Relationships between negative exchanges and depressive symptoms in older couples: The moderating roles of care and gender

Abstract

Marital relationships have a significant impact on older adults’ well-being. However, when contending with spousal illness or disability, negative exchanges may be particularly detrimental. This study examines the extent to which negative spousal exchanges have more impact on caregiving versus non-caregiving couples. Using dyadic analyses, this investigation compares three different groups consisting of (a) couples who did not provide or receive care, (b) couples in which husbands received care from their wives, and (c) couples in which wives received care from their husbands. We tested for gender differences in the effect of negative exchanges on depressive symptoms among caregivers and care recipients. Based on the 2012 and 2014 Health and Retirement Study, the sample consisted of 3,530 couples in which at least one of the spouses was aged 51 or over. Structural equation modeling was used to test the moderating effects of care and gender. Sociodemographic characteristics were included as control variables. Results indicated that there were significant differences in the magnitude of the path between negative exchanges and depressive symptoms across the subgroups. Among husbands, there was a stronger relationship between negative exchanges and depressive symptoms for care-receiving husbands than for caregiving husbands or husbands in non-caregiving relationships. Among wives, however, there was no significant difference in the path from negative exchanges with spouses to depressive symptoms across care status. We also found gender differences in the effect of negative exchanges on depressive symptoms among caregivers, but not among care recipients. Findings suggest that care status and gender of caregivers and care recipients have significant effects on the association between negative interactions within couples and mental health outcomes.

Keywords

Caregiving depression gender differences negative interactions older couples

Marital relationships have a strong influence on mental health in later life (Holahan et al., 2007; Williams & Umberson, 2004), particularly when spouses suffer from illness or disability and need care from their partners (Wolff & Kasper, 2006). Chronic illness or disability can result in significant distress and adjustment for both the ill spouse and his or her partner (Berg & Upchurch, 2007; Sabey & Rauer, 2018). Thus, a dyadic perspective helps in understanding the impact of illness and disability on the mental health outcomes of caregivers and care recipients, as it takes spousal influence into account.

Existing literature suggests that interactions between spouses play an important role in both partners’ mental health outcomes in positive and negative ways (Berkman, Glass, Brissette, & Seeman, 2000). Specifically, research on caregiving couples found that positive social exchanges from caregiving spouses, such as emotional support, were related to more positive care recipient outcomes (e.g., adaptive coping behaviors), fewer negative reactions to loss of functional independence, and fewer depressive symptoms (Matire, Stephens, Druley, & Wojno, 2002; Schulz & Schwarzer, 2004). Emotional support from care-receiving spouses also contributes to their caregiving partners’ lower levels of caregiving burden and stress as well as to their greater well-being (Kleiboer, Kuijer, Hox, Schreurs, & Bensing, 2006; Raschick & Ingersoll-Dayton, 2004; Wright & Aquilino, 1998). Conversely, negative social exchanges with spouses (e.g., criticism and rejection) result in maladaptive coping behaviors, increased distress, and more depressive symptoms among care recipients and greater caregiving burden, stress, and depressive symptoms for caregiving partners (Townsend & Franks, 1997).

Previous studies on caregiving couples have focused on the impact of social exchanges with spouses on depressive symptoms at the individual level (Robles, Slatcher, Trombello, & McGinn, 2014). For married individuals, however, social exchanges between spouses influence each spouse’s depressive symptoms through both intrapersonal and interpersonal processes (Choi, Yorgason, & Johnson, 2015). A growing body of literature emphasizes the importance of examining interpersonal processes to understand the effects of social exchanges with a spouse on both partners’ well-being (Peterson & Smith, 2010). That is, spouses’ perceived quality of social exchanges with their partners affects their own depressive symptoms as well as their partners’ depressive symptoms. In this study, we used a dyadic actor–partner interdependence approach to model the interpersonal effects of negative exchanges with spouses on depressive symptoms. In doing so, the model estimated how we conceptualized that the interpersonal effects of the focal person’s own perception of negative exchanges with his/her spouse would affect the person’s own depressive symptoms (actor effect) as well as the partner’s depressive symptoms (partner effect).

Many studies within the general population of older married couples have found that negative social exchanges between spouses were associated with poor psychological well-being and distress (Ko & Lewis, 2011; Krause & Rook, 2003). However, few studies have compared the extent to which negative social exchanges between spouses have more harmful effects on caregiving couples versus non-caregiving couples. This comparison is important because the sickness of one spouse generally necessitates that couples adjust expectations, roles, and responsibilities in their marital relationships. In such stressful situations, the effects of negative interactions with spouses may be even more salient for caregiving couples than for non-caregiving couples.

Previous research on the general population of married couples has also shown that the effects of negative social exchanges vary by gender (Waite & Gallagher, 2002; Walen & Lachman, 2000). In terms of the influence of spousal support, women are more relationship oriented and are more influenced by interactions with a spouse than are men (Antonucci & Akiyama, 1987; Bolger, DeLongis, Kessler, & Schilling, 1989). Some research indicates that women are particularly susceptible to negative interactions within their marriage (Proulx, Helms, & Buehler, 2007). However, there has been relatively little focus on whether these gender differences in the effects of negative social exchanges on mental health exist for caregiving couples.

The purpose of this study is to understand the extent to which caregiving and care-receiving affect the relationship between negative social exchanges with a spouse and depressive symptoms. This research also explores how the gender of caregivers and care recipients influences the associations between negative social exchanges with spouses and depressive symptoms.

Negative social exchanges with spouses and depressive symptoms in caregiving and non-caregiving couples

Research on the general population of older adults indicates that negative social exchanges with spouses have detrimental effects on psychological health, such as depression, emotional distress, and diminished life satisfaction (Rook, Mavandadi, Sorkin, & Zettel, 2007; Whisman, Robustelli, Beach, Snyder, & Harper, 2015). Furthermore, interpersonal conflict can be a stressor in itself (Rook et al., 2007).

Equity theory provides a useful framework to explain the differences in the association between negative exchanges with a spouse and depressive symptoms between caregiving and non-caregiving couples. Equity theory posits that an inequitable relationship generates distress (Walster, Walster, & Berscheid, 1978). If a person is overbenefitted (i.e., receives more than gives), the person is likely to feel guilt and shame; if a person is underbenefitted (i.e., gives more than receives), the person is likely to feel resentment and anger (Hatfield, Traupmann, Sprecher, Utne, & Hay, 1985). Equity theory is relevant to this topic because caregiving situations usually lead to changes and adjustments in roles and responsibilities among older couples. These changes may erode reciprocity in a marital relationship and cause an imbalance in each spouse’s contribution to the relationship as caregiving spouses take on more responsibilities than do their ill partners (Ybema, Kuijer, Buunk, DeJong, & Sanderman, 2001). This imbalance may make caregiving couples more susceptible than non-caregiving couples to the adverse effects of negative exchanges with spouses. The feelings of guilt and shame associated with overbenefitting and of resentment and anger associated with underbenefitting may enhance the potency of negative exchanges with spouses thereby leading to depression.

Empirical research on caregivers suggests that negative interactions with care recipients are associated with greater distress and depressive symptoms among caregivers (Creasey, Myers, Epperson, & Taylor, 1990; Townsend & Franks, 1997). Although negative social exchanges with care recipients occur infrequently (Carruth, 1996), caregivers may have difficulties in coping with such negative exchanges if care recipients criticize or reject them. This is because caregiving responsibilities may impose burdens and feelings of inequity on caregiving spouses. In addition, caregiving spouses tend to receive less support from other informal and formal networks because caring for an ill partner is seen as part of the marriage contract (Miller & Montgomery, 1990). Given spousal caregivers’ risk of stress and burden, caregiving spouses may be vulnerable to negative exchanges with their partners.

Negative exchanges with caregiving spouses may have more adverse effects on care-receiving older adults than on healthy older adults. This salience of negative social exchanges is likely due to the notion that negative responses from caregiving spouses constitute a violation of care recipients’ expectations that close relationships should be supportive in times of crisis (Newsom, Nishishiba, Morgan, & Rook, 2003). Existing studies have suggested that some caregiving spouses respond negatively to their ill partners (e.g., expressing excessive worry); alternatively, they may avoid emotional connection while providing instrumental help to them (Lutzky & Knight, 1994; Manne, Alfieri, Taylor, & Dougherty, 1999). Caregiving spouses who behave in negative ways toward their ill partner may provide inadequate care by overprotecting them or treating them with less respect (Thompson & Sobolew-Shubin, 1993). Moreover, as spousal caregivers exhibit more negative behaviors toward their care-receiving partners, such as criticizing them, care recipients may try to be undemanding and compliant with caregivers’ decisions to avoid tension in their relationships with caregivers (Lewinter, 2003; Ward-Griffin, Bol, & Oudshoorn, 2006). These unfavorable consequences of negative exchanges with caregivers may lead to unmet needs and feelings of being a burden that, in turn, can contribute to care recipients’ depressive symptoms.

In sum, the caregiving literature demonstrates that negative exchanges between caregivers and care recipients exacerbate the harmful impact of illness and disability on the couple (Ball et al., 2010; Franks et al., 2006). These negative exchanges between spouses may enhance illness and disability, increase stress related to care, and diminish mental health. Few studies, however, have compared caregiving couples to non-caregiving couples to determine the extent to which caregiving impacts the effects of support exchanges between spouses. The present study addresses this gap by making this comparison and examining the impact of negative spousal support on depression. Our first hypothesis is that negative social exchanges with spouses will exert a stronger effect on mental health outcomes for spouses in caregiving couples than for spouses in non-caregiving couples.

Gender, negative exchanges with spouse, and care

Considering the role of gender is important in understanding the relationship between negative exchanges with spouses and depressive symptoms, particularly for caregiving couples. Studies applying equity theory to married couples suggest that women are more sensitive to inequitable relationships than men (DeMaris, 2010). As women are socialized into their gender roles, they tend to place greater emphasis on intimate relationships and invest more of their interpersonal resources, such as emotional support, into these relationships (Mickelson, Claffey, & Williams, 2006). Wives, compared to husbands, are more likely to report being underbenefitted in their relationships (Grote & Clark, 2001). For adults who assumed traditional gender roles throughout their marriages, it was usual for wives to provide the majority of the caregiving work in their families. This long-term caregiving work may make women more susceptible to the ill effects of caregiving later in life. For example, Wright and Aquilino (1998) found that older caregiving wives were frustrated, especially when their husbands did not provide any emotional support. In comparison to caregiving wives, caregiving husbands may be less likely to perceive caregiving situations as inequitable; they may regard caregiving in later life as an opportunity to reciprocate their wives’ efforts to care for the family in earlier stages of marriage (Russell, 2001). Given the gender differences in the perception of caregiving situations, negative exchanges with spouses play a more adverse role for caregiving wives than for caregiving husbands. In contrast, these gender differences in the effects of negative exchanges with spouses may not be as salient for care recipients. Specifically, care-receiving wives are likely to feel less underbenefitted than their caregiving counterparts because they are receiving help from their husbands. This assistance from their caregiving husbands can make them less susceptible to the effects of negative exchanges with their husbands.

There are few empirical studies that consider gender differences in the association between negative interactions with spouses and mental health in the context of caregiving. Most studies focus exclusively on women or men (Ball et al., 2010; Snyder, 2000) or do not investigate gender differences (Symister & Friend, 2003; Townsend & Franks, 1997). In the few studies that have considered gender differences, there is some indication that wife caregivers may be more strongly influenced by spousal negative behaviors than husband caregivers. Caregiving wives are more likely to perceive costs related to caregiving than are caregiving husbands (Raschick & Ingersoll-Dayton, 2004). Caregiving wives also report higher levels of distress and greater feelings of isolation than caregiving husbands (Hooker, Manoogian-O’Dell, Monahan, Frazier, & Shifren, 2000; Pinquart & Sörensen, 2006); thus, the effects of care recipients’ negative reactions to caregivers on caregivers’ depressive symptoms may be more salient for caregiving wives than for caregiving husbands. Based on this small body of literature, our second hypothesis is that, among caregivers, negative social exchanges with spouses will have a greater effect on wives than on husbands.

Because there are few studies that have tested gender differences in the relationship between negative interactions with caregivers and care recipients’ depressive symptoms, we draw from studies on patients with chronic illness and their spouses. One such study on patients with cancer found gender differences in negative social exchanges with spouses (Manne et al., 1999). Male patients were more likely to report negative social exchanges with their spouses (e.g., being demanding) than were female patients while female patients were more likely to report positive support from their spouses (e.g., showing love/affection) than were male patients (Manne et al., 1999). These gender differences may result from the fact that women are more likely than men to focus on positive support and less likely to recall negative interactions from support providers (Yankeelov, Barbee, Cunningham, & Druen, 1991). Our third hypothesis is that wives who are care recipients will experience less ill effects from negative exchanges than their male counterparts.

Method

Sample

This study uses two subsamples of the Health and Retirement Study (HRS) that consist of those respondents who lived with their spouses in the community in 2012 or in 2014. These subsamples were derived from respondents who completed the Leave-Behind Participant Lifestyle Questionnaires (Lifestyle Questionnaires) in 2012 or 2014 which covers psychosocial aspects of the respondents’ lives. As the Lifestyle Questionnaire was administered to a subgroup of respondents who were selected, there was no overlap of respondents in the 2012 and 2014 waves. We used two cross-sectional data sets from the 2012 and 2014 waves of the HRS to increase the sample of caregiving couples. The inclusion criteria for the sample analyzed were that respondents (1) were married or partnered, (2) lived with their spouse/partner in the community, (3) were at least 50 years old in the year when they answered the survey, (4) were a partner in a heterosexual couple, (5) had no missing information about whether or not they received care from their spouses, and (6) completed the Lifestyle Questionnaire during the same wave as their spouse. Those who were dropped from the study were more likely to be females and non-Whites and were less educated, more impaired, more depressed than the current sample. Excluded from the analytic sample were couples in which both spouses provided care to each other (N = 51). We decided to exclude these couples because the provision of care by just one partner is the central focus of this article. These inclusion and exclusion criteria resulted in 3,582 couples. The couples in which both spouses provide care to each other were more likely to be non-White and were less educated, more impaired, and more depressed than the analytic sample in this study. Additionally, 52 couples were dropped because of missing values on exogenous variables (i.e., wives’ or husbands’ years of education); thus, 3,530 couples (7,060 individuals) were included in the final analysis.

Measures

Spousal care and gender

Spousal care was assessed by whether respondents had received any help related to activities of daily living (i.e., bathing, eating, dressing, walking across a room, and getting in/out of bed) or instrumental activities of daily living (i.e., using a telephone, taking medication, handling money, shopping, and preparing meals) from their partner during the past month. If the respondent reported that they had received any help from their partner or spouse, the respondent was coded as the care recipient, and his or her partner was coded as the caregiver. In addition, gender was coded as a dichotomous variable (i.e., 0 = male, 1 = female). The measures of spousal care and gender were used to create subgroups to test moderating effects of caregiving and care receipt (i.e., caregiving wives, care-receiving husbands, caregiving husbands, care-receiving wives, and husbands and wives who did not engage in caregiving or receiving). We provide a detailed explanation of how the subgroups were generated in the section on the analytical plan.

Negative social exchanges with spouses

Respondents’ perceived negative social exchanges with their partners were measured using 4 items, assessed by a scale ranging from 0 (not at all) to 3 (a lot) (Birditt, Newton, Cranford, & Ryan, 2015). The questions were how much the partner (1) demanded of the respondent, (2) criticized the respondent, (3) let the respondent down when the respondent was counting on him or her, and (4) got on the respondent’s nerves. Higher scores indicate more negative social exchanges with the spouse. For husbands, α is .792; for wives, α is .790.

Depressive symptoms

Depressive symptoms were measured by a shortened version of the Center for Epidemiologic Studies Depression Scale (CES-D) (Radloff, 1977). The CES-D score is the sum of eight indicators. Six of the eight indicators measure whether the respondent experienced the following negative sentiments all or most of the time during the past week: depression, everything is an effort, sleep is restless, feel alone, feel sad, and cannot get going. The other two indicators measure positive feelings such as whether the respondent felt happy and enjoyed life all or most of the time. The indicators of positive feelings were reverse coded. Thus, higher scores indicate greater depressive symptoms in the past week, ranging from 0 to 8. Cronbach’s α is .77 and .81 for husbands and wives in 2012, respectively, and .77 and .79 for husbands and wives in 2014, respectively.

Control variables

Respondents’ age, race, and years of education were included as control variables in the analytical model, as previous studies found these variables were associated with negative exchanges and depressive symptoms (Cheng, Li, Leung, & Chan, 2011; Newsom, Rook, Nishishiba, Sorkin, & Mahan, 2005). Respondents’ age and years of education were measured as continuous variables in years. Race was measured as a dichotomous variable (0 = non-White, 1 = White). To control for period effects, we included a dummy variable for year of survey (1 = a respondent answered the survey in 2014).

Analytical plan

The main analytical strategy for this study was the actor–partner interdependence model (APIM; Cook & Kenny, 2005) using Mplus. The APIM is an appropriate analytic method because our data consist of wives and husbands and we are exploring the impact of exchanges between spouses on each spouse. Because spouses’ assessment of depressive symptoms was generally correlated, the lack of independence between their assessments of outcomes is accounted for in the APIM. The APIM also has advantages in taking into account nonindependence between scores of spouses. In addition, the APIM takes into account interpersonal effects of one member in a couple’s perceived negative exchanges on their own (actor effect) and their partner’s (partner effect) depressive symptoms.

Our analysis plan was comprised of several steps. First, using the overall sample, the effects of negative social exchanges with spouses on depressive symptoms were examined after controlling for age, race, education, and waves when the survey was conducted. Model fit was assessed using the comparative fit index (CFI), the root mean square error of approximation (RMSEA), and the standardized root mean square residual (SRMR) since the χ² statistic is more likely to be significant when a sample size is large (Kline, 2011).

Once the overall model was identified, the moderating effect of the provision and receipt of care was tested by conducting subgroup analyses. Three subgroups were generated based on whether individuals received care from their spouse and the gender of caregivers/care-receivers: (1) couples where spouses neither provided care to nor received care from their partner (81%, N = 2,872), (2) couples where husbands received care from their wives (11%, N = 384), and (3) couples where wives received care from their husbands (8%, N = 274). To explore how the provision and receipt of care from spouses influences the association between negative exchanges from spouses and their own and partner’s depressive symptoms, we examined the measurement invariance of the latent construct (i.e., negative social exchanges) across groups. Testing measurement invariance includes a series of restrictive criteria. Configural invariance is satisfied if the pattern of parameters is invariant across groups. Weak factorial invariance, which is also referred to as metric invariance, builds upon configural invariance and additionally examines whether or not factor loadings are equivalent across groups. Strong factorial invariance, which is referred as to scalar invariance, builds upon metric invariance by requiring that item intercepts are equivalent across groups (Meredith & Horn, 2001). After the measurement invariance was confirmed across the three subgroups, we tested the invariance of paths in the APIM across the subgroups. We tested the equality of each path across the three subgroups. The χ² difference test was used to test the first hypothesis about whether the moderating effects of caregiving were significant or not.

Next, we tested the second hypothesis about gender differences in the association between negative exchanges and depressive symptoms. The measurement invariance between caregiving wives and caregiving husbands was first tested. Then, the Wald test was conducted to test gender differences in the relationship between negative social exchanges and depressive symptoms for statistical significance. The same process was conducted to test the third hypothesis about gender differences for care recipients regarding the relationship between negative exchanges and depressive symptoms.

Last, missing data were examined; 289 couples (8.18%) had missing values on at least one of the dependent variables or observed variables. Among them, 198 husbands and 204 wives had missing values on the measure of observed variables for negative exchanges with spouses and one husband on the measure of depressive symptoms. Respondents with missing data were more likely to be non-White and less educated than those without missing data (p < .001). A full-information maximum likelihood estimator (FIML), as implemented in Mplus, was used to estimate the model with missing data (Enders, 2010). The FIML approach includes not only complete cases but also partially recorded cases into the likelihood function for the estimation of parameters under at least missing at random. FIML has been found to be more accurate and more efficient than other methods for handling missing data, such as listwise deletion or pairwise deletion (Enders, 2010).

The HRS has a complex sampling design, so the data were analyzed with household-level weights to correct for oversampling of African Americans, Hispanics, and Floridians. Results of the APIM presented in this study are based on weighted data.

Results

Table 1 displays unweighted means and standard deviations of the study variables. On average, husbands were significantly older than their wives (p < .001), more impaired (p < .05) and were likely to have fewer depressive symptoms (p < .001). Among subgroups of husbands, they were significantly different in terms of age, educational attainment, levels of functional impairment, and depressive symptoms. Post hoc analysis of pairwise comparisons indicated that both care-receiving and caregiving husbands were older (p < .001, p < .01, respectively) and less educated (p < .001, p < .001, respectively) than husbands who provided or received no care (i.e., no care husbands). Care-receiving husbands were likely to be more physically impaired than other groups (p < .001, p < .001). Among subgroups of wives, both care-receiving and caregiving wives were older than wives who provided or received no care (i.e., no care wives; p < .001, p < .001). No care wives had higher levels of education than other groups (p < .001, p < .001), and care-receiving wives reported higher levels of impairment and depressive symptoms than other groups of wives (p < .001, p < .001).

Table 1.
Unweighted descriptive statistics of the overall sample and subgroups.

Husbands Wives p-Value^b

All (n = 3,530) No care (n = 2,872) Care recipient (n = 384) Caregiver (n = 274) p-Value^a All (n = 3,530) No care (n = 2,872) Caregiver (n = 384) Care recipient (n = 274) p-Value ^a

Age (mean, SD) 68.97 (9.64) 68.26 (9.32) 73.22 (10.15) 70.43 (10.43) * 66.06 (9.21) 65.44 (8.87) 68.88 (9.84) 68.59 (10.51) * ***

Race (1 = White) (%) 82.41 82.76 79.95 82.12 n.s. 83.03 83.39 82.29 80.29 n.s. n.s.

Years of education (mean, SD) 13.12 (3.14) 13.35 (3.06) 11.95 (3.30) 12.36 (3.25) * 13.05 (2.85) 13.27 (2.75) 12.31 (2.84) 11.89 (3.44) * n.s.

Functional impairment 0.44 (1.26) 0.13 (0.53) 2.96 (2.25) 0.23 (0.74) * 0.38 (1.17) 0.14 (0.63) 0.34 (1.08) 2.97 (2.15) * *

Negative exchanges with spouse

Too many demands on you 1.01 (0.86) 1.00 (0.84) 1.05 (0.94) 1.15 (0.90) * 1.00 (0.91) 0.96 (0.88) 1.30 (1.00) 0.95 (0.92) *** n.s.

Criticize you 1.12 (0.84) 1.11 (0.83) 1.17 (0.91) 1.15 (0.88) n.s. 0.98 (0.88) 0.97 (0.87) 1.05 (0.92) 1.08 (0.87) * *

Let you down 0.49 (0.76) 0.46 (0.73) 0.57 (0.85) 0.66 (0.87) * 0.72 (0.85) 0.71 (0.84) 0.79 (0.91) 0.75 (0.85) n.s. *

Gets on your nerves 0.90 (0.77) 0.87 (0.76) 1.02 (0.81) 1.01 (0.74) * 1.14 (0.81) 1.12 (0.79) 1.28 (0.91) 1.16 (0.84) *

Depressive symptoms 0.95 (1.54) 0.76 (1.34) 2.28 (2.19) 1.04 (1.48) * 1.20 (1.81) 1.01 (1.64) 1.51 (2.04) 2.77 (2.34) * ***

Note. n.s. = not significant; SD: standard deviation; ANOVA = analysis of variance; ADLs = activities of daily living; IADL = instrumental activities of daily living. Functional impairment was computed by summing the number of ADLs and IADLs that a respondent reported having difficulty in performing.

^aOne-way ANOVA and χ² test were conducted to compare the three subgroups.

^b Paired t-test and McNemar’s test were conducted to compare overall husbands and wives.

p < .05; p < .01; p < .001.

Table 2 presents correlations between the study variables. The high correlations among the observed variables of latent constructs support the construction of the latent variable. For both husbands and wives, the correlations appear to be in the expected direction. More negative social exchanges with a spouse are associated with his or her partner’s greater depressive symptoms.

Table 2.
Correlations between negative social exchanges and depressive symptoms (N = 3,530).

Items 1 2 3 4 5

1. Too many demands on you — .489 .451 .451 .181

2. Criticize you .544 — .486 .498 .177

3. Let you down .421 .431 — .545 .224

4. Gets on your nerves .500 .539 .502 — .240

5. Depressive symptoms .137 .142 .201 .224 —

Note. Correlation values for husbands are below the diagonal; those for wives are above the diagonal. All correlations are significant at the .001 level.

Testing the overall model

A model that tested the effects of negative social exchanges with spouses on depressive symptoms resulted in an acceptable fit. The χ² statistic, χ²(85) = 369.113, p < .001, was likely significant due to the large sample size, but the other fit indices were acceptable: RMSEA = .031; CFI = .951; SRMR = .026. The standardized coefficients of factor loadings for negative exchanges ranged from .623 to .762 for husbands and from .640 to .738 for wives. Figure 1 presents the standardized and unstandardized coefficients for paths and the squared multiple correlations (R²) for each construct.

Figure 1.
Actor–partner interdependent model of the effects of negative social exchanges with spouses on depressive symptoms for the overall sample: χ²(85, N = 3,530) = 396.113, p < .001; CFI = .951; RMSEA = .031; SRMR = .026. Coefficients presented first are standardized, and coefficients in the parentheses are unstandardized. CFI = comparative fit index; RMSEA = root mean square error of approximation; SRMR = standardized root mean square residual. p < .05; p < .01; p < .001.

The results indicated that husbands’ perceptions of greater negative social exchanges with their wives was related to the husbands’ greater depressive symptoms (β = .186, p < .001) as well as the wives’ greater depressive symptoms (β = .079, p < .01). For husbands, being White/Caucasian and more educated were associated with fewer depressive symptoms. For wives, greater negative social exchanges with their husbands were related to more depressive symptoms for the wives (β = .259, p < .001) as well as husbands’ greater depressive symptoms (β = .139, p < .001). In addition, being older and more educated were related to fewer negative social exchanges with spouses, and being more educated and responding to the 2014 survey questionnaire were related to fewer depressive symptoms.

Testing the moderating effect of caregiving and care receipt

To examine whether the effect of negative exchanges on depressive symptoms differs by the provision and receipt of care among older couples, we tested the moderating effect of caregiving and care receipt. The test for measurement invariance across subgroups supported weak factorial invariance (Meredith, 1993) across the subgroups (p = .49), suggesting that factor loadings are equivalent across the subgroups. However, strong factorial invariance (Meredith & Horn, 2001) was not supported (p < .001), indicating that item intercepts are not equivalent across the subgroups. Based on the modification indices, we freed the intercepts of two items (i.e., husbands’ report of wives’ criticizing behaviors and wives’ report of husbands’ demanding behaviors) to vary across the subgroups. In doing so, partial strong factorial invariance was held for other items (p = .29; see Supplementary Table 1).

Using this partial measurement invariance model, we then estimated the structural equation model for each subgroup. The coefficients of the control variables were estimated without equality constraints across the subgroups. Figure 2 illustrates the structural equation model for caregiving and receiving groups. The first coefficients presented in Figure 2 were unstandardized to compare coefficients across the subgroups. Coefficients in the parentheses were standardized to compare the relative strengths of the coefficients within each group. Although the magnitudes of the coefficients were different across the subgroups, the directions of the coefficients were similar.

Figure 2.
Actor–partner interdependent model of the effects of negative social exchanges with spouses on depressive symptoms for the subgroups. (a) No care group, (b) care-receiving husbands and caregiving wives group, and (c) caregiving husbands and care-receiving wives group. Coefficients presented first are unstandardized, and coefficients in the parentheses are standardized. p < .05; p < .01; p < .001.

For all three husband subgroups, husbands’ perception of negative social exchanges with spouses was associated with more depressive symptoms. For no care husbands and care-receiving husbands, greater negative exchanges with wives were significantly related to more depressive symptoms (b = 0.394, p < .001; b = 1.018, p < .001, respectively). For caregiving husbands, the relationship between negative exchanges and depressive symptoms was not significant (b = 0.261, p = .096). Inconsistent partner effects were observed across three groups of husbands. Wives’ perception of negative exchanges with husbands was significantly associated with husbands’ depressive symptoms only for no care husbands (b = 0.323, p < .001). For wives, the paths from wives’ perceptions of negative social exchanges with husbands to wives’ depressive symptoms were statistically significant for no care wives and caregiving wives. Greater negative social exchanges with husbands were associated with more depressive symptoms for no care wives and caregiving wives (b = 0.762, p < .001; b = 1.332, p < .001, respectively). However, for wives who received care from their husbands, the path from negative exchanges to depressive symptoms was not significant (p > .10). In addition, partner effects were observed only for care-receiving wives. Husbands’ perceptions of negative exchanges with wives were significantly related to depressive symptoms only for care-receiving wives (b = 0.825, p < .05).

We next evaluated whether or not the path coefficients were significantly different across the subgroups. The results showed that the path from husbands’ perceptions of negative exchanges with wives to their own depressive symptoms was significantly different across the subgroups, χ²(2)= 8.29, p < .05. The relationship between negative exchanges with wives and their husbands’ depressive symptoms was stronger for husbands who received care (b = 1.018) compared to those who were not involved in care (b = 0.394) or those who provided care (b = 0.261). Post hoc analysis of pairwise comparisons indicated that the path for care-receiving husbands was significantly different from that of caregiving husbands (p < .01) or no care husbands (p < .05). There was no significant difference in the path between caregiving and no care husbands (p = .444). This pattern of results was similar for wives; however, the difference was not significant (p = .102). In addition, partner effects did not significantly differ across the three groups (see Supplementary Table 2).

Testing the moderating effect of gender among caregivers and care recipients

We first tested the measurement invariance between husbands and wives among caregivers and care recipients. The weak factorial invariance (Meredith, 1993) was supported by a χ² difference test (p = .50), indicating that factor loadings are equivalent between gender; however, the strong factorial invariance (Meredith, 1993) was not supported (p < .001). Based on the modification indices, the 3 items (i.e., care recipients’ report on getting on nerves and letting down and caregivers’ report on criticizing) were identified as the source of the misfit. By freeing the intercepts of these 3 items, the partial strong factorial invariance was established (p = .15; see Supplementary Table 3). Using this partial measurement invariance model, we tested the gender differences in the effects of negative exchanges on depressive symptoms for caregivers versus care recipients. Results indicated that, among caregivers, there were significant gender differences in the association between negative exchanges with spouses and depressive symptoms, χ²(1) = 11.90, p < .001. Negative exchanges with care recipients had a stronger actor effect on depressive symptoms among caregiving wives than among caregiving husbands (b = 1.231, p < .001 for caregiving wives, b = 0.263, p = .094 for caregiving husbands). However, among care recipients, there was no significant difference between care-receiving wives and husbands in the actor effect of negative exchanges on their depressive symptoms χ²(1) = 1.234, p = .266. Regarding partner effects, no significant gender differences were found for either caregivers (p = .851) or care recipients (p = .193; see Supplementary Table 4).

Discussion

This study examined how the provision and receipt of care influences the relationship between negative social exchanges with spouses and depressive symptoms, using the 2012 and 2014 waves of the HRS data. The current study extends the previous literature on spousal relationships and psychological well-being by comparing the effects of negative social exchanges with spouses on depressive symptoms across caregiving and non-caregiving couples. This article provides insights about the differential experiences of husbands and wives.

Our first hypothesis concerning differences between caregiving and non-caregiving couples in the actor effect of spousal exchanges on depressive symptoms was supported only for husbands. Consistent with equity theory, among husbands, there were significant differences in the association between negative exchanges with wives and depressive symptoms across the subgroups. This study found that care-receiving husbands were the most vulnerable to negative exchanges when compared with caregiving husbands or no care husbands. Drawing upon equity theory, it may be that care-receiving husbands experience difficulty in establishing a reciprocal relationship with wives and perceive themselves as being overbenefitted. It is possible that, if wives negatively interact with their care-receiving husbands, care-receiving husbands feel that they are a burden and experience more guilt, which may increase the husbands’ depressive symptoms.

Interestingly, we found that, for caregiving husbands, the association between negative exchanges with their spouses and depressive symptoms was similar to the association among husbands who do not engage in caregiving. That is, for caregiving husbands, the caregiving situation did not moderate the association between negative exchanges with spouses and their own depressive symptoms. Gender role specialization hypothesis may explain this finding. Because spousal caregiving is not a traditional role for husbands, caregiving husbands often receive affirmation when having difficulties with caregiving work as well as praise from others for their work (Calasanti & King, 2007). In addition, previous research has found that caregiving husbands utilize managerial strategies to deal with caregiving work, and consequently, caregiving husbands report feelings of competence as caregivers and low levels of burden (Russell, 2001). This previous research may help explain the lack of difference between caregiving husbands and those who are not caregivers. That is, given the lower levels of burden and the affirmation and appreciation of caregiving work from others (Ribeiro, Paúl, & Nogueira, 2007), negative reactions from care recipients may have similar effects both for caregiving husbands and for husbands who do not engage in caregiving.

Among wives, we found no significant differences in the effect of negative exchanges with husbands on the wives’ own depressive symptoms across care status. Existing research indicates that wives may be sensitive to inequity in marital relationships and susceptible to negative exchanges with husbands even in non-caregiving situations (DeMaris, 2010). However, our findings did not support the moderating effects of care status on the association between negative exchanges and depressive symptoms. The effects of negative exchanges with husbands are similar for wives regardless of whether or not they are in caregiving relationships. Due to wives’ long-standing patterns of providing care in the family throughout their marriages, spousal care may not generate significant differences in exacerbating inequitable relationships among wives. That is, caregiving wives are probably less susceptible to the adverse effects of negative social exchanges with husbands than non-caregiving or care-receiving wives because they have already experienced and practiced caregiving duties in their marriage.

The second hypothesis related to gender differences in caregivers was supported by the data. We found wife caregivers were more vulnerable to the effects of negative exchanges with care recipients than husband caregivers. This finding is consistent with the existing literature on gender differences in caregiving stress and coping styles (Pinquart & Sörensen, 2006) as well as gender differences in perceiving inequity in marital relationship (Grote & Clark, 2001).

The third hypothesis concerning gender differences among care recipients was not supported. We found neither actor nor partner effects for the relationship between negative exchanges and depressive symptoms. Our findings are not consistent with those of Yankeelov, Barbee, Cunningham, and Druen (1991) who found that women were more likely to focus on positive interactions than were their male counterparts. Instead, our work supports the findings of Manne, Taylor, Dougherty, and Kemeny (1997) who found no gender differences in the effects of negative social exchanges with spouses on psychological distress. Instead, Manne and colleagues (1997) found that positive exchanges with spouses or impairment levels of patients with cancer were differentially associated with distress depending on the gender of the patient with cancer. At present, the body of research examining the effects of negative exchanges on care recipients is very small. Future research should examine how care recipients perceive and respond to social exchanges with caregiving spouses.

Importantly, the effects of each person’s perceived negative exchanges on his/her spouse’s depressive symptoms (partner effects) were found for both wives and husbands in the overall model. When we conducted subgroup analyses, however, such partner effects were significant only for two subgroups. In couples who do not provide or receive any care, the husbands’ perceived negative exchanges with their wives had significant effects on the wives’ depressive symptoms, but the wives’ perceived negative exchanges with their husbands did not have any partner effects. This result is consistent with a previous study on younger Italian couples (Donato et al., 2015). Another group who showed significant partner effects in the present study was care-receiving wives. Interestingly, caregiving husbands’ perceived negative exchanges were significantly associated with greater depressive symptoms in care-receiving wives, whereas wives’ own perceived negative exchanges were not. Although we found some differences in partner effects across subgroups, those differences were not statistically significant across care status and gender.

Findings from this research point to some important directions for exploring the causal mechanisms in how social exchanges influence the well-being of long-term marital partners when they provide or receive care. Consistent with previous research, we found that negative exchanges with spouses increase depressive symptoms (Ko & Lewis, 2011). However, some of our hypotheses about the ill effects of negative exchanges and how these differed for care-receiving/caregiving husbands and wives were not supported in the present study. It may be that gender identity and gender roles play a more critical function than gender in the relationship between negative exchanges and depression. Specifically, caregiving situations may generate imbalances in gender duties and roles that require readjustments within marital relationships (Ybema et al., 2001). Thus, gender roles and identity may moderate adverse effects of negative interactions with spouses on depressive symptoms by increasing feelings of being overbenefitted or underbenefitted in caregiving contexts. For example, if husbands hold strong “traditional” gender norms and are more oriented toward hegemonic masculinity (e.g., putting a high value on independence and toughness; O’Brien, Hunt, & Hart, 2005), receiving care from stressed caregivers may make them feel overbenefitted and guilty. In contrast, feminine ideals are related to asking for help, caring about health, and taking a primary role in nurturing and monitoring family members’ health (Courtenay, 2000). If older wives have a strong belief in feminine ideals, caregiving and care-receiving roles may not intensify feelings of being overbenefitted or underbenefitted as these roles are not contradictory to their gender identity. There could be a variation in the degree to which older men and women endorse masculinity and femininity ideals. Thus, future research can build upon the present study by examining how gender roles and identity are embedded in older couples’ marital relationships; how such gender roles and identity affect older husbands’ and wives’ perceptions of equity regarding caregiving and care-receiving roles; and how their perceptions of equity and negative exchanges are interwoven and, in turn, affect their well-being.

Despite its innovative approach, this study has several limitations. First, caution is needed in interpreting these results because the sample sizes of caregiving couples and non-caregiving couples are unequal. The sample size of caregiving couples is much smaller than that of non-caregiving couples. For example, as standard errors are sensitive to sample size, the larger standard errors of path coefficients for care-receiving couples seem to make the path from negative exchanges to depressive symptoms insignificant. Future research, using larger samples of caregiving couples, is needed to further compare how caregiving and care-receiving influence the relationship between negative exchanges with spouses and their partners’ depressive symptoms.

Second, this study was based on cross-sectional data, so it is not possible to determine causal relationships between spousal support and depressive symptoms. Although the HRS is a longitudinal survey, caregiving and care-receiving are time-varying variables and the proportion of caregiving/care-receiving couples in the whole sample is very low. To test the moderating effects of changes in care status on the relationship between negative exchanges with spouses and depressive symptoms, a larger sample size is needed within each subgroup. Furthermore, although existing studies have found that an individual’s current emotions have minimal effects on the reporting of negative exchanges (Krause, Liang, & Yatomi, 1989), it may be possible that depressive moods in one partner affect support exchanges between spouses. This relationship should be further explored.

Third, negative exchanges were used as a measure for couples’ interactions in this study. Because positive and negative exchanges have been found to be related to different dimensions of mental health (Ingersoll-Dayton, Morgan, & Antonucci, 1997), we also explored the possibility of testing the effects of positive exchanges with spouses on depressive symptoms, differing by care status and gender. However, the measurement invariance was not supported across subgroups, and we were unable to analyze positive exchanges with spouses in our study. Future research should include measures of positive interactions related to caregiving and care-receiving.

Fourth, previous studies have suggested that the types and the severity of illness may affect a couple’s adjustment and well-being. For example, dementia may be more likely to strain a couple’s interactions and well-being than other types of physical illness (Baikie, 2002). However, given the small number of people with dementia and other types of severe illness, our study could not compare the relationship between negative exchanges with spouses and depressive symptoms by type and seriousness of illness: however, future research should examine this issue.

In conclusion, the present study advances our understanding of the relationship between caregiving/care-receiving, spousal interactions, and depressive symptoms. The findings of this study have important implications for caregiving couples in both practice and research. The findings highlight the importance of negative exchanges when one spouse provides care to his or her partner. This study provides evidence that the adverse mental health effects of negative exchanges between spouses are greater for care-receiving husbands. To reduce depressive symptoms in spouses in caregiving couples, interventions that facilitate supportive interactions between care recipients and caregivers may be effective, especially for couples where wives provide care to husbands. Practitioners can help caregiving couples deal with difficult emotions by educating them in communication skills as well as coping skills. More interventions for caregiving couples should be developed and tested. One promising approach would be to adapt couple-based therapies, such as emotion-focused couples therapy (Johnson & Greenberg, 1995; McLean et al., 2008). These kinds of therapies aim to improve marital functioning and attachment which is especially important for couples where wives provide care to husbands in later life.

Supplemental material

Supplemental Material, Revision2_supplementary_submit - Relationships between negative exchanges and depressive symptoms in older couples: The moderating roles of care and gender

Supplemental Material, Revision2_supplementary_submit for Relationships between negative exchanges and depressive symptoms in older couples: The moderating roles of care and gender by Minyoung Kwak and Berit Ingersoll-Dayton in Journal of Social and Personal Relationships

	Husbands	Wives	p-Value^b
Age (mean, SD)	68.97 (9.64)	68.26 (9.32)	73.22 (10.15)	70.43 (10.43)	***	66.06 (9.21)	65.44 (8.87)	68.88 (9.84)	68.59 (10.51)	***	***
Race (1 = White) (%)	82.41	82.76	79.95	82.12	n.s.	83.03	83.39	82.29	80.29	n.s.	n.s.
Years of education (mean, SD)	13.12 (3.14)	13.35 (3.06)	11.95 (3.30)	12.36 (3.25)	***	13.05 (2.85)	13.27 (2.75)	12.31 (2.84)	11.89 (3.44)	***	n.s.
Functional impairment	0.44 (1.26)	0.13 (0.53)	2.96 (2.25)	0.23 (0.74)	***	0.38 (1.17)	0.14 (0.63)	0.34 (1.08)	2.97 (2.15)	***	*
Negative exchanges with spouse
Too many demands on you	1.01 (0.86)	1.00 (0.84)	1.05 (0.94)	1.15 (0.90)	*	1.00 (0.91)	0.96 (0.88)	1.30 (1.00)	0.95 (0.92)	***	n.s.
Criticize you	1.12 (0.84)	1.11 (0.83)	1.17 (0.91)	1.15 (0.88)	n.s.	0.98 (0.88)	0.97 (0.87)	1.05 (0.92)	1.08 (0.87)	*	***
Let you down	0.49 (0.76)	0.46 (0.73)	0.57 (0.85)	0.66 (0.87)	***	0.72 (0.85)	0.71 (0.84)	0.79 (0.91)	0.75 (0.85)	n.s.	***
Gets on your nerves	0.90 (0.77)	0.87 (0.76)	1.02 (0.81)	1.01 (0.74)	***	1.14 (0.81)	1.12 (0.79)	1.28 (0.91)	1.16 (0.84)	**	***
Depressive symptoms	0.95 (1.54)	0.76 (1.34)	2.28 (2.19)	1.04 (1.48)	***	1.20 (1.81)	1.01 (1.64)	1.51 (2.04)	2.77 (2.34)	***	***

	Items	1	2	3	4	5
1.	Too many demands on you	—	.489	.451	.451	.181
2.	Criticize you	.544	—	.486	.498	.177
3.	Let you down	.421	.431	—	.545	.224
4.	Gets on your nerves	.500	.539	.502	—	.240
5.	Depressive symptoms	.137	.142	.201	.224	—

Footnotes

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

ORCID iD

Minyoung Kwak

Open research statement

As part of IARR’s encouragement of open research practices, the authors have provided the following information: This research was not pre-registered. The data used in the research are available. The data can be obtained at The materials used in the research are available. The materials can be obtained by emailing: mykwak@daegu.ac.kr.

Supplemental material

Supplemental material for this article is available online.

References

Antonucci

T. C.

Akiyama

(1987). An examination of sex differences in social support among older men and women. Sex Roles, 17, 737–749.

Baikie

(2002). The impact of dementia on marital relationships. Sexual and Relationship Therapy, 17, 289–299.

Ball

Snow

A. L.

Steele

A. B.

Morgan

R. O.

Davila

J. A.

Wilson

Kunik

M. E.

(2010). Quality of relationships as a predictor of psychosocial functioning in patients with dementia. Journal of Geriatric Psychiatry and Neurology, 23, 109–114.

Berg

C. A.

Upchurch

(2007). A developmental-contextual model of couples coping with chronic illness across the adult life span. Psychological Bulletin, 133, 920.

Berkman

L. F.

Glass

Brissette

Seeman

T. E.

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

Birditt

K. S.

Newton

N. J.

Cranford

J. A.

Ryan

L. H.

(2015). Stress and negative relationship quality among older couples: Implications for blood pressure. Journals of Gerontology Series B: Psychological Sciences and Social Sciences, 71, 775–785.

Bolger

DeLongis

Kessler

R. C.

Schilling

E. A.

(1989). Effects of daily stress on negative mood. Journal of Personality and Social Psychology, 57, 808.

Calasanti

King

(2007). Taking ‘women’s work’ ‘like a man’: Husbands’ experiences of care work. The Gerontologist, 47, 516–527.

Carruth

A. K.

(1996). Development and testing of the caregiver reciprocity scale. Nursing Research, 45, 92–97.

10.

Cheng

S. T.

K. K.

Leung

E. M.

Chan

A. C.

(2011). Social exchanges and subjective well-being: Do sources of positive and negative exchanges matter? Journals of Gerontology Series B: Psychological Sciences and Social Sciences, 66, 708–718.

11.

Choi

Yorgason

J. B.

Johnson

D. R.

(2015). Marital quality and health in middle and later adulthood: Dyadic associations. Journals of Gerontology Series B: Psychological Sciences and Social Sciences, 71, 154–164.

12.

Cook

W. L.

Kenny

D. A.

(2005). The actor–partner interdependence model: A model of bidirectional effects in developmental studies. International Journal of Behavioral Development, 29, 101–109.

13.

Courtenay

W. H.

(2000). Constructions of masculinity and their influence on men’s well-being: A theory of gender and health. Social Science & Medicine, 50, 1385–1401.

14.

Creasey

G. L.

Myers

B. J.

Epperson

M. J.

Taylor

(1990). Couples with an elderly parent with Alzheimer’s disease: Perceptions of familial relationships. Psychiatry, 53, 44–51.

15.

DeMaris

(2010). The 20-year trajectory of marital quality in enduring marriages: Does equity matter? Journal of Social and Personal Relationships, 27, 449–471.

16.

Donato

Parise

Iafrate

Bertoni

Finkenauer

Bodenmann

(2015). Dyadic coping responses and partners’ perceptions for couple satisfaction: An actor -partner interdependence analysis. Journal of Social and Personal Relationships, 32, 580–600.

17.

Enders

C. K.

(2010). Applied missing data analysis. New York, NY: Guilford Press.

18.

Franks

M. M.

Stephens

M. A. P.

Rook

K. S.

Franklin

B. A.

Keteyian

S. J.

Artinian

N. T.

(2006). Spouses’ provision of health-related support and control to patients participating in cardiac rehabilitation. Journal of Family Psychology, 20, 311.

19.

Grote

N. K.

Clark

M. S.

(2001). Perceiving unfairness in the family: Cause or consequence of marital distress? Journal of Personality and Social Psychology, 80, 281.

20.

Hatfield

Traupmann

Sprecher

Utne

Hay

(1985). Equity and intimate relations: Recent research. In Ickes

(Ed.), Compatible and incompatible relationships (pp. 91–117). New York, NY: Springer.

21.

Holahan

C. J.

Moos

R. H.

Moerkbak

M. L.

Cronkite

R. C.

Holahan

C. K.

Kenney

B. A.

(2007). Spousal similarity in coping and depressive symptoms over 10 years. Journal of Family Psychology, 21, 551.

22.

Hooker

Manoogian-O’Dell

Monahan

D. J.

Frazier

L. D.

Shifren

(2000). Does type of disease matter? Gender differences among Alzheimer’s and Parkinson’s disease spouse caregivers. The Gerontologist, 40, 568–573.

23.

Ingersoll-Dayton

Morgan

Antonucci

(1997). The effects of positive and negative social exchanges on aging adults. The Journals of Gerontology Series B: Psychological Sciences and Social Sciences, 52, S190–S199.

24.

Johnson

S. M.

Greenberg

L. S.

(1995). The emotionally focused approach to problems in adult attachment. In Jacobson

N. S.

Gurman

A. S.

(Eds.), Clinical handbook of couple therapy (pp. 121–141). New York, NY: The Guilford Press.

25.

Kleiboer

A. M.

Kuijer

R. G.

Hox

J. J.

Schreurs

K. M.

Bensing

J. M.

(2006). Receiving and providing support in couples dealing with multiple sclerosis: A diary study using an equity perspective. Personal Relationships, 13, 485–501.

26.

Kline

R. B.

(2011) Principles and Practice of Structural Equation Modeling. New York: Guilford Press.

27.

L. K.

Lewis

M. A.

(2011). The role of giving and receiving emotional support in depressive symptomatology among older couples: An application of the actor-partner interdependence model. Journal of Social and Personal Relationships, 28, 83–99.

28.

Krause

Liang

Yatomi

(1989). Satisfaction with social support and depressive symptoms: A panel analysis. Psychology and Aging, 4, 88.

29.

Krause

Rook

K. S.

(2003). Negative interaction in late life: Issues in the stability and generalizability of conflict across relationships. The Journals of Gerontology Series B: Psychological Sciences and Social Sciences, 58, P88–P99.

30.

Lewinter

(2003). Reciprocities in caregiving relationships in Danish elder care. Journal of Aging Studies, 17, 357–377.

31.

Lutzky

S. M.

Knight

B. G.

(1994). Explaining gender differences in caregiver distress: The roles of emotional attentiveness and coping styles. Psychology and Aging, 9, 513.

32.

Manne

S. L.

Alfieri

Taylor

K. L.

Dougherty

(1999). Spousal negative responses to cancer patients: The role of social restriction, spouse mood, and relationship satisfaction. Journal of Consulting and Clinical Psychology, 67, 352.

33.

Manne

S. L.

Taylor

K. L.

Dougherty

Kemeny

(1997). Supportive and negative responses in the partner relationship: Their association with psychological adjustment among individuals with cancer. Journal of Behavioral Medicine, 20, 101–125.

34.

Matire

L. M.

Stephens

M. A. P.

Druley

J. A.

Wojno

W. C.

(2002). Negative reactions to received spousal care: Predictors and consequences of miscarried support. Health Psychology, 21, 167.

35.

McLean

L. M.

Jones

J. M.

Rydall

A. C.

Walsh

Esplen

M. J.

Zimmermann

Rodin

G. M.

(2008). A couples intervention for patients facing advanced cancer and their spouse caregivers: Outcomes of a pilot study. Psycho-Oncology, 17, 1152–1156.

36.

Meredith

(1993). Measurement invariance, factor analysis and factorial invariance. Psychometrika, 58, 525–543.

37.

Meredith

Horn

(2001). The role of factorial invariance in modeling growth and change. In Collins

L. M.

Sayer

A. G.

(Eds.), Decade of behavior. New methods for the analysis of change (pp. 203–240). Washington, DC: American Psychological Association.

38.

Mickelson

K. D.

Claffey

S. T.

Williams

S. L.

(2006). The moderating role of gender and gender role attitudes on the link between spousal support and marital quality. Sex Roles, 55, 73–82.

39.

Miller

Montgomery

(1990). Family caregivers and limitations in social activities. Research on Aging, 12, 72–93.

40.

Newsom

J. T.

Nishishiba

Morgan

D. L.

Rook

K. S.

(2003). The relative importance of three domains of positive and negative social exchanges: A longitudinal model with comparable measures. Psychology and Aging, 18, 746.

41.

Newsom

J. T.

Rook

K. S.

Nishishiba

Sorkin

D. H.

Mahan

T. L.

(2005). Understanding the relative importance of positive and negative social exchanges: Examining specific domains and appraisals. The Journals of Gerontology Series B: Psychological Sciences and Social Sciences, 60, P304–P312.

42.

O’Brien

Hunt

Hart

(2005). ‘It’s caveman stuff, but that is to a certain extent how guys still operate’: Men’s accounts of masculinity and help seeking. Social Science & Medicine, 61, 503–516.

43.

Peterson

K. M.

Smith

D. A.

(2010). An actor–partner interdependence model of spousal criticism and depression. Journal of Abnormal Psychology, 119, 555.

44.

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, P33–P45.

45.

Proulx

C. M.

Helms

H. M.

Buehler

(2007). Marital quality and personal well-being: A meta-analysis. Journal of Marriage and Family, 69, 576–593.

46.

Radloff

L. S.

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

47.

Raschick

Ingersoll-Dayton

(2004). The costs and rewards of caregiving among aging spouses and adult children. Family Relations, 53, 317–325.

48.

Ribeiro

Paúl

Nogueira

(2007). Real men, real husbands: Caregiving and masculinities in later life. Journal of Aging Studies, 21, 302–313.

49.

Robles

T. F.

Slatcher

R. B.

Trombello

J. M.

McGinn

M. M.

(2014). Marital quality and health: A meta-analytic review. Psychological Bulletin, 140, 140.

50.

Rook

K. S.

Mavandadi

Sorkin

D. H.

Zettel

L. A.

(2007). Optimizing social relationships as a resource for health and well-being in later life. In Aldwin

C. M.

Park

C. L.

Spiro

III (Eds.), Handbook of health psychology and aging (pp. 267–285). New York, NY: Guilford Press.

51.

Russell

(2001). In sickness and in health: A qualitative study of elderly men who care for wives with dementia. Journal of Aging Studies, 15, 351–367.

52.

Sabey

A. K.

Rauer

A. J.

(2018). Changes in older couples’ compassionate love over a year: The roles of gender, health, and attachment avoidance. Journal of Social and Personal Relationships, 35, 1139–1158.

53.

Schulz

Schwarzer

(2004). Long-term effects of spousal support on coping with cancer after surgery. Journal of Social and Clinical Psychology, 23, 716–732.

54.

Snyder

J. R.

(2000). Impact of caregiver-receiver relationship quality on burden and satisfaction. Journal of Women & Aging, 12, 147–167.

55.

Symister

Friend

(2003). The influence of social support and problematic support on optimism and depression in chronic illness: A prospective study evaluating self-esteem as a mediator. Health Psychology, 22, 123.

56.

Thompson

S. C.

Sobolew-Shubin

(1993). Overprotective relationships: A nonsupportive side of social networks. Basic and Applied Social Psychology, 14, 363–383.

57.

Townsend

A. L.

Franks

M. M.

(1997). Quality of the relationship between elderly spouses: Influence on spouse caregivers’ subjective effectiveness. Family Relations, 46, 33–39.

58.

Waite

Gallagher

(2002). The case for marriage: Why married people are happier, healthier and better off financially. New York, NY: Broadway Books.

59.

Walen

H. R.

Lachman

M. E.

(2000). Social support and strain from partner, family, and friends: Costs and benefits for men and women in adulthood. Journal of Social and Personal Relationships, 17, 5–30.

60.

Walster

G. W.

Berscheid

(1978). Equity: Theory and Research. Boston: Allyn and Bacon.

61.

Ward-Griffin

Bol

Oudshoorn

(2006). Perspectives of women with dementia receiving care from their adult daughters. Canadian Journal of Nursing Research, 38, 120–146.

62.

Whisman

M. A.

Robustelli

B. L.

Beach

S. R.

Snyder

D. K.

Harper

J. M.

(2015). Marital discord and depression in middle-aged and older couples: Is taxon status associated with depression? Journal of Social and Personal Relationships, 32, 967–973.

63.

Williams

Umberson

(2004). Marital status, marital transitions, and health: A gendered life course perspective. Journal of Health and Social Behavior, 45, 81–98.

64.

Wolff

J. L.

Kasper

J. D.

(2006). Caregivers of frail elders: Updating a national profile. The Gerontologist, 46, 344–356.

65.

Wright

D. L.

Aquilino

W. S.

(1998). Influence of emotional support exchange in marriage on caregiving wives’ burden and marital satisfaction. Family Relations, 47, 195–204.

66.

Yankeelov

Barbee

Cunningham

Druen

(1991). Interactive coping in romantic relationships. Paper presented at the International Conference on Personal Relationships, Normal, IL.

67.

Ybema

J. F.

Kuijer

R. G.

Buunk

B. P.

DeJong

G. M.

Sanderman

(2001). Depression and perceptions of inequity among couples facing cancer. Personality and Social Psychology Bulletin, 27, 3–13.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.19 MB

	Husbands					Wives					p-Value^b
	All (n = 3,530)	No care (n = 2,872)	Care recipient (n = 384)	Caregiver (n = 274)	p-Value^a	All (n = 3,530)	No care (n = 2,872)	Caregiver (n = 384)	Care recipient (n = 274)	p-Value ^a	p-Value^b
Age (mean, SD)	68.97 (9.64)	68.26 (9.32)	73.22 (10.15)	70.43 (10.43)	***	66.06 (9.21)	65.44 (8.87)	68.88 (9.84)	68.59 (10.51)	***	***
Race (1 = White) (%)	82.41	82.76	79.95	82.12	n.s.	83.03	83.39	82.29	80.29	n.s.	n.s.
Years of education (mean, SD)	13.12 (3.14)	13.35 (3.06)	11.95 (3.30)	12.36 (3.25)	***	13.05 (2.85)	13.27 (2.75)	12.31 (2.84)	11.89 (3.44)	***	n.s.
Functional impairment	0.44 (1.26)	0.13 (0.53)	2.96 (2.25)	0.23 (0.74)	***	0.38 (1.17)	0.14 (0.63)	0.34 (1.08)	2.97 (2.15)	***	*
Negative exchanges with spouse
Too many demands on you	1.01 (0.86)	1.00 (0.84)	1.05 (0.94)	1.15 (0.90)	*	1.00 (0.91)	0.96 (0.88)	1.30 (1.00)	0.95 (0.92)	***	n.s.
Criticize you	1.12 (0.84)	1.11 (0.83)	1.17 (0.91)	1.15 (0.88)	n.s.	0.98 (0.88)	0.97 (0.87)	1.05 (0.92)	1.08 (0.87)	*	***
Let you down	0.49 (0.76)	0.46 (0.73)	0.57 (0.85)	0.66 (0.87)	***	0.72 (0.85)	0.71 (0.84)	0.79 (0.91)	0.75 (0.85)	n.s.	***
Gets on your nerves	0.90 (0.77)	0.87 (0.76)	1.02 (0.81)	1.01 (0.74)	***	1.14 (0.81)	1.12 (0.79)	1.28 (0.91)	1.16 (0.84)	**	***
Depressive symptoms	0.95 (1.54)	0.76 (1.34)	2.28 (2.19)	1.04 (1.48)	***	1.20 (1.81)	1.01 (1.64)	1.51 (2.04)	2.77 (2.34)	***	***