Aging in Place With a Spouse in Need: Neighborhood Cohesion and Older Adult Spouses’ Physical and Mental Health

Abstract

This study examines the association of perceived neighborhood cohesion (NC) with older adults’ health and the buffering effects of NC against the negative effects of spousal caregiving on health. Data of 3329 community-living older adults living with a spouse in need of care from the Health and Retirement Study were collected at two time-points. Multiple regression analyses were computed for each of the four health outcomes. For men, NC predicted fewer depressive symptoms and better cognition. NC buffered the negative effect of providing activities of daily living (ADL) help to the wife on cognition. For women, NC predicted fewer depressive symptoms and better cognition. NC buffered the negative effect of providing ADL help to the husband on ADL difficulties. The results accentuate the importance of residency location for older adults’ physical and mental health. The health benefits of NC may have more implications for older adults providing spousal care.

Keywords

aging in place neighborhood effects caregiving spouse caregivers

Most older adults prefer aging in their homes even when they have difficulties in everyday activities or cognitive functioning to maintain their independence, autonomy, and connection to friends and family (Wiles et al., 2012). Wiles et al. (2012) summarized that older adults gain the advantages of having a sense of attachment or connection and a feeling of security and familiarity with home and community. However, the spouses of those older adults suffering from chronic health conditions and difficulties in activities would be an entirely different story: despite living in the same home and community as their spouses, they are likely to be at risk of caregiver burden, which commonly manifests as health deterioration (Norton et al., 2010; Schulz & Beach, 1999). Then, is living in the community only beneficial for the care-recipient older adults? What benefits can caregiving spouses receive from connection to the community? In the current study, we focus on the subjective assessment of older adult caregiver spouses regarding their neighborhood’s extended and potential social support and the effects it has on their physical and mental health outcomes.

Pearlin et al.’s (1990) stress process model outlines four main areas that contribute to caregiver stress: background context (e.g., caregiving history, family, and network composition), primary stressors (e.g., patients’ cognitive status and problematic behaviors), secondary role strains (e.g., family conflict and economic problems), and secondary intrapsychic strains (e.g., self-esteem, loss of self). This model highlights the roles of internal (e.g., personality, spirituality) and external resources (e.g., social support, knowledge of illness) as moderators between the effects of stressors and outcomes. In the current study, we examine the effects of neighborhood cohesion (NC)—as a potential type of social support—on caregiver spouse’s health outcomes.

Studies on caregiver burden showed that spousal caregiving has deleterious effects on physical and psychological health outcomes such as physical functioning, cognition, depressive symptoms, obesity, and even mortality (Girgis et al., 2013; Nijboer et al., 2001; Norton et al., 2010; Trudeau-Hern & Daneshpour, 2012; Vitaliano et al., 2003). For example, Schulz and Beach (1999) found that older spousal caregivers living with a spouse with mental and emotional strains had a 63% greater risk of mortality within 4 years than noncaregivers.

Research has consistently reported the benefits of social support for lowering the caregiver burden (Chiou et al., 2009; Zhang et al., 2014). For example, Chiou et al. (2009) reported that family caregivers experienced lower levels of caregiver burden when they received more informational, instrumental, and emotional social support from others. Social support promotes recipients’ resilience to stress via psychological and behavioral mechanisms such as motivation to adopt healthy behaviors and reduce risky ones; feeling of being understood; positive appraisal of stressful events; increased self-esteem; and use of active coping strategies (Southwick et al., 2016). However, most studies focused on the social support from family members or friends largely ignore the effect of social context on caregiver burden. In the current study, we examine the influence of neighborhood characteristics, especially NC, as an extended concept of social support on spousal caregivers’ health outcomes.

It is well recognized that the illness of a spouse is a stressful event, and providing subsequent care can bring significant burdens. Research on caregivers’ health consistently reports that spousal caregivers more so than noncaregivers face higher risks of physical and mental health problems such as poor physical functioning (Haley et al., 2000; Schulz & Sherwood, 2008), depression (Haley et al., 2000; Lavela & Ather, 2010; Schulz & Sherwood, 2008), and a decline in cognitive functioning (de Vugt et al., 2006; Lavela & Ather, 2010).

Whereas some caregivers experience a decline in physical and mental health, others manage to avoid them. Stress process models have been widely used to examine the risk and protective factors for caregiver well-being and found that caregivers’ psychosocial resources such as adaptive caregiving appraisals, personality, coping responses, and social supports strongly predicted caregivers’ well-being (e.g., Aneshensel et al., 1995; Chwalisz, 1996; Haley et al., 2003; Lawton et al., 1989; Pakenham, 1999). Haley et al. (2000) for example, supported this theory by reporting that satisfaction with social support and a greater number of social activities and visits with family and friends were related to lower depression of caregivers.

However, these studies explained the effects of social supports only from direct and proximal social networks, without considering the effects of the neighborhood where older adults lived. NC represents cognitive social capital, such as trust and norms, that individuals can access via social connections and networks in their neighborhood (Cramm et al., 2013; Cutrona et al., 2000; Obasaju et al., 2009), and many studies have demonstrated that community cohesion has direct effects on older adults’ health outcomes or buffers the effect from stress (Fone et al., 2007; Gary et al., 2007; Kawachi et al., 1999; Yen et al., 2009). Therefore, as extended social support, older adult caregivers may benefit from NC.

NC works as a form of collective efficacy, that is, a common belief or expectation of controls over their social environment (Bandura, 2000; Sampson et al., 1997). Bandura suggests that perceived collective efficacy plays a key role in human functioning because it affects human behaviors both directly and indirectly via goal setting, outcome expectations, and perception of impediments and opportunities in the environment. Previous studies presented that collective efficacy was strongly associated with health outcomes (Kawachi & Berkman, 2000; Sampson et al., 1997). Kawachi and Berkman (2000) suggest that social capital influences individuals’ (1) health-related behaviors such as exchanging health information and the use of preventive services; (2) access to services and amenities, and; (3) psychosocial process in which individuals experience increased self-esteem and mutual respect.

Despite its potential, NC did not draw much attention in the context of caregiving for older adults. To our best knowledge, only one study tested the effect of the neighborhood on caregivers’ health. Brummett et al. (2005) reported from the same number of caregivers and noncaregiver participants that caregivers with worse neighborhood characteristics than noncaregivers showed poorer glucose functioning—a higher risk factor for diabetes, increased mortality, and cardiovascular disease. Therefore, in the current study, we contribute to the caregiving literature by examining the buffering effect of NC on caregivers’ health outcomes. Additionally, we also test the gender differences regarding the effect of a neighborhood which, according to previous research, was that male spousal caregivers experience significantly lower stress from caregiving (Pinquart & Sörensen, 2006; Thompson et al., 2004). In sum, we hypothesized that (H1) caregiving for a spouse with dementia or activities of daily living (ADL) difficulties would predict older adults’ poor health outcomes; (H2) a higher level of NC would predict better health outcomes, and; (H3) NC would buffer the negative effect of spousal caregiving. Additionally, we examined if there were gender differences in the main effects of NC on older adults’ health and in the buffering effects of NC on caregivers’ health outcomes. Figure 1 illustrates the conceptual model of the present study.

Figure 1

Conceptual model of the moderation effects of neighborhood cohesion on the associations between caregiving and health outcomes.

Method

Data for the current study were collected from the Health and Retirement Study (HRS). The University of Michigan started the HRS in 1992 with a sponsor from the National Institute on Aging. The HRS collects data every second year from the participants with regular questionnaires, and some special modules are used between waves. Data collection was conducted through face-to-face interviews, self-administered questionnaires, and follow-up telephone interviews every 2 years. Questions about community cohesion were included in the self-administered Psychosocial and Lifestyle Questionnaire, and participants respond every 4 years. In the current study, we selected two recent waves (i.e., 2010 and 2012) of data from community-dwelling older adults who were living with a spouse. At the baseline, we selected older adult heterosexual couples who (1) lived by themselves in a community; (2) stayed in the same marital (partnered) relationship; (3) were ages 50 years old or above (both respondents and spouse/partner). We dropped participants who did not respond to questions on community cohesion in 2010. A total of 3329 (1654 men and 1675 women) community-dwelling adults’ data were collected. The participants comprised 2931 Caucasians (88.0%), 257 African Americans (7.7%), and 140 other races (4.2%). Most of them (86.0% for men and 89.6% for women) received at least a high school education. The median income was US$54,496 (SD = US$80,597). We assumed that community-living respondents were providing at least a minimum level of care to their spouse if their spouse was diagnosed with dementia by a doctor or the spouse needed any help in ADL.

Measures

Dependent Variables

Participants’ functioning limitations were measured by ADL and instrumental ADL (IADL; Saliba et al., 2000). Self-reported items on ADL limitation included bathing, dressing, eating, toileting, and getting out of bed (0 = no limitation, 1 = limitation). Self-reported IADL items included preparing meals, managing money, house or yard work, and taking medications (0 = no limitation, 1 = limitation). The summary scores for ADL and IADL were computed, and a higher score indicated more difficulties.

Caregivers’ depressive symptoms were measured at t1 and t2 by the eight-item Center for Epidemiological Studies depression scale (CES-D; Radloff, 1977). The eight items include six negative questions (i.e., depressed, effort, restless, lonely, sad, and could not go on) and two positive items (i.e., happy, enjoyed life). All the items were dichotomous questions (1 = yes, 0 = no). Positive items were reverse-coded so that the high summary score of the items indicate more depressive symptoms. Finally, all the scores were summarized for future use. Cronbach’s α was .79.

The HRS assessed respondents’ cognitive functioning utilizing a measure developed based on the telephone screening instrument (Brandt et al., 1988) and the Mini-Mental Status Exam (Folstein et al., 1975). The measure included word recall and mental status items: word recall items comprised both immediate and delayed recall, and the composite scores ranged from 0 to 20; mental status items included serial 7s test, backward count from 20, naming (objects, date, president, and vice-president). Its possible range was from 0 to 15. Therefore, the total summary score of the recall and mental status indices ranged from 0 to 35.

Independent Variables

Perceived NC was measured by simplified four-item neighborhood social cohesion scales (Stafford et al., 2003). The original scale included two aspects of neighborhood social cohesion—structural and cognitive. The HRS focused on the cognitive aspect and chose four items: “I really feel part of this area,” “Most people in this area can be trusted,” “If you were in trouble, there are lots of people in this area who would help you,” and “Most people in this area are friendly.” According to the original scale, the first two and the latter two items show the respondents’ attachment and trust to their neighborhood, respectively. Respondents used a seven-point Likert scale to indicate the degree to which they agree to the following statements. Because responses were positively skewed, we applied a log-transformation for the summary scores of the four items. Cronbach’s α was .86.

Covariates

Older adults’ age when the interview began was calculated in years using their birth year. Dummy coding was applied to respondents’ race (Black and others, reference = White) and education (less than high school and more than high school, reference = high school). The total household income consisted of employment-related wages, salaries, bonuses, all welfare benefits, such as Social Security benefits, Supplemental Security Income, unemployment compensation, and all incomes derived from assets.

Analytic Strategy

We reported descriptive characteristics of respondents related to their health outcomes, caregiving contexts, and control variables. Results of t-test and χ² tests were reported to compare the differences in the study variables. Separate sets of correlations among variables for men and women were also computed. We used ordinary least square regression analyses to regress each health outcome on NC, caregiving contexts, and control variables. We analyzed the data separately for men and women for the two following reasons: first, in the HRS, quite a number of the men and women respondents came from the same household, which violates the independence assumption; second, previous research suggested that there is a gender difference in caregiving roles and the effects of neighborhood experiences (Pinquart & Sörensen, 2006; Schieman & Meersman, 2004). We multiplied NC by caregiving context such as caregiving a spouse with dementia and providing ADL help to the spouse to create interaction terms. We also controlled for the baseline effects of the health outcomes (i.e., ADL and IADL difficulties, depressive symptoms, and cognition) to assess the longitudinal effects of the predictors. In separate analyses, we tested the models without interaction terms first and then included interaction terms. There was sample attrition as is common for longitudinal studies. The missing ratio due to nonresponse or death was 6.3% for 2012. The correlations of nonresponse with other study variables showed that likelihood of missing was associated with older age (r = .13, p < .001), being male (r = .08, p < .001), less than a high school education (r = .05, p < .001), and more than a high school education (r = −.04, p = .013). Because these results met the assumption of the missing at random (MAR), that missing data can be predicted from available data, we applied a full information maximum likelihood (Newman, 2003). Because only half of the participants responded to the Psychosocial and Lifestyle Questionnaire (PLQ) each wave, all the analyses were weighted using the psychosocial weight.

Results

As shown in Table 1, men, as compared to women, were in poorer physical health in categories such as ADL and IADL, whereas women experienced more depressive symptoms. In the case of a community-living couple-only household, women were more, although marginally, likely than men to serve their spouse with dementia. Table 2 shows the correlations among the study variables. For men, better NC was associated with better health outcomes such as fewer ADL and IADL difficulties, fewer depressive symptoms, and better cognitive function. Men who were providing ADL help to their spouse had more IADL difficulties and poorer cognitive functioning. For women, NC was associated with fewer ADL/IADL difficulties and depressive symptoms, and better and cognitive function. Providing ADL help to their spouse was associated with more depressive symptoms.

Table 1

Descriptive Characteristics.

	Men(N = 1564)		Women(N = 1675)		p
	Mean	SD	Mean	SD	p
ADL difficulties 2012	0.25	0.77	0.19	0.70	.019
ADL difficulties 2010	0.23	0.70	0.19	0.67	.088
IADL difficulties 2012	0.12	0.44	0.07	0.35	.001
IADL difficulties 2010	0.11	0.42	0.06	0.31	<.001
CES-D 2012	0.96	1.56	1.19	1.82	<.001
CES-D 2010	0.88	1.47	1.15	1.76	<.001
Cognition 2012	21.81	4.66	23.11	4.67	<.001
Cognition 2010	21.89	4.55	23.15	4.37	<.001
Neighborhood cohesion (NC)	3.07	0.31	3.08	0.30	.421
Spouse with dementia (%)	1.00		1.50		.084
ADL help to spouse (%)	2.70		3.40		.153
Age	69.94	9.43	66.80	9.18	<.001
Race/ethnicity (%)					.318
White	91.00		92.80
Black	6.30		5.30
Other	2.70		1.90
Education (%)					<.001
<High school	14.70		11.50
High school	32.70		42.50
>High school	52.70		46.00
Household income (US$1,000)	78.15	79.96	79.56	81.24	.611
Household asset (US$1,000)	648.94	908.61	651.53	997.68	.935

Note. ADL = activities of daily living; CES-D = Center for Epidemiological Studies depressionscale; IADL = instrumental activities of daily living.

Table 2

Correlations Among Major Study Variables.

	1	2	3	4	5	6	7	8	9	10	11
1.ADL difficulties 2012	−	+.65^***	+.45^***	+.31^***	+.31^***	+.28^***	−.20^***	−.22^***	−.09^***	+.001	+.02
2.ADL difficulties 2010	+.63^***	−	+.36^***	+.37^***	+.25^***	+.29^***	−.16^***	−.17^***	−.12^***	+.03	+.05
3.IADL difficulties 2012	+.50^***	+.33^***	−	+.50^***	+.25^***	+.25^***	−.33^***	−.30^***	−.05	−.01	−.01
4.IADL difficulties 2010	+.34^***	+.37^***	+.54^***	−	+.20^**	+.24^**	−.21^***	−.19^**	−.07^**	−.01	−.004
5.CES-D 2012	+.38^***	+.31^***	+.31^***	+.29^***	−	+.53^**	−.18^**	−.20^**	−.18^**	+.04	+.10^**
6.CES-D 2010	+.34^***	+.33^***	+.25^***	+.34^***	+.58^***	−	−.21^**	−.2.3^**	−.20^**	+.04	+.01
7.Cognition 2012	−.27^***	−.18^***	−.33^***	−.31^***	−.26^***	−.23^***	−	+.63^**	+.08^*	+.05	+.03
8.Cognition 2010	−.23^***	−.19^***	−.32^***	−.32^***	−.17^***	−.23^***	+.69^***	−	+.08^**	+.04	+.05
9.Neighborhood cohesion	−.09^**	−.09^**	−.06^*	−.11^**	−.21^***	−.18^***	+.14^***	+.15^***	−	+.01	−.03
10.Spouse with dementia	−.03	−.02	−.03	−.03	+.003	+.004	+.00	−.02	+.02	−	+.20^**
11.ADL help to spouse	+.04	+.01	+.06^*	+.03	+.05	+.02	−.02	−.06^*	−.03	+.10^***	−

Note. ADL = activities of daily living; CES-D = Center for Epidemiological Studies depression scale; IADL = instrumental activities of daily living.

Coefficients for husbands are shown below the diagonal and for wives above the diagonal.

^*<.05; ^**<.01; ^***<.001.

Table 3 shows the results of the regression models estimated to predict health outcomes for men after controlling for the effects of the corresponding baseline health outcomes for each model. Better NC predicted fewer depressive symptoms and better cognitive function. The “NC × ADL help” interaction term was significant (b = 5.47, SE = 2.75, p = .047), indicating that better NC may be more beneficial to male older adults living with ADL difficulties. However, none of the interactions (i.e., NC × SP with dementia and NC × ADL help) were statistically significant for the other health outcomes. Thus, the association of NC with ADL and IADL difficulties and depressive symptoms did not vary by their status of caregiving roles.

Table 3

Health Outcomes Regressed on Neighborhood Cohesion, Caregiving, and Interactions (Men Only).

Variable	ADL difficulties		IADL difficulties		CES-D		Cognition
Baseline health outcome	.72^*** (.03)	.72^*** (.03)	.60^*** (.03)	.60^*** (.03)	.62^*** (.03)	.62^*** (.03)	.60^*** (.03)	.60^*** (.03)
NC	−.08(.07)	−.09(.07)	−.03(.04)	−.03(.04)	~−.57^*** (.16)	~−.58^*** (.16)	1.26^*** (.36)	1.21^** (.36)
SP with dementia	−.11(.19)	.62(2.98)	−.06(.12)	⁺ .36(1.82)	−.12(.44)	−7.04(6.78)	0.87(1.02)	26.70(15.64)
ADL help to SP	.04(.15)	−1.65(1.59)	⁺.04(.09)	⁺ −.51(.97)	⁺.19(.35)	~−1.08(3.62)	.95(.80)	−15.62^† (8.35)
Interactions
NC × SP with dementia		−.24(.96)		−.14(.59)		⁺2.23(2.19)		−8.35(5.04)
NC × ADL help to SP		⁺.56(.52)		⁺ .18(.32)		.42(1.19)		5.47*(2.75)

Control variables
Age	.002(.002)	.002(.002)	+.003^* (.001)	⁺ .003^* (.001)	~−.01(.01)	~−.01(.01)	−.10^*** (.02)	−.10^*** (.01)
Black	−.09(.10)	⁺ −.08(.08)	⁺.06(.06)	⁺ .06(.06)	~−.16(.23)	~−.16(.23)	−.68(.53)	.73(.53)
Other race	.10(.09)	.10(.09)	⁺.12^* (.06)	⁺ .12^* (.06)	⁺~.52^* (.21)	⁺~.51^* (.21)	.42(.49)	.42(.49)
<High school	⁺.002(.07)	⁺ .001(.07)	⁺ −.03(.04)	^-−.03(.04)	.11(.16)	.11(.16)	−.74^* (.37)	−.75^* (.37)
>High school	−.05(.04)	−.05(.04)	−.02(.03)	−.02(.03)	~−.34^** (.10)	~−.34^** (.10)	−.11(.23)	−.13(.23)
Income (US$1000)	<−.001(<.001)	<−.001(<.001)	−.001^* (<.001)	<−.001^* (<.001)	<−.001(.001)	<−.001(.001)	.001(.001)	.001(.001)
Log asset	−.01(.02)	−.01(.02)	.02^* (.01)	.02^* (.01)	.02(.03)	.02(.03)	.21^** (.08)	.21^** (.08)

Constant	⁺.37	⁺ .39	⁺−.26	−.26	2.49	^~4.69	9.18	9.30
R ²	⁺.40	⁺ .40	⁺ .30	⁺ .30	⁺.38	~^~.38	.52	.52

Note. ADL = activities of daily living; CES-D = Centerfor Epidemiological Studies depression scale; IADL = instrumental activities of daily living; NC = neighborhood cohesion.

Unstandardized regression coefficients with standard errors in parentheses. Baseline health outcome is the corresponding outcome variable measured in 2010 for each model. N = 1680.

^†< .10; ^*< .05; ^**< .01; ^***< .001.

Related to the findings from control variables, older age predicted more IADL difficulties, whereas it was negatively associated with cognition. Compared with Caucasians, men of other races reported more IADL difficulties and depressive symptoms. More education was associated with fewer depressive symptoms and better cognitive function. More income predicted fewer IADL difficulties. More asset was associated with more IADL difficulties and better cognition.

Table 4 presents the results from the regression models estimated to predict health outcomes for women. NC in 2010 predicted better health outcomes in 2012, such as fewer depressive symptoms and better cognition. Providing ADL help to a husband marginally predicted more depressive symptoms. The model with interaction terms showed that NC marginally buffered the negative effect of providing ADL help to a husband on their ADL difficulties.

Table 4

Health Outcomes Regressed on Neighborhood Cohesion, Caregiving, and Interactions (Women Only).

Variable	ADL difficulties		IADL difficulties		CES-D		Cognition
Baseline health outcome	.69^*** (.03)	.70^*** (.03)	.52^*** (.03)	.52^*** (.03)	.52^*** (.03)	.52^*** (.03)	.56^*** (.03)	.56^*** (.03)
NC	.04(.06)	.05(.06)	−.01(.03)	−.01(.03)	~−.47^* (.19)	~−.43^* (.20)	.72^† (.38)	.79^* (.39)
SP with dementia	−.28(.16)	−5.55^** (2.05)	.02(.09)	.24(1.09)	−.08(.49)	−3.03(6.14)	1.86(1.00)	.60(12.44)
ADL help to SP	−.05(.12)	1.97^† (1.05)	−.09(.07)	−.45(.56)	⁺.67^† (.37)	4.58(3.16)	.68(.76)	7.94(6.40)
Interactions
NC × SP with dementia		1.02(.66)		−.07(.35)		.97(1.99)		.42(4.04)
NC × ADL help to SP		−.67^† (.34)		.12(.18)		−1.29(1.03)		−2.39(2.09)
Control variables
Age	.004(.002)	.003(.002)	+.003^** (.001)	+.003^** (.001)	~−.004(.01)	~−.005(.01)	−.08^*** (.01)	−.08^*** (.01)
Black	.10(.09)	.09(.09)	⁺−.03(.05)	⁺−.03(.05)	.16(.28)	.15(.28)	−.75(.57)	−.75(.57)
Other race	.14(.10)	.14(.10)	⁺.05^* (.05)	⁺.05(.05)	⁺.38^* (.29)	⁺.38^* (.29)	.02(.59)	.03(.59)
<High school	⁺.17(.07)	⁺.17^* (.07)	.13^** (.04)	.13^** (.04)	.05(.22)	.05(.22)	−.69^* (.45)	−.70^* (.45)
>High school	.07(.04)	.07^† (.04)	−.01(.02)	−.01(.02)	~−.10(.12)	~−.10(.12)	.40^† (.24)	.40^† (.24)
Income (US$1000)	<−.001^* (<.001)	<−.001^* (<.001)	−.001(<.001)	.001(<.001)	−.002^* (.001)	−.002^* (.001)	−.001(.001)	−.001(.001)
Log asset	.02(.01)	.02(.01)	−.03^*** (.01)	−.03^*** (.01)	.03(.04)	.03(.04)	.17^* (.09)	.18^* (.09)

Constant	⁺−.55	⁺−.58	⁺.19	⁺.20	2.15	2.06	11.07	10.83
R ²	⁺.43	⁺.44	⁺ .30	⁺ .30	⁺.30	⁺.30	.42	.42

Note. ADL = activitiesof daily living; CES-D = Center for Epidemiological Studies depression scale; IADL = instrumental activities of daily living; NC = neighborhood cohesion; SP = spouse.

Unstandardized regression coefficients with standard errors in parentheses. Baseline health outcome is the corresponding outcome variable measured in 2010 for each model. N = 1702.

^†< .10; ^*< .05; ^**< .01; ^***< .001.

Women’s older age was associated with more IADL difficulties and poorer cognitive functioning. Women of other races besides White or Black had more IADL difficulties and depressive symptoms. More education was associated with fewer ADL/IADL and better cognitive functioning. Participants’ asset was associated with fewer IADL difficulties and better cognitive functioning.

Discussion

The purpose of this study was to examine the association of NC with community-living older adult spousal caregivers’ health using nationally representative data from the Health and Retirement Study. Our findings revealed that spousal caregiving and NC are associated with older adults’ health outcomes. More importantly, NC moderated the caregiving effects on health outcomes such as physical (i.e., ADL difficulties) and cognitive functioning. This study contributes to the literature with new evidence not only about the positive effect of the neighborhood on older adults’ health but also its positive influences in the context of caregiving. The gender differences in those effects found in the current study also add values to the literature on caregiving and the theories on social capital.

The negative main effects of caregiving on older adults’ health outcomes found in the current study were consistent with previous findings (Haley et al., 2000; Schulz & Sherwood, 2008; Lavela & Ather, 2010). However, we additionally found differential effects of caregiving on health by gender: male caregiver spouses had more ADL, IADL, and cognitive difficulties but fewer depressive symptoms than female caregivers; men who provided ADL help to spouse experienced more difficulties in cognitive functioning, whereas female caregivers experienced more ADL difficulties and depressive symptoms. These findings on the negative effects of caregiving on both male and female caregivers’ health are consistent with previous reports that female caregivers experience more caregiver burden and depression but are better off in certain aspects of well-begin such as subjective well-being and physical health (Pinquart & Sörensen, 2006). Women are more likely than men to assume caregiver role, but once men (i.e., husbands) take the caregiver role, men may become more vulnerable in some dimensions in well-being than women caregivers.

As expected, NC buffered the negative effects of spousal caregiving: NC played a protector role in caregivers’ physical and cognitive functioning. Regarding these main effects of caregiving, there were gender differences in the effects of NC—cognitive functions for men and mental health for women.

We can draw some policy implications from the findings of this study. First, at a time where caregiving from family members has been stretched for quite some time due to multiple reasons such as lower birth rates, labor mobility, and higher rates of divorce, this study demonstrates the importance to sustain and improve programs that enhance social networks and interactions at the local level. Policy discussions on aging tend to focus on Social Security and Medicare, often at the expense of community programs supported by the Older Americans Act (OAA). Funding is extremely limited and the gap between the original policy objectives of the OAA and its means continues to be substantial (Hudson & Kingson, 1991). Besides, access to OAA programs is increasingly linked to medical conditions (Vladeck, 2004). This contribution emphasizes the importance of remaking community programs and social services within OAA and at the local level a priority.

Second, with a highly fragmented network of supports for seniors with regard to both the type of providers (e.g., municipalities, private companies, nonprofit association, etc.) and the kinds of services offered, clinicians continue to find it difficult to refer older adults to the appropriate community-based support and services (Siegler et al., 2015). With our study emphasizing the importance of communities and, as a result, community programs, bridging this knowledge gap becomes even more imperative.

There are limitations to be considered when interpreting the results of the current study. First, the measurement of perceived NC partially reflects individuals’ personality traits, which might have influenced both the view of their neighborhood and health status. For example, more extraverted and less neurotic people tend to have a more positive attitude toward others (Bolger & Eckenrode, 1991; Von Dras & Siegler, 1997) and better health status (Martin et al., 2006; Turiano et al., 2012). Although personality measurements were available in the HRS, because half of the sample was missing by design, including personality variables would have decreased the number of current study’s community-dwelling spousal caregivers by half. Therefore, the future study needs to include more objective measurement of NC and consider the effects of participants’ personality traits. Second, there were limitations in the data. Although we included two waves of data, our study could not fully overcome the limitations of a cross-sectional design. With more than two waves’ design, we were not able to collect a good number of older adult spousal caregivers living in the community. Utilizing caregiving-oriented data with more waves including a precaregiving data-point would provide us with better causal relationships among caregiving, neighborhood characteristics, and caregivers’ health. Third, we acknowledge that more information on a caregiving context should be considered. For example, caregiving roles were measured only by binary coding, so the degrees of care-recipients’ care needs were not well-reflected. Formal home care service and informal help caregivers utilized or received were not considered, which might have influenced the caregiver burden (Chen et al., 2017). Finally, care-recipients’ spouses may not necessarily be primary caregivers. We reasoned that if a couple lives in a community when one person is demented or needs ADL help, the spouse would take at least a minimal role of caregiver. However, they might have other primary caregivers such as children or professional caregivers.

Despite these limitations, this study provides evidence that NC is associated not only with general older adults’ health outcomes but is also especially important for spousal caregivers to buffer the negative health effect of caregiving. Given that the population is aging, and older adults want to age in a community, it is important to develop public policies that enhance the positive characteristics of the neighborhood to help the caregivers maintain a healthy life. Further research is recommended to delineate the mechanism of the positive association of neighborhood with caregivers’ health outcomes as well as the determinants of community cohesion.

Footnotes

Declaration of Conflicting Interests

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

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This research was supported by the Daegu University Research Grant, 20190300.

ORCID iD

Kyuho Lee

Author Biographies

Kyuho Lee, PhD, is an assistant professor in the Department of Family Welfare at Daegu University, South Korea. He is also a director of the WePlus Family Institute. His research interests are in older adults’ physical and psychological well-being, social engagement, caregiving, social support, and retirement.

Patrik Marier, PhD, is a professor of Political Science and a Concordia University Research Chair (CURC) in Aging and Public Policy. He is also the Scientific Director of the Center for Research and Expertise in Social Gerontology (CREGÉS). His research focuses on the impact of aging populations on a number of public policy fields including pensions, labor, and social services and programs across comparative cases.

References

Aneshensel

C. S.

Pearlin

Mullan

J. T.

Zarit

S. H.

Whitlatch

C. J

. (1995). Profiles in caregiving: The unexpected career. Academic Press.

Bandura

. (2000). Exercise of human agency through collective efficacy. Current Directions in Psychological Science, 9(3), 75–78.doi:10.1111/1467-8721.00064

Bolger

Eckenrode

. (1991). Social relationships, personality, and anxiety during a major stressful event. Journal of Personality and Social Psychology, 61(3), 440–449.doi:10.1037/0022-3514.61.3.440

http://www.ncbi.nlm.nih.gov/pubmed/1941515

Brandt

Spencer

Folstein

. (1988). The telephone interview for cognitive status. Neuropsychiatry, Neuropsychology Behavioral Neurology, 1(2), 111–117.

Brummett

B. H.

Siegler

I. C.

Rohe

W. M.

Barefoot

J. C.

Vitaliano

P. P.

Surwit

R. S.

Feinglos

M. N.

Williams

R. B

. (2005). Neighborhood characteristics moderate effects of caregiving on glucose functioning. Psychosomatic Medicine, 67(5), 752–758.doi:10.1097/01.psy.0000174171.24930.11

http://www.ncbi.nlm.nih.gov/pubmed/16204434

Chen

M.-C.

Kao

C.-W.

Chiu

Y.-L.

Lin

T.-Y.

Tsai

Y.-T.

Jian

Y.-T. Z.

Tzeng

Y.-M.

Lin

F.-G.

Hwang

S.-L.

S.-R.

Kao

. (2017). Effects of home-based long-term care services on caregiver health according to age. Health and Quality of Life Outcomes, 15(1), 208.doi:10.1186/s12955-017-0786-6

29061145

Chiou

C. J.

Chang

H.-Y.

Chen

I. P.

Wang

H. H

. (2009). Social support and caregiving circumstances as predictors of caregiver burden in Taiwan. Archives of Gerontology and Geriatrics, 48(3), 419–424.doi:10.1016/j.archger.2008.04.001

18602706

Chwalisz

. (1996). The perceived stress model of caregiver burden: Evidence from spouses of persons with brain injuries. Rehabilitation Psychology, 41(2), 91–114.doi:10.1037/0090-5550.41.2.91

Cramm

J. M.

van Dijk

H. M.

Nieboer

A. P

. (2013). The importance of neighborhood social cohesion and social capital for the well being of older adults in the community. The Gerontologist, 53(1), 142–152.doi:10.1093/geront/gns052

http://www.ncbi.nlm.nih.gov/pubmed/22547088

10.

Cutrona

C. E.

Russell

D. W.

Hessling

R. M.

Brown

P. A.

Murry

. (2000). Direct and moderating effects of community context on the psychological well-being of African American women. Journal of Personality and Social Psychology, 79(6), 1088–1101.doi:10.1037/0022-3514.79.6.1088

http://www.ncbi.nlm.nih.gov/pubmed/11138756

11.

de Vugt

M. E.

Jolles

van Osch

Stevens

Aalten

Lousberg

Verhey

F. R. J

. (2006). Cognitive functioning in spousal caregivers of dementia patients: Findings from the prospective MAASBED study. Age and Ageing, 35(2), 160–166.doi:10.1093/ageing/afj044

http://www.ncbi.nlm.nih.gov/pubmed/16495293

12.

Folstein

M. F.

Folstein

S. E.

McHugh

P. R

. (1975). Mini-mental state: A practical method for grading the cognitive state of patients for the clinician. Journal of Psychiatric Research, 12(3), 189–198.doi:10.1016/0022-3956(75)90026-6

http://www.ncbi.nlm.nih.gov/pubmed/1202204

13.

Fone

Dunstan

Lloyd

Williams

Watkins

Palmer

. (2007). Does social cohesion modify the association between area income deprivation and mental health? A multilevel analysis. International Journal of Epidemiology, 36(2), 338–345.doi:10.1093/ije/dym004

http://www.ncbi.nlm.nih.gov/pubmed/17329315

14.

Gary

T. L.

Stark

S. A.

LaVeist

T. A

. (2007). Neighborhood characteristics and mental health among African Americans and whites living in a racially integrated urban community. Health & Place, 13(2), 569–575.doi:10.1016/j.healthplace.2006.06.001

http://www.ncbi.nlm.nih.gov/pubmed/16904931

15.

Girgis

Lambert

Johnson

Waller

Currow

. (2013). Physical, psychosocial, relationship, and economic burden of caring for people with cancer: A review. Journal of Oncology Practice, 9(4), 197–202.doi:10.1200/JOP.2012.000690

http://www.ncbi.nlm.nih.gov/pubmed/23942921

16.

Haley

W. E.

LaMonde

L. A.

Han

Burton

A. M.

Schonwetter

. (2003). Predictors of depression and life satisfaction among spousal caregivers in hospice: Application of a stress process model. Journal of Palliative Medicine, 6(2), 215–224.doi:10.1089/109662103764978461

http://www.ncbi.nlm.nih.gov/pubmed/12854938

17.

Haley

W. E.

LaMonde

L. A.

Han

Narramore

Schonwetter

. (2000). Family caregiving in hospice: Effects on psychological and health functioning among spousal caregivers of hospice patients with lung cancer or dementia. The Hospice Journal, 15(4), 1–18.doi:10.1080/0742-969X.2000.11882959

18.

Hudson

R. B.

Kingson

E. R

. (1991). Inclusive and fair: The case for universality in social programs. Generations, 15(3), 51–56.

19.

Kawachi

Berkman

. (2000). Social cohesion, social capital, and health. Social Epidemiology, 174(7), 174–190.

20.

Kawachi

Kennedy

B. P.

Glass

. (1999). Social capital and self-rated health: A contextual analysis. American Journal of Public Health, 89(8), 1187–1193.doi:10.2105/AJPH.89.8.1187

http://www.ncbi.nlm.nih.gov/pubmed/10432904

21.

Lavela

S. L.

Ather

. (2010). Psychological health in older adult spousal caregivers of older adults. Chronic Illness, 6(1), 67–80.doi:10.1177/1742395309356943

http://www.ncbi.nlm.nih.gov/pubmed/20308352

22.

Lawton

M. P.

Kleban

M. H.

Moss

Rovine

Glicksman

. (1989). Measuring caregiving appraisal. Journal of Gerontology, 44(3), P61–P71.doi:10.1093/geronj/44.3.P61

http://www.ncbi.nlm.nih.gov/pubmed/2715587

23.

Martin

da Rosa

Siegler

I. C

Davey

Macdonald

Poon

L. W

Georgia Centenarian Study . (2006). Personality and longevity: Findings from the Georgia centenarian study. Age, 28(4), 343–352.doi:10.1007/s11357-006-9022-8

22253500

24.

Newman

D. A

. (2003). Longitudinal modeling with randomly and systematically missing data: A simulation of AD hoc, maximum likelihood, and multiple imputation techniques. Organizational Research Methods, 6(3), 328–362.doi:10.1177/1094428103254673

25.

Nijboer

Tempelaar

Triemstra

Sanderman

van den Bos

G. A. M

. (2001). Dynandcs in cancer caregiver’s health over time: Gender-specific patterns and determinants. Psychology & Health, 16(4), 471–488.doi:10.1080/08870440108405520

26.

Norton

M. C.

Smith

K. R.

Østbye

Tschanz

J. T.

Corcoran

Schwartz

Breitner

. (2010). Greater risk of dementia when spouse has dementia? The Cache County study. Journal of the American Geriatrics Society, 58(5), 895–900.doi:10.1111/j.1532-5415.2010.02806.x

20722820

27.

Obasaju

M. A.

Palin

F. L.

Jacobs

Anderson

Kaslow

N. J

. (2009). Won’t you be my neighbor? Using an ecological approach to examine the impact of community on revictimization. Journal of Interpersonal Violence, 24(1), 38–53.doi:10.1177/0886260508314933

http://www.ncbi.nlm.nih.gov/pubmed/18319364

28.

Pakenham

. (1999). Adjustment to multiple sclerosis: Application of a stress and coping model. Health Psychology, 18(4), 383–392.doi:10.1037/0278-6133.18.4.383

http://www.ncbi.nlm.nih.gov/pubmed/10431940

29.

Pearlin

L. I.

Mullan

J. T.

Semple

S. J.

Skaff

M. M

. (1990). Caregiving and the stress process: An overview of concepts and their measures. The Gerontologist, 30(5), 583–594.doi:10.1093/geront/30.5.583

http://www.ncbi.nlm.nih.gov/pubmed/2276631

30.

Pinquart

Sörensen

. (2006). Gender differences in caregiver stressors, social resources, and health: An updated meta-analysis. The Journals of Gerontology Series B: Psychological Sciences and Social Sciences, 61(1), P33–P45.doi:10.1093/geronb/61.1.P33

http://www.ncbi.nlm.nih.gov/pubmed/16399940

31.

Radloff

L. S

. (1977). The CES-D scale: A self-report depression scale for research in the general population. Applied Psychological Measurement, 1(3), 385–401.

32.

Saliba

Orlando

Wenger

N. S.

Hays

R. D.

Rubenstein

L. Z

. (2000). Identifying a short functional disability screen for older persons. The Journals of Gerontology Series A: Biological Sciences and Medical Sciences, 55(12), M750–M756.doi:10.1093/gerona/55.12.M750

33.

Sampson

R. J.

Raudenbush

S. W.

Earls

. (1997). Neighborhoods and violent crime: A multilevel study of collective efficacy. Science, 277(5328), 918–924.doi:10.1126/science.277.5328.918

http://www.ncbi.nlm.nih.gov/pubmed/9252316

34.

Schieman

Meersman

S. C

. (2004). Neighborhood problems and health among older adults: Received and donated social support and the sense of mastery as effect modifiers. The Journals of Gerontology: Series B, 59(2), S89–S97.doi:10.1093/geronb/59.2.S89

http://www.ncbi.nlm.nih.gov/pubmed/15014096

35.

Schulz

Beach

S. R

. (1999). Caregiving as a risk factor for mortality: The caregiver health effects study. JAMA, 282(23), 2215–2219.doi:10.1001/jama.282.23.2215

http://www.ncbi.nlm.nih.gov/pubmed/10605972

36.

Schulz

Sherwood

P. R

. (2008). Physical and mental health effects of family caregiving. Journal of Social Work Education, 108(9 Suppl), 105–113.doi:10.1097/01.NAJ.0000336406.45248.4c

http://www.ncbi.nlm.nih.gov/pubmed/18797217

37.

Siegler

E. L.

Lama

S. D.

Knight

M. G.

Laureano

Reid

M. C

. (2015). Community-based supports and services for older adults: A primer for clinicians. Journal of Geriatrics, 2015, 1–6.doi:10.1155/2015/678625

http://www.ncbi.nlm.nih.gov/pubmed/25729774

38.

Southwick

S. M.

Sippel

Krystal

Charney

Mayes

Pietrzak

. (2016). Why are some individuals more resilient than others: The role of social support. World Psychiatry, 15(1), 77–79.doi:10.1002/wps.20282

http://www.ncbi.nlm.nih.gov/pubmed/26833614

39.

Stafford

Bartley

Sacker

Marmot

Wilkinson

Boreham

Thomas

. (2003). Measuring the social environment: Social cohesion and material deprivation in English and Scottish neighbourhoods. Environment and Planning A: Economy and Space, 35(8), 1459–1475.doi:10.1068/a35257

40.

Thompson

R. L.

Lewis

S. L.

Murphy

M. R.

Hale

J. M.

Blackwell

P. H.

Acton

G. J.

Clough

D. H.

Patrick

G. J.

Bonner

P. N

. (2004). Are there sex differences in emotional and biological responses in spousal caregivers of patients with Alzheimer’s disease? Biological Research for Nursing, 5(4), 319–330.doi:10.1177/1099800404263288

http://www.ncbi.nlm.nih.gov/pubmed/15068661

41.

Trudeau-Hern

Daneshpour

. (2012). Cancer’s impact on spousal caregiver health: A qualitative analysis in grounded theory. Contemporary Family Therapy, 34(4), 534–554.doi:10.1007/s10591-012-9211-9

42.

Turiano

N. A.

Pitzer

Armour

Karlamangla

Ryff

C. D.

Mroczek

D. K

. (2011). Personality trait level and change as predictors of health outcomes: Findings from a national study of Americans (MIDUS). The Journals of Gerontology. Series B, Psychological Sciences and Social Sciences, 67(1), 4–12.doi:10.1093/geronb/gbr072

http://www.ncbi.nlm.nih.gov/pubmed/21765062

43.

Vitaliano

P. P.

Zhang

Scanlan

J. M

. (2003). Is caregiving hazardous to one’s physical health? A meta-analysis. Psychological Bulletin, 129(6), 946–972.doi:10.1037/0033-2909.129.6.946

http://www.ncbi.nlm.nih.gov/pubmed/14599289

44.

Vladeck

F. W

. (2004). Beyond one hip fracture at a time: Rethinking aging services. Journal of Gerontological Social Work, 42(3–4), 151–162.

45.

Von Dras

D. D.

Siegler

I. C

. (1997). Stability in extraversion and aspects of social support at midlife. Journal of Personality and Social Psychology, 72(1), 233–241.doi:10.1037/0022-3514.72.1.233

http://www.ncbi.nlm.nih.gov/pubmed/9008383

46.

Wiles

J. L.

Leibing

Guberman

Reeve

Allen

R. E. S

. (2012). The meaning of “aging in place” to older people. The Gerontologist, 52(3), 357–366.doi:10.1093/geront/gnr098

http://www.ncbi.nlm.nih.gov/pubmed/21983126

47.

Yen

I. H.

Michael

Y. L.

Perdue

. (2009). Neighborhood environment in studies of health of older adults: A systematic review. American Journal of Preventive Medicine, 37(5), 455–463.doi:10.1016/j.amepre.2009.06.022

http://www.ncbi.nlm.nih.gov/pubmed/19840702

48.

Zhang

Guo

Edwards

Yates

. (2014). Self-efficacy moderation and mediation roles on BPSD and social support influences on subjective caregiver burden in Chinese spouse caregivers of dementia patients. International Psychogeriatrics, 26(9), 1465–1473.doi:10.1017/S1041610214000994

http://www.ncbi.nlm.nih.gov/pubmed/24932533