously, we also estimated models with two-way interaction terms between continuous participation in the remaining types of voluntary groups and functional limitations in order to evaluate evidence for this hypothesis (refer to Tables 2 and 3, Models 3–5). The interaction term between developing functional limitations and continuous participation in religious groups achieved statistical significance in models for respondents' personal growth (b =.58, p .05; Table 3, Model 4).
To interpret statistically significant interaction terms, we computed predicted scores for psychological well-being across groups of respondents who differed by whether they developed functional limitations and by whether they continuously participated in the voluntary group of interest. We used estimates from Table 2, Model 3, to interpret the statistically significant three-way interaction among gender, continuous participation in recreational groups, and functional limitations on depressive symptoms, and we used estimates from Table 3, Model 4, to interpret the statistically significant two-way interaction between continuous participation in religious groups and functional limitations on personal growth.
As Figure 3 displays, among men who did not continuously participate in recreational groups, developing functional limitations increased predicted levels of depressive symptoms by two thirds of a standard deviation. Among men who reported continuous participation in these groups, however, developing functional limitations was associated with a smaller increase in depressive symptoms over the 5-year period. Among women, regardless of their continuous participation in recreational groups, developing functional limitations was associated with increasing depressive symptoms from T1 to T2. These results indicate that continuous participation in recreational groups ameliorated the problematic association between functional limitations and depressive symptoms for men but not for women.
|
|
|
DISCUSSION |
|---|
|
TOPAbstractMethodsResultsDiscussionReferences |
|---|
; Taylor & Lynch, 2004). Findings also suggest, however, that deleterious changes in psychological well-being associated with developing functional limitations are not uniform across all adults who experience such limitations. Increases in depressive symptoms associated with developing functional limitations were less severe among men who continuously participated in recreational groups, and developing functional limitations was not associated with lower levels of personal growth among respondents who continuously participated in religious groups. Evidence for these associations held even when statistically controlling for a variety of sociodemographic factors associated with psychological well-being, such as respondents' education, race/ethnicity, and gender. Moreover, supplemental analyses (not shown), in which we separated respondents into three distinct categories of noncontinuous participation (no participation at T1 or T2, participation at T1 but not at T2, and participation at T2 but not at T1), indicated that the pattern of psychological well-being benefits associated with continuous participation in the face of functional decline remained in terms of contrasts with each of the noncontinuous participation subgroups. Although these results provide at least some evidence in support of the beneficial effects of continuous voluntary group participation on adults' psychological well-being, not all types of voluntary group participation demonstrated independent and/or interactive effects on both of the psychological well-being outcomes examined (i.e., depressive symptoms and personal growth). Additionally, although we observed few gender differences in the associations between continuous voluntary group participation and psychological well-being, results indicated some evidence for the idea that the psychological effects of voluntary group participation might differ for men and women. For example, whereas continuous participation in recreational groups protected men from smaller increases in depressive symptoms associated with developing functional limitations, we did not find this protective effect among women. Such complex patterns of significant and nonsignificant associations suggest that the ways and degrees to which particular types of continuous voluntary group participation affect certain aspects of psychological well-being are likely variable. Additional empirical and theory-based work is necessary to better understand the substantive significance of these complex patterns.
The results also indicated that the ways in which continuous participation in voluntary groups moderates the association between functional decline and psychological well-being are similarly complex. Whereas continuous participation in recreational groups served as a moderator for men's psychological well-being in terms of lessening the severity of the association between developing functional limitations and increasing depressive symptoms, continuous participation in religious groups moderated the association between developing functional limitations and lower levels of personal growth by eliminating this association altogether for both men and women. We speculate that because functional decline is a more potent risk factor for poorer psychological well-being in terms of more depressive symptoms than lower levels of personal growth, continuous participation in recreational groups might not be powerful enough to protect individuals against experiencing any increases in depressive symptoms upon developing functional limitations. On the other hand, because developing functional limitations is a less powerful risk factor for personal growth, and because continuous participation in voluntary groups is more robustly associated with this dimension of well-being, continuous participation in religious groups can altogether shield individuals from loss of personal growth following functional decline.
Both types of moderation processes (i.e., the amelioration and elimination of the association between functional decline and psychological well-being) are congruent with continuity theory. Continuity theory suggests that behavioral and mental strategies—such as maintaining participation in formal voluntary groups—can help individuals to achieve a consistent sense of self and their social worlds and thereby experience greater well-being when faced with life changes (Atchley, 1989). Alternative processes, however, might account for the statistically significant interactions found in this study. Although this study draws on the methodological strengths of a longitudinal, prospective design, it remains possible that respondents' maintenance of higher levels of psychological well-being in the face of declining functional health caused them to remain involved in voluntary groups, rather than their continuous participation helping them to maintain their psychological well-being. Additionally, although the associations between the severity of respondents' functional decline and their continuous participation in the voluntary groups across T1 and T2 were modest in size, the potential selection process from risk to protective factor suggests that continuous participation in voluntary groups among respondents with functional limitations could indicate a lesser degree of functional decline within this group. This lesser degree of functional decline might explain why respondents with continuous participation in voluntary groups reported better psychological well-being. Finally, although we investigated voluntary group participation in terms of three types of groups, the remaining heterogeneity of types of groups within each of these categories might mask the implications of particular types of voluntary group participation for psychological well-being.
Despite these limitations, this study's findings contribute to a better understanding of the psychosocial implications of functional limitations and factors comprising processes of risk and resilience in adulthood. Whereas results from previous studies have indicated that various types of voluntary group participation can promote individual well-being within the U.S. adult population at large (Menec, 2003; Morrow-Howell et al., 2003), the findings from this study demonstrate that voluntary group participation might also promote adults' psychological well-being by buffering them against losses in psychological well-being following the onset of functional limitations. Furthermore, the results suggest the advantages of voluntary group participation that extends across years of an individual's adulthood, thereby supporting the importance of efforts not only to engage, but also to maintain, adults' participation in voluntary groups. Future research on the precise mechanisms through which voluntary group participation promotes particular aspects of psychological well-being, as well as on factors that facilitate adults' ongoing participation, holds the promise of contributing to a better understanding of how psychological well-being can be maintained across adulthood—particularly in the face of aging-related challenges.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
Received for publication October 6, 2005. Accepted for publication March 15, 2006.
|
References |
|---|
|
TOPAbstractMethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
 |
|
 |
|
  N. C. Franklin and C. A. Tate Lifestyle and Successful Aging: An Overview American Journal of Lifestyle Medicine, January 1, 2009; 3(1): 6 - 11. [Abstract] [PDF]
|
|
| ||||||||||||||||||||||||||||||||
| HOME | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|---|
