Does Becoming A Volunteer Attenuate Loneliness Among Recently Widowed Older Adults?

Does Becoming A Volunteer Attenuate Loneliness Among Recently Widowed Older Adults? Abstract Objectives Loneliness is a significant public health concern, particularly for those who have lost a spouse through widowhood. This study examines whether becoming a volunteer at the time of widowhood is associated with reduction of these risks. Method A pooled sample of 5,882 married adults age 51+, drawn from the 2006–2014 waves of the Health and Retirement Study, was used to estimate regression models of the relationship between becoming widowed (relative to staying continuously married) and loneliness, and whether the associated loneliness of having lost a spouse is moderated by starting to volunteer (<2 hr, 2+ hr/week). Results Our results show that for those who become widowed, loneliness is significantly higher than those who stay continuously married. However, starting to volunteer 2+ hr per week is related to attenuated loneliness among the widowed such that widows who volunteer at that intensity have levels of loneliness similar to those of continuously married individuals volunteering at the same intensity. Discussion This study suggests higher intensity volunteering may be a particularly important pathway for alleviating loneliness among older adults who have recently become widowed. Results are discussed in light of theory, future research, and potential interventions. Health and Retirement Study, Productive engagement, Psychological health For many older Americans, loneliness is an unfortunate condition that affects everyday life (Queen, Stawski, Ryan, & Smith, 2014). An increasing body of knowledge on this topic suggests that being lonely is no trivial matter. Indeed, loneliness is associated with poorer physical health, more depressive symptoms, and even earlier mortality (Coyle & Dugan, 2012; Luo, Hawkley, Waite, & Cacioppo, 2012). In fact, the mortality risk associated with lack of strong social relationships is similar to smoking, approximately double that of obesity and quadruple that of exposure to air pollution (Holt-Lunstad, Smith, & Layton, 2010). Clearly, the consequences of loneliness have important implications for health and longevity. The extent to which older adults are lonely varies by a variety of factors. For instance, older adults who maintain social connections to other individuals in their lives and remain engaged in organizations tend to fair better than those who are socially disconnected from the everyday lives of others (Berkman & Syme, 1994). Although narrowing of social networks is typical in older adulthood and generally associated with better emotional well-being (English & Carstensen, 2014), they become especially crucial when immediate interpersonal relationships change, such as those who experience the loss of a spouse through widowhood (Antonucci et al., 2001). One way that these at-risk older adults may be able to obtain or maintain social connectedness despite loss of a spouse is to engage in the community by volunteering. To that end, in this article, we use a sample drawn from the Health and Retirement Study to address the following research questions: Does becoming a volunteer moderate the effect of becoming a widow on loneliness? And, does the “dose” of engagement matter? Widowhood and Loneliness In the United States, nearly one in five people over the age of 50 are lonely. Loneliness is the feeling resulting from a perceived gap between the number and quality of existing social relationships and one’s desired social relationships (Cacioppo, Hawkley, & Thisted, 2010), and is particularly problematic for older adults who experience changes in their social environment and in their ability to interact with others in a meaningful way (Smith, 2012; Taube, Jakobsson, Midlöv, & Kristensson, 2016). Becoming a widow, which is a common and difficult transition in later life, can lead to significant loneliness that is difficult to mitigate (van Baarsen, 2002). According to Socioemotional Selectivity Theory (Carstensen, Isaacowitz, & Charles, 1999) older adults are particularly motivated to maintain and foster emotionally gratifying relationships. Marriage is one of the most important relationships as people age, with cultivation of increasingly positive spousal interactions occurring in later life (Carstensen, Fung, & Charles, 2003). As a result, the loss of a spouse produces a significant gap in one’s social support network—a gap that is difficult to overcome—leading to intense and persistent feelings of loneliness following spousal death even when compensatory social support networks are available (van Baarsen, 2002). Two common interventions that have been used to address loneliness include: (a) increased social interaction and (b) provision of social support (Hawkley & Cacioppo, 2010). Losing a spouse to death—and losing the regular interactions that accompany being married—exposes people to feelings of loneliness because of the loss of social, emotional and instrumental support (Berkman & Syme, 1994; Greenfield & Russell, 2011; Lubben & Gironda, 2003; Pinquart, 2003; Shiovitz-Ezra & Ayalon, 2010). Volunteering as a Potential Intervention A large body of evidence points to a range of health benefits associated with engagement in social and productive activities in later life, particularly volunteering (for reviews see Anderson et al., 2014; Carr, Fried, & Rowe, 2015; Heaven et al., 2013; Jenkinson et al., 2013). Such engagement appears to influence health through psychosocial pathways like positive emotional exchange, associated lifestyle factors (increased physical, cognitive, and social activity), self-esteem and purpose in life, as well as through stress-buffering effects that moderate the influence of stressors on health (Cohen, Gottlieb, & Underwood, 2000; Matz-Costa, Carr, McNamara, & James, 2016). The social aspect of volunteering may be especially beneficial for protecting against loneliness among older adults. Meaningful social relationships are a key contributor to well-being among older adults. For example, social support plays an important role in offsetting loneliness among widows, such as through engagement with close friends, close relatives, and children (de Vries, Utz, Caserta, & Lund, 2014; Utz, Swenson, Caserta, Lund, & deVries, 2014) and has a particularly strong impact on those caregiving for a dying spouse with extensive disabilities or cognitive deficits (Dassel & Carr, 2016; Hawkley & Cacioppo, 2010). Volunteering facilitates enhanced emotional health through meaningful social interactions (see a description of the social benefit hypothesis provided by Li and Ferraro, 2005 explaining the related decrease in depression). Although institutional engagement like volunteering plays an important role in loneliness (Steffens, Cruwys, Haslam, Jetten, & Haslam, 2016) not all forms of institutional engagement seem to be sufficient to address the unique challenges that loneliness presents in late life. Working and religious attendance among younger adults are associated with lower levels of loneliness (Kuwert, Knaevelsrud, & Pietrzak, 2014; Rote, Hill, & Ellison, 2013). However, these activities are not related to reduced loneliness for older adults, perhaps because they do not consistently allow older people to feel that they are contributing in ways that are valued in a mutually beneficial way (Greenfield & Russell, 2011; Hawkley & Cacioppo, 2010; Kahn, Hessling, & Russell, 2003) rendering these activities less emotionally meaningful. Additionally, although small amounts or infrequent volunteer engagement may be advantageous for some aspects of health and well-being in later life, volunteering at low levels—for instance, volunteering at a soup kitchen five times a year—may not be sufficient to combat significant loneliness for those at particularly high risk. However, there is some evidence that higher levels of volunteer engagement, such as volunteering regularly for a few hours a week, may lead to more significant mental health benefits (Morrow-Howell, Hinterlong, Rozario, & Tang, 2003). Other factors have been shown to play a role in shaping loneliness. First, personality may orient certain individuals to be more likely to feel lonely (Hawkley & Cacioppo, 2010), which may also explain why personality is also associated with the decision to volunteer (King, Jackson, Morrow-Howell, & Oltmanns, 2015). Second, loneliness is strongly associated with physical and mental health, with lonely individuals experiencing more symptoms of depression, poorer self-rated health, and poorer cognitive performance (Cacioppo & Hawkley, 2009; Hawkley & Cacioppo, 2010; Shiovitz-Ezra & Ayalon, 2010). Finally, higher socioeconomic status is associated with decreased risk of loneliness (Hawkley et al., 2008) and increased likelihood of volunteering (Morrow-Howell, 2010). Although widowhood is associated with higher levels of loneliness (Sha’ked, 2014), it is not yet clear whether the effects of volunteering on loneliness are the same for people who are continually married as they are for those who experience the loss of a spouse. On the one hand, the number of hours one volunteers is associated with decreases in depressive symptoms and increases in self-efficacy among widows (Li, 2007). On the other hand, no research has explored whether higher levels of volunteering offer greater benefit to those at risk for loneliness. This is an important omission to the literature on the health and emotional benefits of volunteering. Unlike those who are never-married—who presumably never experience the social connectedness associated with being married—those who experience the death of a spouse experience the loss of interaction and support to which they have been accustom. Moreover, although divorce is traumatic for many people, the death of a spouse represents a unique form of loss, and has been referred to as “the most stressful of life events” (Carr, House, Wortman, Nesse, & Kessler, 2001). Thus, reintegration into social networks—in this case through volunteering—may be particularly important for people who have experienced the loss of a spouse to death. To that end, we developed the following hypothesis: Among those who become widowed, those who start a new volunteer role will experience less loneliness than those who do not, with more extensive volunteering associated with greater reduction in loneliness. Method Data The data for this study come from the Health and Retirement Study (HRS). The HRS, a longitudinal study of older Americans, began in 1992 sampling adults born between 1931 and 1941 (individuals initially ages 51 to 61) plus their spouses, oversampling for Hispanics, Blacks, and Florida residents. Follow-up occurred every 2 years, with additional cohorts added in subsequent waves. Since 1998 the HRS has been nationally representative of individuals ages 51 and older, with detailed information collected on demographics, housing situation, household and family, economic, employment, and health information. For the purposes of this study, we use data from the 2006–2014 HRS survey waves. We limited our analyses to these waves because the primary outcome variable is available only in the Leave Behind Questionnaire (LBQ) beginning in the 2006 study wave. The LBQ is collected from half the HRS respondents each wave with an alternating sample such that for every two-wave period, the entire HRS sample is surveyed. Our sample is pooled from respondents ages 51 or older who completed the LBQ in 2006, 2008, or 2010 (i.e., baseline) and again two study waves later when this item was measured again, 4 years later (i.e., Time 2). Because we are interested in the effect of volunteering on loneliness resulting from onset of widowhood, we include only individuals who are married at baseline, and who either remain continuously married or become widowed by Time 2. Our total pooled sample of 5,649 includes 5,057 individuals who remain continuously married, and 592 who become widowed. Measures Outcome variable The outcome variable for this study is a three-item composite measure of loneliness located in the LBQ of the HRS. Although an 11-item loneliness measure is available in the LBQ beginning in 2008, given the limited number of individuals who completed the LBQ questionnaire at two time points and also experienced the loss of a spouse, we used the measure of loneliness available consistently beginning in 2006, allowing us to identify a sufficient sample to address the research questions. This three-item measure has been shown to be a valid assessment of loneliness in the HRS sample (Hughes, Waite, Hawkley, & Cacioppo, 2004). This composite measure is the sum of three indicators, with response options of “1” (often), “2” (some of the time), and “3” (hardly ever or never). These indicators were how often do you feel: (1) you lack companionship; (2) left out; and (3) isolated from others. Each item was reverse coded such that higher numbers indicate greater loneliness. Individuals who did not respond to all three questions were coded missing. The outcome measure is the measure of loneliness measured at Time 2. Key independent variables Our first key independent variable was whether or not someone became widowed by Time 2 (relative to remaining continuously married). Our second key independent variable is a measure of change in volunteer engagement behavior between baseline and Time 2. The HRS measures formal volunteer engagement based on a series of questions about volunteer status, and how much time was spent volunteering for an organization during the previous year. Although it is not possible to measure volunteer hours continuously, these questions make it possible to identify changes in engagement according to three time categories: 1–99 hr per year, 100–199 hr per year, and 200 or more hr per year. We identify five volunteer groups based on these questions. First, we identified those who never volunteered (i.e., non-volunteers at baseline and at Time 2). Second, we identified individuals who continuously volunteered (i.e., were volunteers at both baseline and Time 2). Third, we identified individuals who stopped volunteering (i.e., were volunteers at baseline, but not at Time 2). Finally, we identified individuals who started volunteering. Within this group, we identified those who started volunteering 1–99 hr per year (i.e., non-volunteers at baseline and reported volunteering 1–99 hr in the last year at Time 2), and those who started volunteering 100+ hr per year (i.e., non-volunteers at baseline, and volunteering 100+ hr at Time 2). Due to the small number of individuals who started volunteering 200+ hr per year, we were unable to differentiate individuals in that group. Among those who started volunteering, some were likely volunteers prior to the study period. Due to the goals of this study, however, we are not concerned about individuals who have never reported volunteering. Rather, we focus on the potential for those not volunteering at the time of a transition into widowhood to experience benefits from starting a volunteer activity. We conducted sensitivity analyses of our findings using a lagged volunteer variable (i.e., controlling for volunteer status 2 years prior to baseline), with results remaining unchanged. Details of these analyses are available upon request. Control variables All control variables included in our models are measured during the baseline wave. We control for several demographic factors, including gender, race, years of education, wealth, and age which have been associated with volunteerism in past studies (Anderson et al., 2014). Female was a dichotomously coded indicator of gender. Race/ethnicity was measured with four indicator variables: Non-Hispanic White (reference group); Black; Hispanic; and Other Race. Education was measured in years (range from 0 to 17 years). Wealth was measured as total wealth in dollars, including total housing and nonhousing wealth (excluding pensions) at baseline. Age was measured as the age in years of the respondent at baseline (ranging from age 51 to 99). In addition to demographics, we also control for baseline health factors. Self-rated health is based on a five-point score ranging from “1” (poor health—reference group) to “5” (excellent health). Cognitive performance is based on tests of immediate and delayed recall, with a possible score ranging from 0 to 20. This is a test of episodic memory where individuals are read a list of 10 common words and asked to repeat as many of these words as possible right after being provided the words, and again after the interview continues on other topics 10–15 min later (Ofstedal, Fisher, & Herzog, 2005). The score is the total number of correct answers out of 20. Memory is a well-established metric for cognitive performance, which plays an important role in influencing capacity to engage in executive function tasks, and thus, volunteer engagement. Depressive symptoms is measured using a standard Center for Epidemiological Studies Depression measure, with scores ranging from 0 to 8, and higher numbers indicating more depressive symptoms. Although depression and loneliness are both commonly associated with caregiving and widowhood, recent research shows that they are distinct from one another (Hawkley & Cacioppo, 2010). The health status of a spouse is predictive of the intensity of care provided by the respondent, and plays an important role in shaping the well-being of a spouse in the period following onset of widowhood (Dassel & Carr, 2016; Dassel, Carr, & Vitaliano 2015). Most spouses help care for one another across the duration of a marriage, and it is, therefore, difficult to identify individuals providing care. As a result, we assess the potential impact of a spouse’s health at baseline on the surviving spouse by controlling for objective health measures. First, we control for spousal activities of daily living (ADL) limitations which is a count of the items in which he or she has difficulty: bathing, eating, dressing, walking across the room, or getting in and out of the bed or a chair. Second, we control for spousal memory disease diagnosis, which is a measure of whether or not a doctor has ever provided a diagnosis of any memory related disease (e.g., dementia, Alzheimer’s Disease). Recent research suggests personality plays an important role in shaping likelihood of seeking out volunteer roles and the benefits derived from them (King et al., 2015). Personality also is related to mental health (Cloninger & Zohar, 2011). For these reasons, we control for personality by including measures for the “Big 5” personality traits. These five characteristics (neuroticism, extroversion, openness to experience, conscientiousness, and agreeableness) are based on a series of 31 questions about individual characteristics (e.g., creative, responsible, careful, friendly). Respondents are asked how much they believe that characteristic is like them (coded as: 4 = a lot; 3 = some, 2 = a little, or 1 = not at all), based on Lachman and Weaver (1997). Inclusion of agreeableness did not contribute to improved model fit, did not affect variables of interest, and for the sake of parsimony, was excluded from final models. Details are available upon request. As noted earlier, social interaction and social support are related to decreased symptoms of loneliness (Hawkley & Cacioppo, 2010). Volunteering can be seen as one form of formal social engagement, and is the primary focus of the current article. However, in order to isolate the unique effect of taking on a new volunteer role on alleviating loneliness, we control for other forms of formal and informal social participation at baseline. First, we include three measures related to the number of individuals in respondents’ social networks. Specifically, individuals were asked how many friends, children, and other (non-spouse) family members they are close with. We included a continuous measure of the number of individuals respondents identified as close to them in each of these categories. Second, we controlled for church attendance, identifying individuals who indicated attending church once a month or more. Finally, we identified individuals who are employed and work 1 or more hr per week. Our sample is drawn from individuals who completed the entire LBQ questionnaire at both baseline and at Time 2 (N = 6,217; i.e., 2006 & 2010, 2008 & 2012, or 2010 & 2014). Among these individuals, 150 (2.4%) had missing data on the health of their spouse at baseline, and were excluded from the analyses. Among the remaining (N = 6,067), 115 (1.8%) did not complete the cognitive and mental health questions at baseline, and 19 (0.3%) did not complete the volunteer questions at baseline and at Time 2, and an additional 51 (0.8%) had missing data on education or race. Our total sample includes 5,882 individuals, including 667 who become widowed between baseline and Time 2, and 5,215 individuals who remain continuously married. Analysis Plan To test our research question and hypothesis, Ordinary Least Squares (OLS) regression was used to assess the impact of starting volunteering at a level of 1–99 hr per year or 100+ hr per year (relative to not starting volunteering or no volunteering) on loneliness at Time 2 for those who are continuously married versus those who become widowed. Each person had one observation per variable in the final data file and, thus, multilevel modeling was not necessary (To adjust for the potential interdependencies in the data due to continuously married individuals living in the same household, we conducted our analyses clustered by household, which showed very similar results. Results are available upon request.). First, we entered the widowhood variable into the model while controlling for the raw loneliness score at baseline as well as demographic, health, and personality factors at baseline. Next, we added the volunteer engagement behavior variables to the model while controlling for other forms of social engagement (baseline; Utz, Carr, Nesse, & Wortman, 2002). Finally, we added interaction terms between widowhood and volunteer behavior to examine whether starting volunteering at a level of 1–99 hr per year or 100+ hr per year moderates the relationship between widowhood and change in loneliness. Moderation effects were further examined by calculating marginal effects, holding all other factors at the mean. Results Descriptive Information and Bivariate Analysis Table 1 provides the descriptive data for the sample, which shows that the average loneliness score of the sample at Time 2 is 4.1 (SD = 1.5). About 48% of the sample were non-volunteers, 12.1% stopped volunteering, and nearly one-third (32.2%) volunteered at both baseline and Time 2. Almost 8% of the sample were non-volunteers at baseline but started volunteering at Time 2: 6.3% of the sample started volunteering at an intensity of less than 100 hr per year, and about 1.5% started volunteering at an intensity of 100 hr per year. Table 1. Descriptive Characteristics of Samples   All (N = 5,882)  Became widowed (N = 667)  Continuously married (N = 5,215)  Variable  Meana (SD)  Meana (SD)  Meana (SD)  Loneliness score (wave t+2)  4.142 (1.474)  4.954 (1.666)***  4.038 (1.415)  Loneliness score (wave t)  4.055 (1.418)  4.231 (1.484)***  4.033 (1.408)  Became widowed  0.113      Volunteer behavior   Started Vol 1–99 hr/year  0.063  0.084***  0.060   Started Vol 100+ hr/year  0.015  0.033*  0.013   Continuous volunteer  0.322  0.232***  0.334   Stopped volunteering  0.121  0.133  0.119   Never volunteered  0.479  0.517*  0.474  Female  0.532  0.735***  0.506  Race/ethnicity   White  0.839  0.841***  0.839   Black  0.071  0.090***  0.068   Hispanic  0.069  0.055***  0.071   Other race  0.021  0.013*  0.022  Years of education  13.134 (2.845)  12.648 (2.808)***  13.197 (2.844)  Household wealth (in $100,000s)  5.253 (10.013)  5.211 (15.563)  5.259 (9.063)  Age  67.912 (8.410)  71.510 (8.507)***  67.452 (8.286)  Self-rated health  2.634 (1.003)  2.706 (1.018)*  2.625 (1.001)  Cognitive score  10.278 (3.083)  9.802 (3.090)***  10.339 (3.077)  Depressive symptoms  0.975 (1.604)  1.282 (1.799)***  0.935 (1.573)  Spouse ADL limitations  0.238 (1.604)  0.867 (1.508)***  0.157 (1.573)  Spouse memory disease diag  0.028  0.111***  0.017  Neuroticism  1.958 (0.574)  2.005 (0.571)*  1.952 (0.574)  Extroversion  3.218 (0.548)  3.252 (0.527)  3.213 (0.550)  Openness to experience  2.954 (0.537)  2.927 (0.549)  2.957 (0.535)  Conscientiousness  3.364 (0.430)  3.372 (0.549)  3.363 (0.428)  Number of close friends  4.251 (4.980)  3.961 (4.389)  4.288 (5.049)  Number of close relatives  3.460 (5.126)  3.387 (4.205)  3.469 (5.232)  Number of close children  2.463 (2.601)  2.625 (2.630)  2.442 (2.596)  Attends church monthly  0.572  0.588  0.570  Works  0.374  0.238***  0.391    All (N = 5,882)  Became widowed (N = 667)  Continuously married (N = 5,215)  Variable  Meana (SD)  Meana (SD)  Meana (SD)  Loneliness score (wave t+2)  4.142 (1.474)  4.954 (1.666)***  4.038 (1.415)  Loneliness score (wave t)  4.055 (1.418)  4.231 (1.484)***  4.033 (1.408)  Became widowed  0.113      Volunteer behavior   Started Vol 1–99 hr/year  0.063  0.084***  0.060   Started Vol 100+ hr/year  0.015  0.033*  0.013   Continuous volunteer  0.322  0.232***  0.334   Stopped volunteering  0.121  0.133  0.119   Never volunteered  0.479  0.517*  0.474  Female  0.532  0.735***  0.506  Race/ethnicity   White  0.839  0.841***  0.839   Black  0.071  0.090***  0.068   Hispanic  0.069  0.055***  0.071   Other race  0.021  0.013*  0.022  Years of education  13.134 (2.845)  12.648 (2.808)***  13.197 (2.844)  Household wealth (in $100,000s)  5.253 (10.013)  5.211 (15.563)  5.259 (9.063)  Age  67.912 (8.410)  71.510 (8.507)***  67.452 (8.286)  Self-rated health  2.634 (1.003)  2.706 (1.018)*  2.625 (1.001)  Cognitive score  10.278 (3.083)  9.802 (3.090)***  10.339 (3.077)  Depressive symptoms  0.975 (1.604)  1.282 (1.799)***  0.935 (1.573)  Spouse ADL limitations  0.238 (1.604)  0.867 (1.508)***  0.157 (1.573)  Spouse memory disease diag  0.028  0.111***  0.017  Neuroticism  1.958 (0.574)  2.005 (0.571)*  1.952 (0.574)  Extroversion  3.218 (0.548)  3.252 (0.527)  3.213 (0.550)  Openness to experience  2.954 (0.537)  2.927 (0.549)  2.957 (0.535)  Conscientiousness  3.364 (0.430)  3.372 (0.549)  3.363 (0.428)  Number of close friends  4.251 (4.980)  3.961 (4.389)  4.288 (5.049)  Number of close relatives  3.460 (5.126)  3.387 (4.205)  3.469 (5.232)  Number of close children  2.463 (2.601)  2.625 (2.630)  2.442 (2.596)  Attends church monthly  0.572  0.588  0.570  Works  0.374  0.238***  0.391  Notes: ADL = activities of daily living. aMeans are provided for continuous variables, proportions for dichotomous variables. Significant differences between widowed and continuously married individuals is noted in the following way: *p < .05. ***p < .001. Significance is determined using chi-square tests for dichotomous variables, and t tests for continuous variables. View Large We performed bivariate analyses (t tests and chi-square tests) to identify significant differences in characteristics of those who became widowed relative to those who remained married. Results indicate that the average loneliness score varies by marital status. Those who became widowed had a significantly greater loneliness at baseline and at Time 2 relative to those who were married (p < .001). With respect to volunteering, engagement varies by marital status such that those who became widowed were more likely to start volunteering than those who stayed married, particularly at the highest intensity level (p < .001). Those who became widows were also less likely to be continuous volunteers (p < .001) and more likely to be non-volunteers (p < .05). In this sample, those who went on to become widowed were, at baseline: more likely to be women (p < .001), Black (p < .05) and not Hispanic (p < .001) or another race (p < .005), have less education (p < .001), older (p < .001), have poorer self-rated health (p < .05), lower cognitive performance (p < .001); and more depressive symptoms (p < .001). They are also more likely to have a more disabled spouse (p < .001), who is more likely to have been diagnosed with a memory disorder (p < .001). They have a higher neuroticism score, and are less likely to be working (p < .001) than their continuously married counterparts. OLS Regression and Moderation Analysis Table 2 shows the results from OLS regression models predicting loneliness score at Time 2. In Model 1, we examine the direct association between change in marital status and change in loneliness, adjusting for loneliness at baseline, demographic factors, health and personality. Results indicate that widowhood is related to significantly higher loneliness at Time 2 (β = 0.505, p < .001). In Model 2, we assess the direct relation between change in volunteer behavior and loneliness, adjusting for other forms of social engagement and all factors from Model 1. Results show that volunteer behavior was not significantly related to loneliness score at Time 2. Further, when controlling for volunteer engagement, the direct effect of marital status on loneliness did not change. Table 2. OLS Regression Models Predicting Change in Loneliness (N = 5,882)   Model 1  Model 2  Model 3    Coeff.  SE  Coeff.  SE  Coeff.  SE  Loneliness score (wave t)  0.465  0.013***  0.462  0.013***  0.462  0.013***  Became widoweda  0.789  0.053***  0.785  0.053***  0.809  0.055***  Volunteer behaviorb   1–99 hr/year      −0.010  0.067  −0.007  0.073   100+ hr/year      −0.067  0.129  0.078  0.147   Continuous volunteer      −0.055  0.042  −0.053  0.043   Stopped volunteering      0.018  0.052  0.022  0.055  Number of close friends      −0.007  0.004  −0.007  0.004  Number of close relatives      −0.001  0.003  −0.001  0.003  Number of close children      −0.005  0.007  −0.005  0.007  Attends church monthly      0.060  0.036  0.059  0.036  Works      0.008  0.037  0.009  0.037  Start Vol 1–99 hr × Became widowed          −0.022  0.182  Start Vol 100+ hr × Became widowed          −0.608  0.299*  Continuous volunteer × Became widowed          −0.013  0.122  Stopped volunteering × Became widowed          −0.027  0.153  Femalec  0.133  0.034***  0.130  0.034***  0.130  0.034***  Raced   White               Black  −0.035  0.063  −0.043  0.064  −0.041  0.064   Hispanic  −0.211  0.109  −0.217  0.109*  −0.214  0.109*   Other race  −0.269  0.067***  −0.277  0.067***  −0.277  0.067***  Years of education  0.009  0.007  0.010  0.007  0.010  0.007  Household wealth  0.000  0.000*  0.000  0.000*  0.000  0.000*  Age  0.001  0.002  0.001  0.002  0.001  0.002  Self-rated health  0.078  0.018***  0.078  0.018***  0.077  0.018***  Cognitive score  −0.009  0.006  −0.008  0.006  −0.008  0.006  Depressive symptoms  0.056  0.012***  0.056  0.012***  0.055  0.012***  Spouse ADL limitations  −0.034  0.022  −0.033  0.022  −0.032  0.022  Spouse memory disease diagnosed  −0.051  0.099  −0.053  0.100  −0.040  0.100  Neuroticism  0.207  0.032***  0.206  0.032***  0.207  0.032***  Extroversion  −0.186  0.036***  −0.178  0.036***  −0.178  0.036***  Openness to experience  0.154  0.037***  0.163  0.038***  0.165  0.038***  Conscientiousness  −0.114  0.042**  −0.120  0.042**  −0.121  0.042**  Constant  1.932  0.268***  1.911  0.278***  1.903  0.278***  Log likelihood (model)  −9397.838  −9392.248  −9390.174    Model 1  Model 2  Model 3    Coeff.  SE  Coeff.  SE  Coeff.  SE  Loneliness score (wave t)  0.465  0.013***  0.462  0.013***  0.462  0.013***  Became widoweda  0.789  0.053***  0.785  0.053***  0.809  0.055***  Volunteer behaviorb   1–99 hr/year      −0.010  0.067  −0.007  0.073   100+ hr/year      −0.067  0.129  0.078  0.147   Continuous volunteer      −0.055  0.042  −0.053  0.043   Stopped volunteering      0.018  0.052  0.022  0.055  Number of close friends      −0.007  0.004  −0.007  0.004  Number of close relatives      −0.001  0.003  −0.001  0.003  Number of close children      −0.005  0.007  −0.005  0.007  Attends church monthly      0.060  0.036  0.059  0.036  Works      0.008  0.037  0.009  0.037  Start Vol 1–99 hr × Became widowed          −0.022  0.182  Start Vol 100+ hr × Became widowed          −0.608  0.299*  Continuous volunteer × Became widowed          −0.013  0.122  Stopped volunteering × Became widowed          −0.027  0.153  Femalec  0.133  0.034***  0.130  0.034***  0.130  0.034***  Raced   White               Black  −0.035  0.063  −0.043  0.064  −0.041  0.064   Hispanic  −0.211  0.109  −0.217  0.109*  −0.214  0.109*   Other race  −0.269  0.067***  −0.277  0.067***  −0.277  0.067***  Years of education  0.009  0.007  0.010  0.007  0.010  0.007  Household wealth  0.000  0.000*  0.000  0.000*  0.000  0.000*  Age  0.001  0.002  0.001  0.002  0.001  0.002  Self-rated health  0.078  0.018***  0.078  0.018***  0.077  0.018***  Cognitive score  −0.009  0.006  −0.008  0.006  −0.008  0.006  Depressive symptoms  0.056  0.012***  0.056  0.012***  0.055  0.012***  Spouse ADL limitations  −0.034  0.022  −0.033  0.022  −0.032  0.022  Spouse memory disease diagnosed  −0.051  0.099  −0.053  0.100  −0.040  0.100  Neuroticism  0.207  0.032***  0.206  0.032***  0.207  0.032***  Extroversion  −0.186  0.036***  −0.178  0.036***  −0.178  0.036***  Openness to experience  0.154  0.037***  0.163  0.038***  0.165  0.038***  Conscientiousness  −0.114  0.042**  −0.120  0.042**  −0.121  0.042**  Constant  1.932  0.268***  1.911  0.278***  1.903  0.278***  Log likelihood (model)  −9397.838  −9392.248  −9390.174  Notes: ADL = activities of daily living. Italic values indicate the standard errors (SE). Reference groups include: aContinuously married individuals; bNever volunteerd; cMale; dNon-Hispanic White. *p < .05. **p < .01. ***p < .001. View Large Finally, in Model 3, we examine the effect of new engagement in volunteering at two intensity levels on loneliness by assessing the interaction effect between volunteer behaviors and onset of widowhood. Results show a significant interaction between widowhood and starting volunteering at 100+ hr per year. Specifically, the associated benefits of volunteering 100+ hr per year on loneliness is dependent on whether an individual became widowed (p < .05). A post-estimation Wald significance test further confirms that the moderation of volunteer behavior and becoming widowed is significant (p < .001). To more carefully examine these relationships, we calculate the marginal effects (i.e., predicted probabilities) for loneliness at Time 2 (holding all other factors at the mean) in relation to volunteer intensity for those who became widows versus those who stayed married. Figure 1 shows that starting volunteering is related to reduced loneliness among those who became widows, but only at the highest intensity level (i.e., volunteering 100+ hr per year). Specifically, relative to those who were continuously married, those who became widowed had a statistically higher predicted loneliness score if they never volunteered (4.88 compared to 4.07), stopped volunteering (4.87 compared to 4.09), were continuously volunteers (4.81 compared to 4.01), or started volunteering, but only 1–99 hr per year (4.81 compared to 4.06). Volunteering 100+ hr per year moderates the effect of widowhood on loneliness, and these volunteers had a similar predicted loneliness score (4.34 compared to 4.15) to those who were continuously married. What this means is that more intense levels of loneliness persist for new widows relative to continuously married individuals, but loneliness is lower when widows start volunteering approximately 2 hr per week, on average, or more. New widows who start volunteering at that threshold experience similar levels of loneliness as continuously married individuals. Figure 1. View largeDownload slide Predicted loneliness score by volunteer group and marital status. Note: Error bars are the 95% confidence intervals. Figure 1. View largeDownload slide Predicted loneliness score by volunteer group and marital status. Note: Error bars are the 95% confidence intervals. Discussion Becoming a widow is one of the most difficult transitions that people face in later life. Widowhood is associated with significant increases in loneliness, leaving widows at risk for additional health consequences, including depression, physical health decline, and increased mortality risk. In this article, we examined the relation between starting volunteering and loneliness among older adults who became widowed over a 4-year period relative to those who remained continuously married. We discovered that volunteering moderates the negative effects of loneliness for those who become widowed, but only in relation to engagement in 2 or more hr per week, on average. Although our research cannot explain the mechanisms by which volunteering is beneficial, there are several plausible reasons why volunteering on a regular basis might have the potential to significantly alleviate loneliness in new widows. Previous research shows that loneliness can be addressed through increases in social interaction and increases in social support (Hawkley & Cacioppo, 2010). Although volunteering comes in many forms, some likely requiring more social interaction than others, volunteering almost always involves social engagement with others. Individuals who have recently lost their primary form of social support (i.e., their spouse) have the most to benefit from new sources of social engagement. Additionally, beyond the social aspect of volunteering, it may be that engaging at a significant intensity in work that has a social purpose—as volunteer work often does—bolsters the kinds of personal resources (e.g., health behaviors, self-esteem, purpose in life, sense of control) that are needed to manage the grief and loss associated with widowhood (Matz-Costa et al., 2016). Starting to volunteer on a regular basis may help in establishing an integrated sense of self that compensates for the loss of this salient life role (i.e., identity as married) and social resource and protects against a sense of loneliness (Hobfoll & Wells, 1998). However, it is important to note that not all older adults—particularly widows—have the same access or resources to volunteer. In general, older adults are more likely to volunteer if they are highly educated, have higher incomes, work part time, are married, and have a spouse who also volunteers (Bureau of Labor Statistics, U.S. Department of Labor, 2013; Rotolo & Wilson, 2006). Widowhood is often associated with significant grief and depressive symptoms, including loss of interest in activity (Bonanno, 2004); a substantial loss of income and often times poverty (poverty rates among elderly female widows is 3–4 times higher than rates among elderly married women (McGarry & Schoeni, 2005) and increased household responsibilities, all of which are significant barriers to starting a new volunteer role (McBride, 2007). Future research should explore disparities in economic and health resources as well as other structural barriers (e.g., lack of knowledge about volunteer opportunities, lack of skills, time constraints) among widows that may restrict access to and availability of meaningful volunteer roles. Findings from this study need to be understood in light of its limitations. First, we had to limit our sample to those who completed both the HRS core and the LBQ questionnaires during both time points in which loneliness was measured. As a result, the final sample size for this analysis is smaller than what could be obtained if using the HRS core survey data alone. This decreased the potential power of our analyses. Second, we are unable to draw conclusions regarding the directionality of these effects because, despite our longitudinal design, it was not possible to control for all other potential factors that may differentiate continuously married individuals from those who became widowed. Thus, it is possible that the moderation effect of starting volunteering is due to widows who are lonelier choosing to volunteer to off-set these feelings rather than the act of volunteering contributing to decreased loneliness. Lastly, we were unable to take into account one’s history of volunteer engagement prior to baseline in this study. It is possible that some non-volunteers had previously engaged in volunteer activities prior to widowhood, lost this role due to their caregiving responsibilities, and returned following onset of widowhood. This explanation is in line with Utz and colleagues (2002) who point out that “older adults, particularly those experiencing a devastating or stressful loss such as widowhood, tend to rely on the convoy of lifelong social relations from whom they derive support, self-definition, and a sense of stability and continuity” and that widowed persons, “in both their intentions and their behaviors, prefer to maintain the relationships and activities that have been most meaningful and rewarding throughout the life course” (p. 532). Although the goals of this study focus on whether starting a new volunteering role relates to lower levels of loneliness following widowhood regardless of previous volunteer history, future research might examine transitions into and out of volunteer work, its co-occurrence with caregiving and its effect on loneliness to determine if these effects vary based on history of volunteer engagement. Despite these limitations, the results from this study provide encouraging support for the potential of volunteer engagement as an intervention for alleviating loneliness among widows. There are few volunteer programs designed to target the widows, and interventions designed to facilitate starting a new volunteer role at this critical life juncture could play an important role in delaying onset of subsequent health consequences associated with spousal loss. In addition, given the numerous benefits associated with engaging in volunteering in general, and with 2 hr or more per week in particular (Windsor, Anstey, & Rodgers, 2008), individuals who start a new volunteer role and continue to remain engaged over an extended period may experience health and well-being benefits that facilitate more successful aging. In conclusion, loneliness is a significant public health concern for individuals who become widowed. Formal volunteering, however, may offer a means by which to cultivate personal resources at the transition into widowhood. The findings of the current study suggest that interventions that encourage volunteering, particularly higher intensity volunteering, among those who become widowed may be effective in helping to mitigate loneliness. Funding No funding was received in relation to this research project. Acknowledgments D. C. Carr planned the study, performed all statistical analysis and reporting of results, wrote the abstract, contributed to development of the background and discussion sections, and supervised the writing of the article. B. L. Kail provided methodological consultation, contributed to development of the methodology section, and contributed to introduction and literature review. C. Matz-Costa contributed to the introduction and discussion sections, and assisted with formatting and references. Y. Z. Shavit contributed to the introduction section, and consulted on development of measures. Conflict of Interest None of the authors associated with publication have a conflict of interest in relation to this research project. References Anderson, N. D., Damianakis, T., Kröger, E., Wagner, L. M., Dawson, D. R., Binns, M. A.,… Cook, S. L.; BRAVO Team. ( 2014). The benefits associated with volunteering among seniors: A critical review and recommendations for future research. Psychological Bulletin , 140, 1505– 1533. doi: 10.1037/a0037610 Google Scholar CrossRef Search ADS PubMed  Antonucci, T. C., Lansford, J. E., Schaberg, L., Baltes, M., Takahashi, K., Dartigues, J. F.,… Fuhrer, R. ( 2001). Widowhood and illness: A comparison of social network characteristics in France, Germany, Japan, and the United States. Psychology and Aging , 16, 655– 665. doi: 10.1037/0882-7974.16.4.655 Google Scholar CrossRef Search ADS PubMed  Berkman, L. F., & Syme, S. L. ( 1994). Social networks, host resistance, and mortality: A nine-year follow-up study of Alameda County residents. In A. Steptoe & J. Wardle (Eds.), Psychosocial processes and health  (pp. 43– 67). Cambridge, UK: Cambridge University Press. Retrieved from http://ebooks.cambridge.org/ref/id/CBO9780511759048A010 Bonanno, G. A. ( 2004). Loss, trauma, and human resilience: Have we underestimated the human capacity to thrive after extremely aversive events? The American Psychologist , 59, 20– 28. doi: 10.1037/0003-066X.59.1.20 Google Scholar CrossRef Search ADS PubMed  Bureau of Labor Statistics, U.S. Department of Labor. ( 2013). Volunteering in the United States, 2013. Retrieved from http://www.bls.gov/news.release/volun.nr0.htm Cacioppo, J. T., & Hawkley, L. C. ( 2009). Perceived social isolation and cognition. Trends in Cognitive Sciences , 13, 447– 454. doi: 10.1016/j.tics.2009.06.005 Google Scholar CrossRef Search ADS PubMed  Cacioppo, J. T., Hawkley, L. C., & Thisted, R. A. ( 2010). Perceived social isolation makes me sad: 5-year cross-lagged analyses of loneliness and depressive symptomatology in the Chicago Health, Aging, and Social Relations Study. Psychology and Aging , 25, 453– 463. doi: 10.1037/a0017216 Google Scholar CrossRef Search ADS PubMed  Carr, D., House, J. S., Wortman, C., Nesse, R., & Kessler, R. C. ( 2001). Psychological adjustment to sudden and anticipated spousal loss among older widowed persons. The Journals of Gerontology. Series B, Psychological Sciences and Social Sciences , 56, S237– S248. doi: 10.1093/geronb/56.4.S237 Google Scholar CrossRef Search ADS PubMed  Carr, D. C., Fried, L. P., & Rowe, J. W. ( 2015). Productivity and engagement in an aging America: The role of volunteerism. Daedalus , 144, 55– 67. doi: 10.1162/DAED_a_00330 Google Scholar CrossRef Search ADS   Carstensen, L., Fung, H., & Charles, S. T. ( 2003). Socioemotional selectivity theory and the regulation of emotion in the second half of life. Motivation and Emotion , 27, 103– 123. doi: 10.1023/A:1024569803230 Google Scholar CrossRef Search ADS   Carstensen, L. L., Isaacowitz, D. M., & Charles, S. T. ( 1999). Taking time seriously. A theory of socioemotional selectivity. The American Psychologist , 54, 165– 181. doi: 10.1037/0003-066X.54.3.165 Google Scholar CrossRef Search ADS PubMed  Cloninger, C. R., & Zohar, A. H. ( 2011). Personality and the perception of health and happiness. Journal of Affective Disorders , 128, 24– 32. doi: 10.1016/j.jad.2010.06.012 Google Scholar CrossRef Search ADS PubMed  Cohen, S., Gottlieb, B., & Underwood, L. ( 2000). Social relationships and health. In S. Cohen, L. Underwood, & B. Gottlieb (Eds.), Measuring and intervening in social support  (pp. 3–25). New York, NY: Oxford University Press. Coyle, C. E., & Dugan, E. ( 2012). Social isolation, loneliness and health among older adults. Journal of Aging and Health , 24( 8), 1346– 1363. doi: 10.1177/0898264312460275 Google Scholar CrossRef Search ADS PubMed  Dassel, K. B., & Carr, D. C. ( 2016). Does dementia caregiving accelerate frailty? Findings from the Health and Retirement Study. The Gerontologist , 56, 444– 450. doi: 10.1093/geront/gnu078 Google Scholar CrossRef Search ADS PubMed  Dassel, K. B., Carr, D. C., & Vitaliano, P. ( 2015). Does caring for a spouse with dementia accelerate cognitive decline? Findings from the Health and Retirement Study. The Gerontologist . Advance online publication. doi: 10.1093/geront/39.2.177 de Vries, B., Utz, R., Caserta, M., & Lund, D. ( 2014). Friend and family contact and support in early widowhood. The Journals of Gerontology. Series B, Psychological Sciences and Social Sciences , 69, 75– 84. doi: 10.1093/geronb/gbt078 Google Scholar CrossRef Search ADS PubMed  English, T., & Carstensen, L. L. ( 2014). Selective narrowing of social networks across adulthood is associated with improved emotional experience in daily life. International Journal of Behavioral Development , 38, 195– 202. doi: 10.1177/0165025413515404 Google Scholar CrossRef Search ADS PubMed  Greenfield, E. A., & Russell, D. ( 2011). Identifying living arrangements that heighten risk for loneliness in later life: Evidence from the U.S. National Social Life, Health, and Aging Project. Journal of Applied Gerontology , 30, 524– 534. doi: 10.1177/0733464810364985 Google Scholar CrossRef Search ADS   Hawkley, L. C., & Cacioppo, J. T. ( 2010). Loneliness matters: A theoretical and empirical review of consequences and mechanisms. Annals of Behavioral Medicine , 40, 218– 227. doi: 10.1007/s12160-010-9210-8 Google Scholar CrossRef Search ADS PubMed  Hawkley, L. C., Hughes, M. E., Waite, L. J., Masi, C. M., Thisted, R. A., & Cacioppo, J. T. ( 2008). From social structural factors to perceptions of relationship quality and loneliness: The Chicago health, aging, and social relations study. The Journals of Gerontology. Series B, Psychological Sciences and Social Sciences , 63, S375– S384. doi: 10.1093/geronb/63.6.S375 Google Scholar CrossRef Search ADS PubMed  Heaven, B., Brown, L. J., White, M., Errington, L., Mathers, J. C., & Moffatt, S. ( 2013). Supporting well-being in retirement through meaningful social roles: Systematic review of intervention studies. The Milbank Quarterly , 91, 222– 287. doi: 10.1111/milq.12013 Google Scholar CrossRef Search ADS PubMed  Hobfoll, S. E., & Wells, J. D. ( 1998). Conservation of resources, stress, and aging: Why do some slide and some spring? In J. Lomranz (Ed.), The handbook of aging and mental health: An integrative approach  (pp. 121– 134). New York & London: Plenum Press. Google Scholar CrossRef Search ADS   Holt-Lunstad, J., Smith, T. B., & Layton, J. B. ( 2010). Social relationships and mortality risk: A meta-analytic review. PLoS Medicine , 7, e1000316. doi: 10.1371/journal.pmed.1000316 Google Scholar CrossRef Search ADS PubMed  Hughes, M. E., Waite, L. J., Hawkley, L. C., & Cacioppo, J. T. ( 2004). A short scale for measuring loneliness in large surveys: Results from two population-based studies. Research on Aging , 26, 655– 672. doi: 10.1177/0164027504268574 Google Scholar CrossRef Search ADS PubMed  Jenkinson, C. E., Dickens, A. P., Jones, K., Thompson-Coon, J., Taylor, R. S., Rogers, M.,… Richards, S. H. ( 2013). Is volunteering a public health intervention? A systematic review and meta-analysis of the health and survival of volunteers. BMC Public Health , 13, 773. doi: 10.1186/1471-2458-13-773 Google Scholar CrossRef Search ADS PubMed  Kahn, J. H., Hessling, R. M., & Russell, D. W. ( 2003). Social support, health, and well-being among the elderly: What is the role of negative affectivity? Personality and Individual Differences , 35, 5– 17. doi: 10.1016/S0191-8869(02)00135-6 Google Scholar CrossRef Search ADS   King, H. R., Jackson, J. J., Morrow-Howell, N., & Oltmanns, T. F. ( 2015). Personality accounts for the connection between volunteering and health. The Journals of Gerontology. Series B, Psychological Sciences and Social Sciences , 70, 691– 697. doi: 10.1093/geronb/gbu012 Google Scholar CrossRef Search ADS PubMed  Kuwert, P., Knaevelsrud, C., & Pietrzak, R. H. ( 2014). Loneliness among older veterans in the United States: Results from the National Health and Resilience in Veterans Study. The American Journal of Geriatric Psychiatry , 22( 6), 564– 569. doi: 10.1016/ j.jagp.2013.02.013 Google Scholar CrossRef Search ADS PubMed  Lachman, M., & Weaver, S. ( 1997). Midlife devlopment inventory (MIDI) personality sacales: scale construction and scoring. Brandeis University. Retrieved from http://www.brandeis.edu/projects/lifesapn/scales.html Li, Y. ( 2007). Recovering from spousal bereavement in later life: Does volunteer participation play a role? The Journals of Gerontology. Series B, Psychological Sciences and Social Sciences , 62, S257– S266. doi: 10.1093/geronb/62.4.S257 Google Scholar CrossRef Search ADS PubMed  Li, Y., & Ferraro, K. F. ( 2005). Volunteering and depression in later life: Social benefit or selection processes? Journal of Health and Social Behavior , 46, 68– 84. doi: 10.1177/002214650504600106 Google Scholar CrossRef Search ADS PubMed  Lubben, J., & Gironda, M. ( 2003). Centrality of social ties to the health and well-being of older adults. In B. Berkman & L. Harootyan (Eds.), Social work and health care in an aging world: Informing education, policy, practice, and research  (pp. 319– 350). New York, NY: Springer Press. Luo, Y., Hawkley, L. C., Waite, L. J., & Cacioppo, J. T. ( 2012). Loneliness, health, and mortality in old age: A national longitudinal study. Social Science & Medicine , 74( 6), 907– 914. doi: 10.1016/j.socscimed.2011.11.028 Google Scholar CrossRef Search ADS   Matz-Costa, C., Carr, D. C., McNamara, T. K., & James, J. B. ( 2016). Physical, cognitive, social, and emotional mediators of activity involvement and health in later life. Research on Aging , 38, 791– 815. doi: 10.1177/0164027515606182 Google Scholar CrossRef Search ADS PubMed  McBride, A. ( 2007). Civic engagement, older adults, and inclusion. Generations , 30, 66– 71. McGarry, K., & Schoeni, R. F. ( 2005). Medicare gaps and widow poverty. Social Security Bulletin , 66, 58– 74. Google Scholar PubMed  Morrow-Howell, N. ( 2010). Volunteering in later life: Research frontiers. The Journals of Gerontology. Series B, Psychological Sciences and Social Sciences , 65, 461– 469. doi: 10.1093/geronb/gbq024 Google Scholar CrossRef Search ADS PubMed  Morrow-Howell, N., Hinterlong, J., Rozario, P. A., & Tang, F. ( 2003). Effects of volunteering on the well-being of older adults. The Journals of Gerontology. Series B, Psychological Sciences and Social Sciences , 58, S137– S145. doi: 10.1093/geronb/58.3.S137 Google Scholar CrossRef Search ADS PubMed  Ofstedal, M. B., Fisher, G. G., & Herzog, A. R. ( 2005). HRS/AHEAD documentation report: Documentation of cognitive functioning measures in the Health and Retirement Study . Ann Arbor, MI: HRS Health Working Group, Survey Research Center, University of Michigan. Retrieved from http://hrsonline.isr.umich.edu/sitedocs/userg/dr-006.pdf Pinquart, M. ( 2003). Loneliness in married, widowed, divorced, and never-married older adults. Journal of Social and Personal Relationships , 20, 31– 53. Google Scholar CrossRef Search ADS   Queen, T. L., Stawski, R. S., Ryan, L. H., & Smith, J. ( 2014). Loneliness in a day: Activity engagement, time alone, and experienced emotions. Psychology and Aging , 29( 2), 297– 305. doi: 10.1037/a0036889 Google Scholar CrossRef Search ADS PubMed  Rote, S., Hill, T. D., & Ellison, C. G. ( 2013). Religious attendance and loneliness in later life. The Gerontologist , 53( 1), 39– 50. doi: 10.1093/geront/gns063 Google Scholar CrossRef Search ADS PubMed  Rotolo, T., & Wilson, J. ( 2006). Substitute or complement? Spousal influence on volunteering. Journal of Marriage and Family , 68, 305– 319. doi: 10.1111/j.1741-3737.2006.00254.x Google Scholar CrossRef Search ADS   Sha’ked, A. ( 2014). Loss and loneliness in close relationships. Psychology Journal , 11, 13– 23. Shiovitz-Ezra, S., & Ayalon, L. ( 2010). Situational versus chronic loneliness as risk factors for all-cause mortality. International Psychogeriatrics , 22, 455– 462. doi: 10.1017/S1041610209991426 Google Scholar CrossRef Search ADS PubMed  Smith, J. M. ( 2012). Toward a better understanding of loneliness in community-dwelling older adults. The Journal of Psychology , 146, 293– 311. doi: 10.1080/00223980.2011.602132 Google Scholar CrossRef Search ADS PubMed  Steffens, N. K., Cruwys, T., Haslam, C., Jetten, J., & Haslam, S. A. ( 2016). Social group memberships in retirement are associated with reduced risk of premature death: Evidence from a longitudinal cohort study. BMJ Open , 6, e010164. doi: 10.1136/bmjopen-2015-010164 Google Scholar CrossRef Search ADS PubMed  Taube, E., Jakobsson, U., Midlöv, P., & Kristensson, J. ( 2016). Being in a Bubble: The experience of loneliness among frail older people. Journal of Advanced Nursing , 72, 631– 640. doi: 10.1111/jan.12853 Google Scholar CrossRef Search ADS PubMed  Utz, R. L., Carr, D., Nesse, R., & Wortman, C. B. ( 2002). The effect of widowhood on older adults’ social participation: An evaluation of activity, disengagement, and continuity theories. The Gerontologist , 42, 522– 533. doi: 10.1093/geront/42.4.522 Google Scholar CrossRef Search ADS PubMed  Utz, R. L., Swenson, K. L., Caserta, M., Lund, D., & deVries, B. ( 2014). Feeling lonely versus being alone: Loneliness and social support among recently bereaved persons. The Journals of Gerontology. Series B, Psychological Sciences and Social Sciences , 69, 85– 94. doi: 10.1093/geronb/gbt075 Google Scholar CrossRef Search ADS PubMed  van Baarsen, B. ( 2002). Theories on coping with loss: The impact of social support and self-esteem on adjustment to emotional and social loneliness following a partner’s death in later life. The Journals of Gerontology. Series B, Psychological Sciences and Social Sciences , 57, S33– S42. doi: 10.1093/geronb/57.1.S33 Google Scholar CrossRef Search ADS PubMed  Windsor, T. D., Anstey, K. J., & Rodgers, B. ( 2008). Volunteering and psychological well-being among young-old adults: How much is too much? The Gerontologist , 48, 59– 70. doi: 10.1093/geront/48.1.59 Google Scholar CrossRef Search ADS PubMed  © The Author(s) 2017. Published by Oxford University Press on behalf of The Gerontological Society of America. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com. http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png The Journals of Gerontology Series B: Psychological Sciences and Social Sciences Oxford University Press

Does Becoming A Volunteer Attenuate Loneliness Among Recently Widowed Older Adults?

Loading next page...
 
/lp/ou_press/does-becoming-a-volunteer-attenuate-loneliness-among-recently-widowed-CEcDP91cyj
Publisher
Oxford University Press
Copyright
© The Author(s) 2017. Published by Oxford University Press on behalf of The Gerontological Society of America. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com.
ISSN
1079-5014
eISSN
1758-5368
D.O.I.
10.1093/geronb/gbx092
Publisher site
See Article on Publisher Site

Abstract

Abstract Objectives Loneliness is a significant public health concern, particularly for those who have lost a spouse through widowhood. This study examines whether becoming a volunteer at the time of widowhood is associated with reduction of these risks. Method A pooled sample of 5,882 married adults age 51+, drawn from the 2006–2014 waves of the Health and Retirement Study, was used to estimate regression models of the relationship between becoming widowed (relative to staying continuously married) and loneliness, and whether the associated loneliness of having lost a spouse is moderated by starting to volunteer (<2 hr, 2+ hr/week). Results Our results show that for those who become widowed, loneliness is significantly higher than those who stay continuously married. However, starting to volunteer 2+ hr per week is related to attenuated loneliness among the widowed such that widows who volunteer at that intensity have levels of loneliness similar to those of continuously married individuals volunteering at the same intensity. Discussion This study suggests higher intensity volunteering may be a particularly important pathway for alleviating loneliness among older adults who have recently become widowed. Results are discussed in light of theory, future research, and potential interventions. Health and Retirement Study, Productive engagement, Psychological health For many older Americans, loneliness is an unfortunate condition that affects everyday life (Queen, Stawski, Ryan, & Smith, 2014). An increasing body of knowledge on this topic suggests that being lonely is no trivial matter. Indeed, loneliness is associated with poorer physical health, more depressive symptoms, and even earlier mortality (Coyle & Dugan, 2012; Luo, Hawkley, Waite, & Cacioppo, 2012). In fact, the mortality risk associated with lack of strong social relationships is similar to smoking, approximately double that of obesity and quadruple that of exposure to air pollution (Holt-Lunstad, Smith, & Layton, 2010). Clearly, the consequences of loneliness have important implications for health and longevity. The extent to which older adults are lonely varies by a variety of factors. For instance, older adults who maintain social connections to other individuals in their lives and remain engaged in organizations tend to fair better than those who are socially disconnected from the everyday lives of others (Berkman & Syme, 1994). Although narrowing of social networks is typical in older adulthood and generally associated with better emotional well-being (English & Carstensen, 2014), they become especially crucial when immediate interpersonal relationships change, such as those who experience the loss of a spouse through widowhood (Antonucci et al., 2001). One way that these at-risk older adults may be able to obtain or maintain social connectedness despite loss of a spouse is to engage in the community by volunteering. To that end, in this article, we use a sample drawn from the Health and Retirement Study to address the following research questions: Does becoming a volunteer moderate the effect of becoming a widow on loneliness? And, does the “dose” of engagement matter? Widowhood and Loneliness In the United States, nearly one in five people over the age of 50 are lonely. Loneliness is the feeling resulting from a perceived gap between the number and quality of existing social relationships and one’s desired social relationships (Cacioppo, Hawkley, & Thisted, 2010), and is particularly problematic for older adults who experience changes in their social environment and in their ability to interact with others in a meaningful way (Smith, 2012; Taube, Jakobsson, Midlöv, & Kristensson, 2016). Becoming a widow, which is a common and difficult transition in later life, can lead to significant loneliness that is difficult to mitigate (van Baarsen, 2002). According to Socioemotional Selectivity Theory (Carstensen, Isaacowitz, & Charles, 1999) older adults are particularly motivated to maintain and foster emotionally gratifying relationships. Marriage is one of the most important relationships as people age, with cultivation of increasingly positive spousal interactions occurring in later life (Carstensen, Fung, & Charles, 2003). As a result, the loss of a spouse produces a significant gap in one’s social support network—a gap that is difficult to overcome—leading to intense and persistent feelings of loneliness following spousal death even when compensatory social support networks are available (van Baarsen, 2002). Two common interventions that have been used to address loneliness include: (a) increased social interaction and (b) provision of social support (Hawkley & Cacioppo, 2010). Losing a spouse to death—and losing the regular interactions that accompany being married—exposes people to feelings of loneliness because of the loss of social, emotional and instrumental support (Berkman & Syme, 1994; Greenfield & Russell, 2011; Lubben & Gironda, 2003; Pinquart, 2003; Shiovitz-Ezra & Ayalon, 2010). Volunteering as a Potential Intervention A large body of evidence points to a range of health benefits associated with engagement in social and productive activities in later life, particularly volunteering (for reviews see Anderson et al., 2014; Carr, Fried, & Rowe, 2015; Heaven et al., 2013; Jenkinson et al., 2013). Such engagement appears to influence health through psychosocial pathways like positive emotional exchange, associated lifestyle factors (increased physical, cognitive, and social activity), self-esteem and purpose in life, as well as through stress-buffering effects that moderate the influence of stressors on health (Cohen, Gottlieb, & Underwood, 2000; Matz-Costa, Carr, McNamara, & James, 2016). The social aspect of volunteering may be especially beneficial for protecting against loneliness among older adults. Meaningful social relationships are a key contributor to well-being among older adults. For example, social support plays an important role in offsetting loneliness among widows, such as through engagement with close friends, close relatives, and children (de Vries, Utz, Caserta, & Lund, 2014; Utz, Swenson, Caserta, Lund, & deVries, 2014) and has a particularly strong impact on those caregiving for a dying spouse with extensive disabilities or cognitive deficits (Dassel & Carr, 2016; Hawkley & Cacioppo, 2010). Volunteering facilitates enhanced emotional health through meaningful social interactions (see a description of the social benefit hypothesis provided by Li and Ferraro, 2005 explaining the related decrease in depression). Although institutional engagement like volunteering plays an important role in loneliness (Steffens, Cruwys, Haslam, Jetten, & Haslam, 2016) not all forms of institutional engagement seem to be sufficient to address the unique challenges that loneliness presents in late life. Working and religious attendance among younger adults are associated with lower levels of loneliness (Kuwert, Knaevelsrud, & Pietrzak, 2014; Rote, Hill, & Ellison, 2013). However, these activities are not related to reduced loneliness for older adults, perhaps because they do not consistently allow older people to feel that they are contributing in ways that are valued in a mutually beneficial way (Greenfield & Russell, 2011; Hawkley & Cacioppo, 2010; Kahn, Hessling, & Russell, 2003) rendering these activities less emotionally meaningful. Additionally, although small amounts or infrequent volunteer engagement may be advantageous for some aspects of health and well-being in later life, volunteering at low levels—for instance, volunteering at a soup kitchen five times a year—may not be sufficient to combat significant loneliness for those at particularly high risk. However, there is some evidence that higher levels of volunteer engagement, such as volunteering regularly for a few hours a week, may lead to more significant mental health benefits (Morrow-Howell, Hinterlong, Rozario, & Tang, 2003). Other factors have been shown to play a role in shaping loneliness. First, personality may orient certain individuals to be more likely to feel lonely (Hawkley & Cacioppo, 2010), which may also explain why personality is also associated with the decision to volunteer (King, Jackson, Morrow-Howell, & Oltmanns, 2015). Second, loneliness is strongly associated with physical and mental health, with lonely individuals experiencing more symptoms of depression, poorer self-rated health, and poorer cognitive performance (Cacioppo & Hawkley, 2009; Hawkley & Cacioppo, 2010; Shiovitz-Ezra & Ayalon, 2010). Finally, higher socioeconomic status is associated with decreased risk of loneliness (Hawkley et al., 2008) and increased likelihood of volunteering (Morrow-Howell, 2010). Although widowhood is associated with higher levels of loneliness (Sha’ked, 2014), it is not yet clear whether the effects of volunteering on loneliness are the same for people who are continually married as they are for those who experience the loss of a spouse. On the one hand, the number of hours one volunteers is associated with decreases in depressive symptoms and increases in self-efficacy among widows (Li, 2007). On the other hand, no research has explored whether higher levels of volunteering offer greater benefit to those at risk for loneliness. This is an important omission to the literature on the health and emotional benefits of volunteering. Unlike those who are never-married—who presumably never experience the social connectedness associated with being married—those who experience the death of a spouse experience the loss of interaction and support to which they have been accustom. Moreover, although divorce is traumatic for many people, the death of a spouse represents a unique form of loss, and has been referred to as “the most stressful of life events” (Carr, House, Wortman, Nesse, & Kessler, 2001). Thus, reintegration into social networks—in this case through volunteering—may be particularly important for people who have experienced the loss of a spouse to death. To that end, we developed the following hypothesis: Among those who become widowed, those who start a new volunteer role will experience less loneliness than those who do not, with more extensive volunteering associated with greater reduction in loneliness. Method Data The data for this study come from the Health and Retirement Study (HRS). The HRS, a longitudinal study of older Americans, began in 1992 sampling adults born between 1931 and 1941 (individuals initially ages 51 to 61) plus their spouses, oversampling for Hispanics, Blacks, and Florida residents. Follow-up occurred every 2 years, with additional cohorts added in subsequent waves. Since 1998 the HRS has been nationally representative of individuals ages 51 and older, with detailed information collected on demographics, housing situation, household and family, economic, employment, and health information. For the purposes of this study, we use data from the 2006–2014 HRS survey waves. We limited our analyses to these waves because the primary outcome variable is available only in the Leave Behind Questionnaire (LBQ) beginning in the 2006 study wave. The LBQ is collected from half the HRS respondents each wave with an alternating sample such that for every two-wave period, the entire HRS sample is surveyed. Our sample is pooled from respondents ages 51 or older who completed the LBQ in 2006, 2008, or 2010 (i.e., baseline) and again two study waves later when this item was measured again, 4 years later (i.e., Time 2). Because we are interested in the effect of volunteering on loneliness resulting from onset of widowhood, we include only individuals who are married at baseline, and who either remain continuously married or become widowed by Time 2. Our total pooled sample of 5,649 includes 5,057 individuals who remain continuously married, and 592 who become widowed. Measures Outcome variable The outcome variable for this study is a three-item composite measure of loneliness located in the LBQ of the HRS. Although an 11-item loneliness measure is available in the LBQ beginning in 2008, given the limited number of individuals who completed the LBQ questionnaire at two time points and also experienced the loss of a spouse, we used the measure of loneliness available consistently beginning in 2006, allowing us to identify a sufficient sample to address the research questions. This three-item measure has been shown to be a valid assessment of loneliness in the HRS sample (Hughes, Waite, Hawkley, & Cacioppo, 2004). This composite measure is the sum of three indicators, with response options of “1” (often), “2” (some of the time), and “3” (hardly ever or never). These indicators were how often do you feel: (1) you lack companionship; (2) left out; and (3) isolated from others. Each item was reverse coded such that higher numbers indicate greater loneliness. Individuals who did not respond to all three questions were coded missing. The outcome measure is the measure of loneliness measured at Time 2. Key independent variables Our first key independent variable was whether or not someone became widowed by Time 2 (relative to remaining continuously married). Our second key independent variable is a measure of change in volunteer engagement behavior between baseline and Time 2. The HRS measures formal volunteer engagement based on a series of questions about volunteer status, and how much time was spent volunteering for an organization during the previous year. Although it is not possible to measure volunteer hours continuously, these questions make it possible to identify changes in engagement according to three time categories: 1–99 hr per year, 100–199 hr per year, and 200 or more hr per year. We identify five volunteer groups based on these questions. First, we identified those who never volunteered (i.e., non-volunteers at baseline and at Time 2). Second, we identified individuals who continuously volunteered (i.e., were volunteers at both baseline and Time 2). Third, we identified individuals who stopped volunteering (i.e., were volunteers at baseline, but not at Time 2). Finally, we identified individuals who started volunteering. Within this group, we identified those who started volunteering 1–99 hr per year (i.e., non-volunteers at baseline and reported volunteering 1–99 hr in the last year at Time 2), and those who started volunteering 100+ hr per year (i.e., non-volunteers at baseline, and volunteering 100+ hr at Time 2). Due to the small number of individuals who started volunteering 200+ hr per year, we were unable to differentiate individuals in that group. Among those who started volunteering, some were likely volunteers prior to the study period. Due to the goals of this study, however, we are not concerned about individuals who have never reported volunteering. Rather, we focus on the potential for those not volunteering at the time of a transition into widowhood to experience benefits from starting a volunteer activity. We conducted sensitivity analyses of our findings using a lagged volunteer variable (i.e., controlling for volunteer status 2 years prior to baseline), with results remaining unchanged. Details of these analyses are available upon request. Control variables All control variables included in our models are measured during the baseline wave. We control for several demographic factors, including gender, race, years of education, wealth, and age which have been associated with volunteerism in past studies (Anderson et al., 2014). Female was a dichotomously coded indicator of gender. Race/ethnicity was measured with four indicator variables: Non-Hispanic White (reference group); Black; Hispanic; and Other Race. Education was measured in years (range from 0 to 17 years). Wealth was measured as total wealth in dollars, including total housing and nonhousing wealth (excluding pensions) at baseline. Age was measured as the age in years of the respondent at baseline (ranging from age 51 to 99). In addition to demographics, we also control for baseline health factors. Self-rated health is based on a five-point score ranging from “1” (poor health—reference group) to “5” (excellent health). Cognitive performance is based on tests of immediate and delayed recall, with a possible score ranging from 0 to 20. This is a test of episodic memory where individuals are read a list of 10 common words and asked to repeat as many of these words as possible right after being provided the words, and again after the interview continues on other topics 10–15 min later (Ofstedal, Fisher, & Herzog, 2005). The score is the total number of correct answers out of 20. Memory is a well-established metric for cognitive performance, which plays an important role in influencing capacity to engage in executive function tasks, and thus, volunteer engagement. Depressive symptoms is measured using a standard Center for Epidemiological Studies Depression measure, with scores ranging from 0 to 8, and higher numbers indicating more depressive symptoms. Although depression and loneliness are both commonly associated with caregiving and widowhood, recent research shows that they are distinct from one another (Hawkley & Cacioppo, 2010). The health status of a spouse is predictive of the intensity of care provided by the respondent, and plays an important role in shaping the well-being of a spouse in the period following onset of widowhood (Dassel & Carr, 2016; Dassel, Carr, & Vitaliano 2015). Most spouses help care for one another across the duration of a marriage, and it is, therefore, difficult to identify individuals providing care. As a result, we assess the potential impact of a spouse’s health at baseline on the surviving spouse by controlling for objective health measures. First, we control for spousal activities of daily living (ADL) limitations which is a count of the items in which he or she has difficulty: bathing, eating, dressing, walking across the room, or getting in and out of the bed or a chair. Second, we control for spousal memory disease diagnosis, which is a measure of whether or not a doctor has ever provided a diagnosis of any memory related disease (e.g., dementia, Alzheimer’s Disease). Recent research suggests personality plays an important role in shaping likelihood of seeking out volunteer roles and the benefits derived from them (King et al., 2015). Personality also is related to mental health (Cloninger & Zohar, 2011). For these reasons, we control for personality by including measures for the “Big 5” personality traits. These five characteristics (neuroticism, extroversion, openness to experience, conscientiousness, and agreeableness) are based on a series of 31 questions about individual characteristics (e.g., creative, responsible, careful, friendly). Respondents are asked how much they believe that characteristic is like them (coded as: 4 = a lot; 3 = some, 2 = a little, or 1 = not at all), based on Lachman and Weaver (1997). Inclusion of agreeableness did not contribute to improved model fit, did not affect variables of interest, and for the sake of parsimony, was excluded from final models. Details are available upon request. As noted earlier, social interaction and social support are related to decreased symptoms of loneliness (Hawkley & Cacioppo, 2010). Volunteering can be seen as one form of formal social engagement, and is the primary focus of the current article. However, in order to isolate the unique effect of taking on a new volunteer role on alleviating loneliness, we control for other forms of formal and informal social participation at baseline. First, we include three measures related to the number of individuals in respondents’ social networks. Specifically, individuals were asked how many friends, children, and other (non-spouse) family members they are close with. We included a continuous measure of the number of individuals respondents identified as close to them in each of these categories. Second, we controlled for church attendance, identifying individuals who indicated attending church once a month or more. Finally, we identified individuals who are employed and work 1 or more hr per week. Our sample is drawn from individuals who completed the entire LBQ questionnaire at both baseline and at Time 2 (N = 6,217; i.e., 2006 & 2010, 2008 & 2012, or 2010 & 2014). Among these individuals, 150 (2.4%) had missing data on the health of their spouse at baseline, and were excluded from the analyses. Among the remaining (N = 6,067), 115 (1.8%) did not complete the cognitive and mental health questions at baseline, and 19 (0.3%) did not complete the volunteer questions at baseline and at Time 2, and an additional 51 (0.8%) had missing data on education or race. Our total sample includes 5,882 individuals, including 667 who become widowed between baseline and Time 2, and 5,215 individuals who remain continuously married. Analysis Plan To test our research question and hypothesis, Ordinary Least Squares (OLS) regression was used to assess the impact of starting volunteering at a level of 1–99 hr per year or 100+ hr per year (relative to not starting volunteering or no volunteering) on loneliness at Time 2 for those who are continuously married versus those who become widowed. Each person had one observation per variable in the final data file and, thus, multilevel modeling was not necessary (To adjust for the potential interdependencies in the data due to continuously married individuals living in the same household, we conducted our analyses clustered by household, which showed very similar results. Results are available upon request.). First, we entered the widowhood variable into the model while controlling for the raw loneliness score at baseline as well as demographic, health, and personality factors at baseline. Next, we added the volunteer engagement behavior variables to the model while controlling for other forms of social engagement (baseline; Utz, Carr, Nesse, & Wortman, 2002). Finally, we added interaction terms between widowhood and volunteer behavior to examine whether starting volunteering at a level of 1–99 hr per year or 100+ hr per year moderates the relationship between widowhood and change in loneliness. Moderation effects were further examined by calculating marginal effects, holding all other factors at the mean. Results Descriptive Information and Bivariate Analysis Table 1 provides the descriptive data for the sample, which shows that the average loneliness score of the sample at Time 2 is 4.1 (SD = 1.5). About 48% of the sample were non-volunteers, 12.1% stopped volunteering, and nearly one-third (32.2%) volunteered at both baseline and Time 2. Almost 8% of the sample were non-volunteers at baseline but started volunteering at Time 2: 6.3% of the sample started volunteering at an intensity of less than 100 hr per year, and about 1.5% started volunteering at an intensity of 100 hr per year. Table 1. Descriptive Characteristics of Samples   All (N = 5,882)  Became widowed (N = 667)  Continuously married (N = 5,215)  Variable  Meana (SD)  Meana (SD)  Meana (SD)  Loneliness score (wave t+2)  4.142 (1.474)  4.954 (1.666)***  4.038 (1.415)  Loneliness score (wave t)  4.055 (1.418)  4.231 (1.484)***  4.033 (1.408)  Became widowed  0.113      Volunteer behavior   Started Vol 1–99 hr/year  0.063  0.084***  0.060   Started Vol 100+ hr/year  0.015  0.033*  0.013   Continuous volunteer  0.322  0.232***  0.334   Stopped volunteering  0.121  0.133  0.119   Never volunteered  0.479  0.517*  0.474  Female  0.532  0.735***  0.506  Race/ethnicity   White  0.839  0.841***  0.839   Black  0.071  0.090***  0.068   Hispanic  0.069  0.055***  0.071   Other race  0.021  0.013*  0.022  Years of education  13.134 (2.845)  12.648 (2.808)***  13.197 (2.844)  Household wealth (in $100,000s)  5.253 (10.013)  5.211 (15.563)  5.259 (9.063)  Age  67.912 (8.410)  71.510 (8.507)***  67.452 (8.286)  Self-rated health  2.634 (1.003)  2.706 (1.018)*  2.625 (1.001)  Cognitive score  10.278 (3.083)  9.802 (3.090)***  10.339 (3.077)  Depressive symptoms  0.975 (1.604)  1.282 (1.799)***  0.935 (1.573)  Spouse ADL limitations  0.238 (1.604)  0.867 (1.508)***  0.157 (1.573)  Spouse memory disease diag  0.028  0.111***  0.017  Neuroticism  1.958 (0.574)  2.005 (0.571)*  1.952 (0.574)  Extroversion  3.218 (0.548)  3.252 (0.527)  3.213 (0.550)  Openness to experience  2.954 (0.537)  2.927 (0.549)  2.957 (0.535)  Conscientiousness  3.364 (0.430)  3.372 (0.549)  3.363 (0.428)  Number of close friends  4.251 (4.980)  3.961 (4.389)  4.288 (5.049)  Number of close relatives  3.460 (5.126)  3.387 (4.205)  3.469 (5.232)  Number of close children  2.463 (2.601)  2.625 (2.630)  2.442 (2.596)  Attends church monthly  0.572  0.588  0.570  Works  0.374  0.238***  0.391    All (N = 5,882)  Became widowed (N = 667)  Continuously married (N = 5,215)  Variable  Meana (SD)  Meana (SD)  Meana (SD)  Loneliness score (wave t+2)  4.142 (1.474)  4.954 (1.666)***  4.038 (1.415)  Loneliness score (wave t)  4.055 (1.418)  4.231 (1.484)***  4.033 (1.408)  Became widowed  0.113      Volunteer behavior   Started Vol 1–99 hr/year  0.063  0.084***  0.060   Started Vol 100+ hr/year  0.015  0.033*  0.013   Continuous volunteer  0.322  0.232***  0.334   Stopped volunteering  0.121  0.133  0.119   Never volunteered  0.479  0.517*  0.474  Female  0.532  0.735***  0.506  Race/ethnicity   White  0.839  0.841***  0.839   Black  0.071  0.090***  0.068   Hispanic  0.069  0.055***  0.071   Other race  0.021  0.013*  0.022  Years of education  13.134 (2.845)  12.648 (2.808)***  13.197 (2.844)  Household wealth (in $100,000s)  5.253 (10.013)  5.211 (15.563)  5.259 (9.063)  Age  67.912 (8.410)  71.510 (8.507)***  67.452 (8.286)  Self-rated health  2.634 (1.003)  2.706 (1.018)*  2.625 (1.001)  Cognitive score  10.278 (3.083)  9.802 (3.090)***  10.339 (3.077)  Depressive symptoms  0.975 (1.604)  1.282 (1.799)***  0.935 (1.573)  Spouse ADL limitations  0.238 (1.604)  0.867 (1.508)***  0.157 (1.573)  Spouse memory disease diag  0.028  0.111***  0.017  Neuroticism  1.958 (0.574)  2.005 (0.571)*  1.952 (0.574)  Extroversion  3.218 (0.548)  3.252 (0.527)  3.213 (0.550)  Openness to experience  2.954 (0.537)  2.927 (0.549)  2.957 (0.535)  Conscientiousness  3.364 (0.430)  3.372 (0.549)  3.363 (0.428)  Number of close friends  4.251 (4.980)  3.961 (4.389)  4.288 (5.049)  Number of close relatives  3.460 (5.126)  3.387 (4.205)  3.469 (5.232)  Number of close children  2.463 (2.601)  2.625 (2.630)  2.442 (2.596)  Attends church monthly  0.572  0.588  0.570  Works  0.374  0.238***  0.391  Notes: ADL = activities of daily living. aMeans are provided for continuous variables, proportions for dichotomous variables. Significant differences between widowed and continuously married individuals is noted in the following way: *p < .05. ***p < .001. Significance is determined using chi-square tests for dichotomous variables, and t tests for continuous variables. View Large We performed bivariate analyses (t tests and chi-square tests) to identify significant differences in characteristics of those who became widowed relative to those who remained married. Results indicate that the average loneliness score varies by marital status. Those who became widowed had a significantly greater loneliness at baseline and at Time 2 relative to those who were married (p < .001). With respect to volunteering, engagement varies by marital status such that those who became widowed were more likely to start volunteering than those who stayed married, particularly at the highest intensity level (p < .001). Those who became widows were also less likely to be continuous volunteers (p < .001) and more likely to be non-volunteers (p < .05). In this sample, those who went on to become widowed were, at baseline: more likely to be women (p < .001), Black (p < .05) and not Hispanic (p < .001) or another race (p < .005), have less education (p < .001), older (p < .001), have poorer self-rated health (p < .05), lower cognitive performance (p < .001); and more depressive symptoms (p < .001). They are also more likely to have a more disabled spouse (p < .001), who is more likely to have been diagnosed with a memory disorder (p < .001). They have a higher neuroticism score, and are less likely to be working (p < .001) than their continuously married counterparts. OLS Regression and Moderation Analysis Table 2 shows the results from OLS regression models predicting loneliness score at Time 2. In Model 1, we examine the direct association between change in marital status and change in loneliness, adjusting for loneliness at baseline, demographic factors, health and personality. Results indicate that widowhood is related to significantly higher loneliness at Time 2 (β = 0.505, p < .001). In Model 2, we assess the direct relation between change in volunteer behavior and loneliness, adjusting for other forms of social engagement and all factors from Model 1. Results show that volunteer behavior was not significantly related to loneliness score at Time 2. Further, when controlling for volunteer engagement, the direct effect of marital status on loneliness did not change. Table 2. OLS Regression Models Predicting Change in Loneliness (N = 5,882)   Model 1  Model 2  Model 3    Coeff.  SE  Coeff.  SE  Coeff.  SE  Loneliness score (wave t)  0.465  0.013***  0.462  0.013***  0.462  0.013***  Became widoweda  0.789  0.053***  0.785  0.053***  0.809  0.055***  Volunteer behaviorb   1–99 hr/year      −0.010  0.067  −0.007  0.073   100+ hr/year      −0.067  0.129  0.078  0.147   Continuous volunteer      −0.055  0.042  −0.053  0.043   Stopped volunteering      0.018  0.052  0.022  0.055  Number of close friends      −0.007  0.004  −0.007  0.004  Number of close relatives      −0.001  0.003  −0.001  0.003  Number of close children      −0.005  0.007  −0.005  0.007  Attends church monthly      0.060  0.036  0.059  0.036  Works      0.008  0.037  0.009  0.037  Start Vol 1–99 hr × Became widowed          −0.022  0.182  Start Vol 100+ hr × Became widowed          −0.608  0.299*  Continuous volunteer × Became widowed          −0.013  0.122  Stopped volunteering × Became widowed          −0.027  0.153  Femalec  0.133  0.034***  0.130  0.034***  0.130  0.034***  Raced   White               Black  −0.035  0.063  −0.043  0.064  −0.041  0.064   Hispanic  −0.211  0.109  −0.217  0.109*  −0.214  0.109*   Other race  −0.269  0.067***  −0.277  0.067***  −0.277  0.067***  Years of education  0.009  0.007  0.010  0.007  0.010  0.007  Household wealth  0.000  0.000*  0.000  0.000*  0.000  0.000*  Age  0.001  0.002  0.001  0.002  0.001  0.002  Self-rated health  0.078  0.018***  0.078  0.018***  0.077  0.018***  Cognitive score  −0.009  0.006  −0.008  0.006  −0.008  0.006  Depressive symptoms  0.056  0.012***  0.056  0.012***  0.055  0.012***  Spouse ADL limitations  −0.034  0.022  −0.033  0.022  −0.032  0.022  Spouse memory disease diagnosed  −0.051  0.099  −0.053  0.100  −0.040  0.100  Neuroticism  0.207  0.032***  0.206  0.032***  0.207  0.032***  Extroversion  −0.186  0.036***  −0.178  0.036***  −0.178  0.036***  Openness to experience  0.154  0.037***  0.163  0.038***  0.165  0.038***  Conscientiousness  −0.114  0.042**  −0.120  0.042**  −0.121  0.042**  Constant  1.932  0.268***  1.911  0.278***  1.903  0.278***  Log likelihood (model)  −9397.838  −9392.248  −9390.174    Model 1  Model 2  Model 3    Coeff.  SE  Coeff.  SE  Coeff.  SE  Loneliness score (wave t)  0.465  0.013***  0.462  0.013***  0.462  0.013***  Became widoweda  0.789  0.053***  0.785  0.053***  0.809  0.055***  Volunteer behaviorb   1–99 hr/year      −0.010  0.067  −0.007  0.073   100+ hr/year      −0.067  0.129  0.078  0.147   Continuous volunteer      −0.055  0.042  −0.053  0.043   Stopped volunteering      0.018  0.052  0.022  0.055  Number of close friends      −0.007  0.004  −0.007  0.004  Number of close relatives      −0.001  0.003  −0.001  0.003  Number of close children      −0.005  0.007  −0.005  0.007  Attends church monthly      0.060  0.036  0.059  0.036  Works      0.008  0.037  0.009  0.037  Start Vol 1–99 hr × Became widowed          −0.022  0.182  Start Vol 100+ hr × Became widowed          −0.608  0.299*  Continuous volunteer × Became widowed          −0.013  0.122  Stopped volunteering × Became widowed          −0.027  0.153  Femalec  0.133  0.034***  0.130  0.034***  0.130  0.034***  Raced   White               Black  −0.035  0.063  −0.043  0.064  −0.041  0.064   Hispanic  −0.211  0.109  −0.217  0.109*  −0.214  0.109*   Other race  −0.269  0.067***  −0.277  0.067***  −0.277  0.067***  Years of education  0.009  0.007  0.010  0.007  0.010  0.007  Household wealth  0.000  0.000*  0.000  0.000*  0.000  0.000*  Age  0.001  0.002  0.001  0.002  0.001  0.002  Self-rated health  0.078  0.018***  0.078  0.018***  0.077  0.018***  Cognitive score  −0.009  0.006  −0.008  0.006  −0.008  0.006  Depressive symptoms  0.056  0.012***  0.056  0.012***  0.055  0.012***  Spouse ADL limitations  −0.034  0.022  −0.033  0.022  −0.032  0.022  Spouse memory disease diagnosed  −0.051  0.099  −0.053  0.100  −0.040  0.100  Neuroticism  0.207  0.032***  0.206  0.032***  0.207  0.032***  Extroversion  −0.186  0.036***  −0.178  0.036***  −0.178  0.036***  Openness to experience  0.154  0.037***  0.163  0.038***  0.165  0.038***  Conscientiousness  −0.114  0.042**  −0.120  0.042**  −0.121  0.042**  Constant  1.932  0.268***  1.911  0.278***  1.903  0.278***  Log likelihood (model)  −9397.838  −9392.248  −9390.174  Notes: ADL = activities of daily living. Italic values indicate the standard errors (SE). Reference groups include: aContinuously married individuals; bNever volunteerd; cMale; dNon-Hispanic White. *p < .05. **p < .01. ***p < .001. View Large Finally, in Model 3, we examine the effect of new engagement in volunteering at two intensity levels on loneliness by assessing the interaction effect between volunteer behaviors and onset of widowhood. Results show a significant interaction between widowhood and starting volunteering at 100+ hr per year. Specifically, the associated benefits of volunteering 100+ hr per year on loneliness is dependent on whether an individual became widowed (p < .05). A post-estimation Wald significance test further confirms that the moderation of volunteer behavior and becoming widowed is significant (p < .001). To more carefully examine these relationships, we calculate the marginal effects (i.e., predicted probabilities) for loneliness at Time 2 (holding all other factors at the mean) in relation to volunteer intensity for those who became widows versus those who stayed married. Figure 1 shows that starting volunteering is related to reduced loneliness among those who became widows, but only at the highest intensity level (i.e., volunteering 100+ hr per year). Specifically, relative to those who were continuously married, those who became widowed had a statistically higher predicted loneliness score if they never volunteered (4.88 compared to 4.07), stopped volunteering (4.87 compared to 4.09), were continuously volunteers (4.81 compared to 4.01), or started volunteering, but only 1–99 hr per year (4.81 compared to 4.06). Volunteering 100+ hr per year moderates the effect of widowhood on loneliness, and these volunteers had a similar predicted loneliness score (4.34 compared to 4.15) to those who were continuously married. What this means is that more intense levels of loneliness persist for new widows relative to continuously married individuals, but loneliness is lower when widows start volunteering approximately 2 hr per week, on average, or more. New widows who start volunteering at that threshold experience similar levels of loneliness as continuously married individuals. Figure 1. View largeDownload slide Predicted loneliness score by volunteer group and marital status. Note: Error bars are the 95% confidence intervals. Figure 1. View largeDownload slide Predicted loneliness score by volunteer group and marital status. Note: Error bars are the 95% confidence intervals. Discussion Becoming a widow is one of the most difficult transitions that people face in later life. Widowhood is associated with significant increases in loneliness, leaving widows at risk for additional health consequences, including depression, physical health decline, and increased mortality risk. In this article, we examined the relation between starting volunteering and loneliness among older adults who became widowed over a 4-year period relative to those who remained continuously married. We discovered that volunteering moderates the negative effects of loneliness for those who become widowed, but only in relation to engagement in 2 or more hr per week, on average. Although our research cannot explain the mechanisms by which volunteering is beneficial, there are several plausible reasons why volunteering on a regular basis might have the potential to significantly alleviate loneliness in new widows. Previous research shows that loneliness can be addressed through increases in social interaction and increases in social support (Hawkley & Cacioppo, 2010). Although volunteering comes in many forms, some likely requiring more social interaction than others, volunteering almost always involves social engagement with others. Individuals who have recently lost their primary form of social support (i.e., their spouse) have the most to benefit from new sources of social engagement. Additionally, beyond the social aspect of volunteering, it may be that engaging at a significant intensity in work that has a social purpose—as volunteer work often does—bolsters the kinds of personal resources (e.g., health behaviors, self-esteem, purpose in life, sense of control) that are needed to manage the grief and loss associated with widowhood (Matz-Costa et al., 2016). Starting to volunteer on a regular basis may help in establishing an integrated sense of self that compensates for the loss of this salient life role (i.e., identity as married) and social resource and protects against a sense of loneliness (Hobfoll & Wells, 1998). However, it is important to note that not all older adults—particularly widows—have the same access or resources to volunteer. In general, older adults are more likely to volunteer if they are highly educated, have higher incomes, work part time, are married, and have a spouse who also volunteers (Bureau of Labor Statistics, U.S. Department of Labor, 2013; Rotolo & Wilson, 2006). Widowhood is often associated with significant grief and depressive symptoms, including loss of interest in activity (Bonanno, 2004); a substantial loss of income and often times poverty (poverty rates among elderly female widows is 3–4 times higher than rates among elderly married women (McGarry & Schoeni, 2005) and increased household responsibilities, all of which are significant barriers to starting a new volunteer role (McBride, 2007). Future research should explore disparities in economic and health resources as well as other structural barriers (e.g., lack of knowledge about volunteer opportunities, lack of skills, time constraints) among widows that may restrict access to and availability of meaningful volunteer roles. Findings from this study need to be understood in light of its limitations. First, we had to limit our sample to those who completed both the HRS core and the LBQ questionnaires during both time points in which loneliness was measured. As a result, the final sample size for this analysis is smaller than what could be obtained if using the HRS core survey data alone. This decreased the potential power of our analyses. Second, we are unable to draw conclusions regarding the directionality of these effects because, despite our longitudinal design, it was not possible to control for all other potential factors that may differentiate continuously married individuals from those who became widowed. Thus, it is possible that the moderation effect of starting volunteering is due to widows who are lonelier choosing to volunteer to off-set these feelings rather than the act of volunteering contributing to decreased loneliness. Lastly, we were unable to take into account one’s history of volunteer engagement prior to baseline in this study. It is possible that some non-volunteers had previously engaged in volunteer activities prior to widowhood, lost this role due to their caregiving responsibilities, and returned following onset of widowhood. This explanation is in line with Utz and colleagues (2002) who point out that “older adults, particularly those experiencing a devastating or stressful loss such as widowhood, tend to rely on the convoy of lifelong social relations from whom they derive support, self-definition, and a sense of stability and continuity” and that widowed persons, “in both their intentions and their behaviors, prefer to maintain the relationships and activities that have been most meaningful and rewarding throughout the life course” (p. 532). Although the goals of this study focus on whether starting a new volunteering role relates to lower levels of loneliness following widowhood regardless of previous volunteer history, future research might examine transitions into and out of volunteer work, its co-occurrence with caregiving and its effect on loneliness to determine if these effects vary based on history of volunteer engagement. Despite these limitations, the results from this study provide encouraging support for the potential of volunteer engagement as an intervention for alleviating loneliness among widows. There are few volunteer programs designed to target the widows, and interventions designed to facilitate starting a new volunteer role at this critical life juncture could play an important role in delaying onset of subsequent health consequences associated with spousal loss. In addition, given the numerous benefits associated with engaging in volunteering in general, and with 2 hr or more per week in particular (Windsor, Anstey, & Rodgers, 2008), individuals who start a new volunteer role and continue to remain engaged over an extended period may experience health and well-being benefits that facilitate more successful aging. In conclusion, loneliness is a significant public health concern for individuals who become widowed. Formal volunteering, however, may offer a means by which to cultivate personal resources at the transition into widowhood. The findings of the current study suggest that interventions that encourage volunteering, particularly higher intensity volunteering, among those who become widowed may be effective in helping to mitigate loneliness. Funding No funding was received in relation to this research project. Acknowledgments D. C. Carr planned the study, performed all statistical analysis and reporting of results, wrote the abstract, contributed to development of the background and discussion sections, and supervised the writing of the article. B. L. Kail provided methodological consultation, contributed to development of the methodology section, and contributed to introduction and literature review. C. Matz-Costa contributed to the introduction and discussion sections, and assisted with formatting and references. Y. Z. Shavit contributed to the introduction section, and consulted on development of measures. Conflict of Interest None of the authors associated with publication have a conflict of interest in relation to this research project. References Anderson, N. D., Damianakis, T., Kröger, E., Wagner, L. M., Dawson, D. R., Binns, M. A.,… Cook, S. L.; BRAVO Team. ( 2014). The benefits associated with volunteering among seniors: A critical review and recommendations for future research. Psychological Bulletin , 140, 1505– 1533. doi: 10.1037/a0037610 Google Scholar CrossRef Search ADS PubMed  Antonucci, T. C., Lansford, J. E., Schaberg, L., Baltes, M., Takahashi, K., Dartigues, J. F.,… Fuhrer, R. ( 2001). Widowhood and illness: A comparison of social network characteristics in France, Germany, Japan, and the United States. Psychology and Aging , 16, 655– 665. doi: 10.1037/0882-7974.16.4.655 Google Scholar CrossRef Search ADS PubMed  Berkman, L. F., & Syme, S. L. ( 1994). Social networks, host resistance, and mortality: A nine-year follow-up study of Alameda County residents. In A. Steptoe & J. Wardle (Eds.), Psychosocial processes and health  (pp. 43– 67). Cambridge, UK: Cambridge University Press. Retrieved from http://ebooks.cambridge.org/ref/id/CBO9780511759048A010 Bonanno, G. A. ( 2004). Loss, trauma, and human resilience: Have we underestimated the human capacity to thrive after extremely aversive events? The American Psychologist , 59, 20– 28. doi: 10.1037/0003-066X.59.1.20 Google Scholar CrossRef Search ADS PubMed  Bureau of Labor Statistics, U.S. Department of Labor. ( 2013). Volunteering in the United States, 2013. Retrieved from http://www.bls.gov/news.release/volun.nr0.htm Cacioppo, J. T., & Hawkley, L. C. ( 2009). Perceived social isolation and cognition. Trends in Cognitive Sciences , 13, 447– 454. doi: 10.1016/j.tics.2009.06.005 Google Scholar CrossRef Search ADS PubMed  Cacioppo, J. T., Hawkley, L. C., & Thisted, R. A. ( 2010). Perceived social isolation makes me sad: 5-year cross-lagged analyses of loneliness and depressive symptomatology in the Chicago Health, Aging, and Social Relations Study. Psychology and Aging , 25, 453– 463. doi: 10.1037/a0017216 Google Scholar CrossRef Search ADS PubMed  Carr, D., House, J. S., Wortman, C., Nesse, R., & Kessler, R. C. ( 2001). Psychological adjustment to sudden and anticipated spousal loss among older widowed persons. The Journals of Gerontology. Series B, Psychological Sciences and Social Sciences , 56, S237– S248. doi: 10.1093/geronb/56.4.S237 Google Scholar CrossRef Search ADS PubMed  Carr, D. C., Fried, L. P., & Rowe, J. W. ( 2015). Productivity and engagement in an aging America: The role of volunteerism. Daedalus , 144, 55– 67. doi: 10.1162/DAED_a_00330 Google Scholar CrossRef Search ADS   Carstensen, L., Fung, H., & Charles, S. T. ( 2003). Socioemotional selectivity theory and the regulation of emotion in the second half of life. Motivation and Emotion , 27, 103– 123. doi: 10.1023/A:1024569803230 Google Scholar CrossRef Search ADS   Carstensen, L. L., Isaacowitz, D. M., & Charles, S. T. ( 1999). Taking time seriously. A theory of socioemotional selectivity. The American Psychologist , 54, 165– 181. doi: 10.1037/0003-066X.54.3.165 Google Scholar CrossRef Search ADS PubMed  Cloninger, C. R., & Zohar, A. H. ( 2011). Personality and the perception of health and happiness. Journal of Affective Disorders , 128, 24– 32. doi: 10.1016/j.jad.2010.06.012 Google Scholar CrossRef Search ADS PubMed  Cohen, S., Gottlieb, B., & Underwood, L. ( 2000). Social relationships and health. In S. Cohen, L. Underwood, & B. Gottlieb (Eds.), Measuring and intervening in social support  (pp. 3–25). New York, NY: Oxford University Press. Coyle, C. E., & Dugan, E. ( 2012). Social isolation, loneliness and health among older adults. Journal of Aging and Health , 24( 8), 1346– 1363. doi: 10.1177/0898264312460275 Google Scholar CrossRef Search ADS PubMed  Dassel, K. B., & Carr, D. C. ( 2016). Does dementia caregiving accelerate frailty? Findings from the Health and Retirement Study. The Gerontologist , 56, 444– 450. doi: 10.1093/geront/gnu078 Google Scholar CrossRef Search ADS PubMed  Dassel, K. B., Carr, D. C., & Vitaliano, P. ( 2015). Does caring for a spouse with dementia accelerate cognitive decline? Findings from the Health and Retirement Study. The Gerontologist . Advance online publication. doi: 10.1093/geront/39.2.177 de Vries, B., Utz, R., Caserta, M., & Lund, D. ( 2014). Friend and family contact and support in early widowhood. The Journals of Gerontology. Series B, Psychological Sciences and Social Sciences , 69, 75– 84. doi: 10.1093/geronb/gbt078 Google Scholar CrossRef Search ADS PubMed  English, T., & Carstensen, L. L. ( 2014). Selective narrowing of social networks across adulthood is associated with improved emotional experience in daily life. International Journal of Behavioral Development , 38, 195– 202. doi: 10.1177/0165025413515404 Google Scholar CrossRef Search ADS PubMed  Greenfield, E. A., & Russell, D. ( 2011). Identifying living arrangements that heighten risk for loneliness in later life: Evidence from the U.S. National Social Life, Health, and Aging Project. Journal of Applied Gerontology , 30, 524– 534. doi: 10.1177/0733464810364985 Google Scholar CrossRef Search ADS   Hawkley, L. C., & Cacioppo, J. T. ( 2010). Loneliness matters: A theoretical and empirical review of consequences and mechanisms. Annals of Behavioral Medicine , 40, 218– 227. doi: 10.1007/s12160-010-9210-8 Google Scholar CrossRef Search ADS PubMed  Hawkley, L. C., Hughes, M. E., Waite, L. J., Masi, C. M., Thisted, R. A., & Cacioppo, J. T. ( 2008). From social structural factors to perceptions of relationship quality and loneliness: The Chicago health, aging, and social relations study. The Journals of Gerontology. Series B, Psychological Sciences and Social Sciences , 63, S375– S384. doi: 10.1093/geronb/63.6.S375 Google Scholar CrossRef Search ADS PubMed  Heaven, B., Brown, L. J., White, M., Errington, L., Mathers, J. C., & Moffatt, S. ( 2013). Supporting well-being in retirement through meaningful social roles: Systematic review of intervention studies. The Milbank Quarterly , 91, 222– 287. doi: 10.1111/milq.12013 Google Scholar CrossRef Search ADS PubMed  Hobfoll, S. E., & Wells, J. D. ( 1998). Conservation of resources, stress, and aging: Why do some slide and some spring? In J. Lomranz (Ed.), The handbook of aging and mental health: An integrative approach  (pp. 121– 134). New York & London: Plenum Press. Google Scholar CrossRef Search ADS   Holt-Lunstad, J., Smith, T. B., & Layton, J. B. ( 2010). Social relationships and mortality risk: A meta-analytic review. PLoS Medicine , 7, e1000316. doi: 10.1371/journal.pmed.1000316 Google Scholar CrossRef Search ADS PubMed  Hughes, M. E., Waite, L. J., Hawkley, L. C., & Cacioppo, J. T. ( 2004). A short scale for measuring loneliness in large surveys: Results from two population-based studies. Research on Aging , 26, 655– 672. doi: 10.1177/0164027504268574 Google Scholar CrossRef Search ADS PubMed  Jenkinson, C. E., Dickens, A. P., Jones, K., Thompson-Coon, J., Taylor, R. S., Rogers, M.,… Richards, S. H. ( 2013). Is volunteering a public health intervention? A systematic review and meta-analysis of the health and survival of volunteers. BMC Public Health , 13, 773. doi: 10.1186/1471-2458-13-773 Google Scholar CrossRef Search ADS PubMed  Kahn, J. H., Hessling, R. M., & Russell, D. W. ( 2003). Social support, health, and well-being among the elderly: What is the role of negative affectivity? Personality and Individual Differences , 35, 5– 17. doi: 10.1016/S0191-8869(02)00135-6 Google Scholar CrossRef Search ADS   King, H. R., Jackson, J. J., Morrow-Howell, N., & Oltmanns, T. F. ( 2015). Personality accounts for the connection between volunteering and health. The Journals of Gerontology. Series B, Psychological Sciences and Social Sciences , 70, 691– 697. doi: 10.1093/geronb/gbu012 Google Scholar CrossRef Search ADS PubMed  Kuwert, P., Knaevelsrud, C., & Pietrzak, R. H. ( 2014). Loneliness among older veterans in the United States: Results from the National Health and Resilience in Veterans Study. The American Journal of Geriatric Psychiatry , 22( 6), 564– 569. doi: 10.1016/ j.jagp.2013.02.013 Google Scholar CrossRef Search ADS PubMed  Lachman, M., & Weaver, S. ( 1997). Midlife devlopment inventory (MIDI) personality sacales: scale construction and scoring. Brandeis University. Retrieved from http://www.brandeis.edu/projects/lifesapn/scales.html Li, Y. ( 2007). Recovering from spousal bereavement in later life: Does volunteer participation play a role? The Journals of Gerontology. Series B, Psychological Sciences and Social Sciences , 62, S257– S266. doi: 10.1093/geronb/62.4.S257 Google Scholar CrossRef Search ADS PubMed  Li, Y., & Ferraro, K. F. ( 2005). Volunteering and depression in later life: Social benefit or selection processes? Journal of Health and Social Behavior , 46, 68– 84. doi: 10.1177/002214650504600106 Google Scholar CrossRef Search ADS PubMed  Lubben, J., & Gironda, M. ( 2003). Centrality of social ties to the health and well-being of older adults. In B. Berkman & L. Harootyan (Eds.), Social work and health care in an aging world: Informing education, policy, practice, and research  (pp. 319– 350). New York, NY: Springer Press. Luo, Y., Hawkley, L. C., Waite, L. J., & Cacioppo, J. T. ( 2012). Loneliness, health, and mortality in old age: A national longitudinal study. Social Science & Medicine , 74( 6), 907– 914. doi: 10.1016/j.socscimed.2011.11.028 Google Scholar CrossRef Search ADS   Matz-Costa, C., Carr, D. C., McNamara, T. K., & James, J. B. ( 2016). Physical, cognitive, social, and emotional mediators of activity involvement and health in later life. Research on Aging , 38, 791– 815. doi: 10.1177/0164027515606182 Google Scholar CrossRef Search ADS PubMed  McBride, A. ( 2007). Civic engagement, older adults, and inclusion. Generations , 30, 66– 71. McGarry, K., & Schoeni, R. F. ( 2005). Medicare gaps and widow poverty. Social Security Bulletin , 66, 58– 74. Google Scholar PubMed  Morrow-Howell, N. ( 2010). Volunteering in later life: Research frontiers. The Journals of Gerontology. Series B, Psychological Sciences and Social Sciences , 65, 461– 469. doi: 10.1093/geronb/gbq024 Google Scholar CrossRef Search ADS PubMed  Morrow-Howell, N., Hinterlong, J., Rozario, P. A., & Tang, F. ( 2003). Effects of volunteering on the well-being of older adults. The Journals of Gerontology. Series B, Psychological Sciences and Social Sciences , 58, S137– S145. doi: 10.1093/geronb/58.3.S137 Google Scholar CrossRef Search ADS PubMed  Ofstedal, M. B., Fisher, G. G., & Herzog, A. R. ( 2005). HRS/AHEAD documentation report: Documentation of cognitive functioning measures in the Health and Retirement Study . Ann Arbor, MI: HRS Health Working Group, Survey Research Center, University of Michigan. Retrieved from http://hrsonline.isr.umich.edu/sitedocs/userg/dr-006.pdf Pinquart, M. ( 2003). Loneliness in married, widowed, divorced, and never-married older adults. Journal of Social and Personal Relationships , 20, 31– 53. Google Scholar CrossRef Search ADS   Queen, T. L., Stawski, R. S., Ryan, L. H., & Smith, J. ( 2014). Loneliness in a day: Activity engagement, time alone, and experienced emotions. Psychology and Aging , 29( 2), 297– 305. doi: 10.1037/a0036889 Google Scholar CrossRef Search ADS PubMed  Rote, S., Hill, T. D., & Ellison, C. G. ( 2013). Religious attendance and loneliness in later life. The Gerontologist , 53( 1), 39– 50. doi: 10.1093/geront/gns063 Google Scholar CrossRef Search ADS PubMed  Rotolo, T., & Wilson, J. ( 2006). Substitute or complement? Spousal influence on volunteering. Journal of Marriage and Family , 68, 305– 319. doi: 10.1111/j.1741-3737.2006.00254.x Google Scholar CrossRef Search ADS   Sha’ked, A. ( 2014). Loss and loneliness in close relationships. Psychology Journal , 11, 13– 23. Shiovitz-Ezra, S., & Ayalon, L. ( 2010). Situational versus chronic loneliness as risk factors for all-cause mortality. International Psychogeriatrics , 22, 455– 462. doi: 10.1017/S1041610209991426 Google Scholar CrossRef Search ADS PubMed  Smith, J. M. ( 2012). Toward a better understanding of loneliness in community-dwelling older adults. The Journal of Psychology , 146, 293– 311. doi: 10.1080/00223980.2011.602132 Google Scholar CrossRef Search ADS PubMed  Steffens, N. K., Cruwys, T., Haslam, C., Jetten, J., & Haslam, S. A. ( 2016). Social group memberships in retirement are associated with reduced risk of premature death: Evidence from a longitudinal cohort study. BMJ Open , 6, e010164. doi: 10.1136/bmjopen-2015-010164 Google Scholar CrossRef Search ADS PubMed  Taube, E., Jakobsson, U., Midlöv, P., & Kristensson, J. ( 2016). Being in a Bubble: The experience of loneliness among frail older people. Journal of Advanced Nursing , 72, 631– 640. doi: 10.1111/jan.12853 Google Scholar CrossRef Search ADS PubMed  Utz, R. L., Carr, D., Nesse, R., & Wortman, C. B. ( 2002). The effect of widowhood on older adults’ social participation: An evaluation of activity, disengagement, and continuity theories. The Gerontologist , 42, 522– 533. doi: 10.1093/geront/42.4.522 Google Scholar CrossRef Search ADS PubMed  Utz, R. L., Swenson, K. L., Caserta, M., Lund, D., & deVries, B. ( 2014). Feeling lonely versus being alone: Loneliness and social support among recently bereaved persons. The Journals of Gerontology. Series B, Psychological Sciences and Social Sciences , 69, 85– 94. doi: 10.1093/geronb/gbt075 Google Scholar CrossRef Search ADS PubMed  van Baarsen, B. ( 2002). Theories on coping with loss: The impact of social support and self-esteem on adjustment to emotional and social loneliness following a partner’s death in later life. The Journals of Gerontology. Series B, Psychological Sciences and Social Sciences , 57, S33– S42. doi: 10.1093/geronb/57.1.S33 Google Scholar CrossRef Search ADS PubMed  Windsor, T. D., Anstey, K. J., & Rodgers, B. ( 2008). Volunteering and psychological well-being among young-old adults: How much is too much? The Gerontologist , 48, 59– 70. doi: 10.1093/geront/48.1.59 Google Scholar CrossRef Search ADS PubMed  © The Author(s) 2017. Published by Oxford University Press on behalf of The Gerontological Society of America. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com.

Journal

The Journals of Gerontology Series B: Psychological Sciences and Social SciencesOxford University Press

Published: Mar 1, 2018

There are no references for this article.

You’re reading a free preview. Subscribe to read the entire article.


DeepDyve is your
personal research library

It’s your single place to instantly
discover and read the research
that matters to you.

Enjoy affordable access to
over 12 million articles from more than
10,000 peer-reviewed journals.

All for just $49/month

Explore the DeepDyve Library

Unlimited reading

Read as many articles as you need. Full articles with original layout, charts and figures. Read online, from anywhere.

Stay up to date

Keep up with your field with Personalized Recommendations and Follow Journals to get automatic updates.

Organize your research

It’s easy to organize your research with our built-in tools.

Your journals are on DeepDyve

Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.

All the latest content is available, no embargo periods.

See the journals in your area

Monthly Plan

  • Read unlimited articles
  • Personalized recommendations
  • No expiration
  • Print 20 pages per month
  • 20% off on PDF purchases
  • Organize your research
  • Get updates on your journals and topic searches

$49/month

Start Free Trial

14-day Free Trial

Best Deal — 39% off

Annual Plan

  • All the features of the Professional Plan, but for 39% off!
  • Billed annually
  • No expiration
  • For the normal price of 10 articles elsewhere, you get one full year of unlimited access to articles.

$588

$360/year

billed annually
Start Free Trial

14-day Free Trial