Predictors of Secondary Role Strains Among Spousal Caregivers of Older Adults With Functional Disability

Predictors of Secondary Role Strains Among Spousal Caregivers of Older Adults With Functional... Abstract Background and Objectives Aging spouses commonly care for a partner with functional disability, but little is known about how spousal caregiving may impact different life domains. This study evaluated how caregiving characteristics are associated with secondary role strains among spousal caregivers. Research Design and Methods This cross-sectional study examined 367 spousal caregivers and their partners from the 2011 National Health and Aging Trends Study and National Study of Caregiving. Hierarchical regressions were estimated to determine how caregiver background factors (sociodemographics, health conditions) along with primary objective (care activities, care recipient health conditions, and dementia status) and subjective (emotional caregiving difficulties, role overload) stressors are linked to care-related valued activity restriction, negative caregiving relationship quality, and care-related family disagreements. Gender differences were considered. Results After accounting for all predictors, older caregivers and caregivers providing more help with activities of daily living and health system interactions (e.g., scheduling appointments) were more likely to report activity restriction, whereas caregivers with more emotional difficulties reported higher negative caregiving relationship quality. Role overload was positively associated with all three secondary strains. For husbands only, caring for a partner with more chronic conditions was linked to higher negative caregiving relationship quality and caring for a partner with dementia was associated with a greater likelihood of family disagreements. Discussion and Implications Secondary role strains may develop through similar and unique pathways for caregiving wives and husbands. Further research is needed to identify those who could benefit from support in managing their care responsibilities alongside other life areas. Caregiving-informal, Caregiver stress, Disabilities, Marriage, Family issues, Gender Introduction Nearly 18 million U.S. adults provide family care for a person aged 65 or older due to limitations in physical, mental, or cognitive capacities (Schulz & Eden, 2016). When spouses or cohabiting partners are available, they often assist with a variety of activities including personal care, household chores, and medical tasks (e.g., managing medication; Reinhard, Levine, & Samis, 2014; Schulz & Eden, 2016). Although it may be perceived as a normative part of marriage, spousal caregiving can take a heavy toll on well-being. Relative to adult child caregivers, caregiving spouses report higher depressive symptoms and worse physical health (Pinquart & Sörenson, 2007, 2011). Spousal caregivers are also more likely to experience chronic conditions and declines in physical function that diminish their ability to provide care (Navaie-Waliser et al., 2002; Oldenkamp et al., 2016; Piette, Rosland, Silveira, Kabeto, & Langa, 2010). Yet, relatively little is known about which aspects of spousal caregiving may be most consequential for other important life areas, such as caregivers’ valued activity engagement and the quality of their marital and family relationships (Schulz & Eden, 2016). Understanding the broader implications of the spousal care role for aging wives and husbands would enhance efforts to sustain the long-term well-being of a particularly vulnerable caregiving subgroup. Guided by stress process models of caregiving, this study examined how spousal caregivers’ background characteristics and caregiving stressors are linked to care-related valued activity restriction, negative caregiving relationship quality, and care-related family disagreements. Stress Process Perspectives on Spousal Caregiving According to stress process models, caring for an older spouse with functional disability is a chronic source of stress that can lead to emotional and physical problems (Aneshensel et al., 1995; Pearlin, Mullan, Semple, & Skaff, 1990). As we describe below, however, the extent to which caregivers experience adverse outcomes depends in part on several key components of the care situation. Caregiver Background Characteristics Background characteristics reflect caregivers’ exposure to various stressors as well as their resources to manage stress (Aneshensel et al., 1995; Pearlin et al., 1990). As such, these factors help to identify caregivers who may be especially likely to suffer negative health consequences. Research has shown, for example, that caregivers who are older, less educated, and employed tend to report greater emotional distress, higher burden, and worse physical health (e.g., Chappell, Dujela, & Smith, 2015; Hsu et al., 2014; Pinquart & Sörensen, 2007; Schulz & Eden, 2016). Likewise, caregivers’ own health problems are linked to greater caregiving burden and may hinder their ability to cope with care-related stress (Navaie-Waliser et al., 2002; Pinquart & Sörensen, 2011; Roth, Fredman, & Haley, 2015). Primary Stressors Primary stressors are difficulties directly related to providing care, and involve objective and subjective elements. Primary objective stressors such as care tasks and care recipient impairments are considered to be the starting point of the caregiving stress process (Aneshensel et al., 1995). Primary subjective stressors represent caregivers’ emotional and psychological responses to these stressors, including self-appraisals of caregiving as difficult or too much to handle. Both types of primary stressors have been linked to adverse mental and physical health outcomes (Pinquart & Sörensen, 2006, 2007, 2011; Zarit, Polenick, DePasquale, Liu, & Bangerter, in press). Differentiating between these stressors aids in understanding how the caregiving stress process unfolds, which is valuable in determining targets for interventions to mitigate or prevent care-related distress. Moreover, stress process models posit that primary objective stressors lead to negative outcomes through increases in primary subjective stressors (Aneshensel et al., 1995), forming the basis for exploring potential mechanistic pathways. Secondary Role Strains Caregivers commonly encounter secondary role strains, or hardships experienced in roles and activities indirectly related to caregiving that emanate from primary care-related stressors (Aneshensel et al., 1995; Pearlin et al., 1990). That is, primary caregiving stress proliferates or “spreads” to other life areas. The time, energy, and resources devoted to providing care, for instance, can restrict caregivers’ participation in valued noncare activities (Gaugler et al., 2008; Schulz & Eden, 2016). Reduced participation is also partly a result of caregivers’ feelings of guilt about engaging in noncare activities and/or their reluctance to partake in activities without their partner (Badr et al., 2016; Rudman, Hebert, & Reid, 2006). Cross-sectional studies show that caregivers’ activity restriction is associated with poor health indicators such as higher blood pressure and worse sleep quality (Chattillion et al., 2013; Moore et al., 2011). These associations are likely dynamic and bidirectional, as evidenced by the reciprocal links between care-related activity restriction and depressive symptoms over time (Mausbach et al., 2011; Smith, Williamson, Miller, & Schulz, 2011). Social isolation could also result from activity restriction and has been associated with an increased risk of detrimental outcomes including coronary heart disease, stroke, and mortality (Holt-Lunstad, Smith, Baker, Harris, & Stephenson, 2015; Valtorta, Kanaan, Gilbody, Ronzi, & Hanratty, 2016). Therefore, it is essential to consider how caregiving stress proliferation may restrict spouses’ valued activities. Caregiving stress proliferation may also have reverberating adverse consequences for family relationships. Within the care dyad, a partner’s chronic illness or disability is a shared stressor that can negatively impact marital interactions (Baucom, Porter, Kirby, & Hudepohl, 2012; Berg & Upchurch, 2007). The stress of giving and receiving care, for example, may aggravate preexisting relationship issues and/or contribute to spousal discord that stems from shifting roles and activities in the marriage (Lyons, Zarit, Sayer, & Whitlatch, 2002; Quinn, Clare, & Woods, 2009; Savla, Roberto, Blieszner, Cox, & Gwazdauskas, 2011). Caregiving spouses may also resent their care role because they view their partner as controlling or demanding (Smith et al., 2011). Such negative relationship qualities can amplify caregivers’ emotional distress and result in potentially harmful caregiving behaviors (e.g., yelling, hitting) that pose a serious threat to care recipients’ long-term health and well-being (Smith et al., 2011). Beyond the care dyad, care-related conflict can arise within the larger family system. Although few studies have considered this topic, existing research indicates that caregiving stress is diffused throughout the family and not just limited to active helpers (Schulz & Eden, 2016; Zarit et al., in press). Consequently, families may disagree about multiple aspects of the care situation, including the amount and quality of care, decisions about medical treatment, and the adequacy of one another’s contributions (Aneshensel et al., 1995; Gwyther & Matchar, 2015; Pearlin et al., 1990; Semple, 1992). These disagreements can be an enduring source of stress for caregivers (Aneshensel et al., 1995; Scharlach, Li, & Dalvi, 2006) and reduce the support they receive from their family networks (Gaugler, Mendiondo, Smith, & Schmitt, 2003). As a whole, the literature on marital and family stress in the caregiving context underscores the need for a better understanding of factors linked to these secondary role strains for spousal caregivers. Caregiver Gender as a Moderator of Caregiving Stress Proliferation The nature and consequences of spousal caregiving often vary by gender (e.g., Aneshensel et al., 1995; Calasanti, 2010; Chappell et al., 2015; Pinquart & Sörensen, 2006; Swinkels, Verbakel, & Broese van Groenou, 2017; Thomeer, Reczek, & Umberson, 2015). Caregiving wives and husbands may therefore differ in how their background characteristics and primary care-related stressors are associated with secondary role strains. Compared with men, women perceive more societal expectation to place family members’ needs above their own as well as more family barriers to self-care (Calasanti, 2010; Rosland, Heisler, Choi, Silveira, & Piette, 2010; Thomeer et al., 2015). Women also provide more hours of caregiving than men, help with a higher number and wider range of care tasks, and experience more secondary stressors (Pinquart & Sörensen, 2006; Swinkels et al., 2017). Furthermore, whereas women tend to be emotionally entrenched in the care role, men commonly adopt a managerial caregiving style that emphasizes delegating tasks to other helpers and preserves their own well-being (Calasanti, 2010; Thompson, 2002). Perhaps in part as a result, caregiving women report higher caregiver burden and depressive symptoms, lower subjective well-being, and worse overall physical health than their male counterparts (Pinquart & Sörensen, 2006). In order to more effectively address the needs of spousal caregivers in managing their care role along with other central life domains, it is imperative to evaluate whether and how the experiences of caregiving wives and husbands differ. Considering women’s greater care demands in tandem with their heightened vulnerability to care-related stress, we propose that caregivers’ background characteristics and primary caregiving stressors may have stronger associations with secondary role strains for wives than for husbands. The Present Study Drawing from a nationally representative U.S. sample of spousal caregivers and their partners with functional disability, we examined the associations among caregivers’ background characteristics, primary caregiving stressors, and secondary role strains (care-related valued activity restriction, negative caregiving relationship quality, and care-related family disagreements). We predicted the following: 1) Caregivers’ older age, lower educational attainment, employment, and higher number of chronic health conditions will be positively associated with secondary role strains. 2) Caregivers’ higher levels of primary objective and subjective care-related stressors will be positively associated with secondary role strains. 3) Primary subjective stressors will mediate associations between primary objective stressors and secondary role strains, accounting for background characteristics. 4) The above associations will be significantly stronger for wives than for husbands. Design and Methods Sample and Procedures The sample for this cross-sectional study included 367 community-dwelling adults aged 65 and older with functional disability and their spousal caregivers drawn from the 2011 National Health and Aging Trends Study (NHATS) and the National Study of Caregiving (NSOC). Participants were eligible for NHATS if they were Medicare enrollees aged 65 and older, lived in the contiguous United States, and received health-related assistance in the last month for one or more of the following: mobility, personal care, and/or household chores. Participants were recruited from a Medicare enrollment database using a stratified three-stage sampling design. NHATS participants were eligible for NSOC if they had at least one family or unpaid nonfamily helper who provided health-related assistance with mobility, self-care, household chores, and/or medical care activities. The 2,423 eligible NHATS participants had 4,935 eligible caregivers. Contact information was provided for 3,362 (68.1%) caregivers, from which 2,007 (59.7%) completed a 30-minute telephone interview. In total, 519 (37.9%) of the 1,368 NHATS participants with one or more caregivers who participated in NSOC were married. Of these, 422 (81.3%) spousal caregivers were interviewed. We were interested in spouses of community-dwelling older adults who lived together, and so we removed 32 spouses whose care recipients lived in residential care and two spouses who did not live with their care recipient. A total of 388 spousal caregivers resided with their partner in the community. Of these, 21 were removed due to missing data on study variables. Hence, the analytic sample included 367 spousal caregivers and care recipients with complete data. On average, caregivers had been married to the care recipient for 42.97 years (SD = 18.07, range = 0–72). Table 1 shows caregivers’ background characteristics and scores on study variables. Table 1. Background Characteristics and Key Variables for Caregiving Wives and Husbands Characteristic  Wives (n = 218)  Husbands (n = 149)  t  df  M  SD  M  SD  CG age in years  71.81  9.63  76.22  7.06  −4.89***  357.40  CR age in years  78.00  7.44  75.58  7.13  3.12**  365.00  CG chronic health conditions  2.22  1.43  1.96  1.39  1.73  365.00  CR chronic health conditions  3.04  1.51  3.09  1.39  −0.37  365.00  CG ADL assistance  1.84  2.03  1.49  1.57  1.83  359.45  CG IADL assistance  3.73  0.99  2.98  1.27  6.11***  265.95  CG health system tasks  2.82  1.60  2.11  1.61  4.14***  365.00  CG medical/nursing tasks  2.84  2.06  1.91  1.66  4.79***  355.07  CG emotional difficulties  1.69  1.80  0.80  1.34  5.44***  362.42  CG role overload  1.78  0.65  1.40  0.49  6.27***  361.40  CG negative caregiving relationship quality  2.41  0.84  2.08  0.70  4.10***  350.72    Proportions    χ2  df  CG educational attainment   High school graduate  0.34  0.24  4.40*  1   College graduate  0.08  0.13  0.78  1   Post graduate  0.06  0.09  2.38  1  CG employed for pay  0.16  0.14  0.26  1  CR dementia status  0.22  0.11  6.84**  1  CG valued activity restriction  0.32  0.17  9.85**  1  CG family disagreements  0.13  0.13  0.001  1  Characteristic  Wives (n = 218)  Husbands (n = 149)  t  df  M  SD  M  SD  CG age in years  71.81  9.63  76.22  7.06  −4.89***  357.40  CR age in years  78.00  7.44  75.58  7.13  3.12**  365.00  CG chronic health conditions  2.22  1.43  1.96  1.39  1.73  365.00  CR chronic health conditions  3.04  1.51  3.09  1.39  −0.37  365.00  CG ADL assistance  1.84  2.03  1.49  1.57  1.83  359.45  CG IADL assistance  3.73  0.99  2.98  1.27  6.11***  265.95  CG health system tasks  2.82  1.60  2.11  1.61  4.14***  365.00  CG medical/nursing tasks  2.84  2.06  1.91  1.66  4.79***  355.07  CG emotional difficulties  1.69  1.80  0.80  1.34  5.44***  362.42  CG role overload  1.78  0.65  1.40  0.49  6.27***  361.40  CG negative caregiving relationship quality  2.41  0.84  2.08  0.70  4.10***  350.72    Proportions    χ2  df  CG educational attainment   High school graduate  0.34  0.24  4.40*  1   College graduate  0.08  0.13  0.78  1   Post graduate  0.06  0.09  2.38  1  CG employed for pay  0.16  0.14  0.26  1  CR dementia status  0.22  0.11  6.84**  1  CG valued activity restriction  0.32  0.17  9.85**  1  CG family disagreements  0.13  0.13  0.001  1  Notes. ADL = activities of daily living; CG = caregiver; CR = care recipient; df = degree of freedom; IADL = instrumental activities of daily living; M = mean; SD = standard deviation. N = 367 spousal caregivers. *p ≤ .05. **p ≤ .01. ***p ≤ .001. View Large Table 1. Background Characteristics and Key Variables for Caregiving Wives and Husbands Characteristic  Wives (n = 218)  Husbands (n = 149)  t  df  M  SD  M  SD  CG age in years  71.81  9.63  76.22  7.06  −4.89***  357.40  CR age in years  78.00  7.44  75.58  7.13  3.12**  365.00  CG chronic health conditions  2.22  1.43  1.96  1.39  1.73  365.00  CR chronic health conditions  3.04  1.51  3.09  1.39  −0.37  365.00  CG ADL assistance  1.84  2.03  1.49  1.57  1.83  359.45  CG IADL assistance  3.73  0.99  2.98  1.27  6.11***  265.95  CG health system tasks  2.82  1.60  2.11  1.61  4.14***  365.00  CG medical/nursing tasks  2.84  2.06  1.91  1.66  4.79***  355.07  CG emotional difficulties  1.69  1.80  0.80  1.34  5.44***  362.42  CG role overload  1.78  0.65  1.40  0.49  6.27***  361.40  CG negative caregiving relationship quality  2.41  0.84  2.08  0.70  4.10***  350.72    Proportions    χ2  df  CG educational attainment   High school graduate  0.34  0.24  4.40*  1   College graduate  0.08  0.13  0.78  1   Post graduate  0.06  0.09  2.38  1  CG employed for pay  0.16  0.14  0.26  1  CR dementia status  0.22  0.11  6.84**  1  CG valued activity restriction  0.32  0.17  9.85**  1  CG family disagreements  0.13  0.13  0.001  1  Characteristic  Wives (n = 218)  Husbands (n = 149)  t  df  M  SD  M  SD  CG age in years  71.81  9.63  76.22  7.06  −4.89***  357.40  CR age in years  78.00  7.44  75.58  7.13  3.12**  365.00  CG chronic health conditions  2.22  1.43  1.96  1.39  1.73  365.00  CR chronic health conditions  3.04  1.51  3.09  1.39  −0.37  365.00  CG ADL assistance  1.84  2.03  1.49  1.57  1.83  359.45  CG IADL assistance  3.73  0.99  2.98  1.27  6.11***  265.95  CG health system tasks  2.82  1.60  2.11  1.61  4.14***  365.00  CG medical/nursing tasks  2.84  2.06  1.91  1.66  4.79***  355.07  CG emotional difficulties  1.69  1.80  0.80  1.34  5.44***  362.42  CG role overload  1.78  0.65  1.40  0.49  6.27***  361.40  CG negative caregiving relationship quality  2.41  0.84  2.08  0.70  4.10***  350.72    Proportions    χ2  df  CG educational attainment   High school graduate  0.34  0.24  4.40*  1   College graduate  0.08  0.13  0.78  1   Post graduate  0.06  0.09  2.38  1  CG employed for pay  0.16  0.14  0.26  1  CR dementia status  0.22  0.11  6.84**  1  CG valued activity restriction  0.32  0.17  9.85**  1  CG family disagreements  0.13  0.13  0.001  1  Notes. ADL = activities of daily living; CG = caregiver; CR = care recipient; df = degree of freedom; IADL = instrumental activities of daily living; M = mean; SD = standard deviation. N = 367 spousal caregivers. *p ≤ .05. **p ≤ .01. ***p ≤ .001. View Large Measures Caregiver Gender Gender (1 = female, 0 = male) was examined as a moderator to consider whether associations significantly differed for caregiving wives and husbands. Caregiver Background Characteristics Sociodemographics included caregivers’ age, educational attainment (1 = no schooling completed to 9 = masters, professional, or doctoral degree), and employment status (1 = worked for pay in the last week, 0 = did not work for pay in the last week). Caregivers reported whether they had been diagnosed with seven chronic health conditions (1 = yes, 0 = no): arthritis, cancer, diabetes, heart disease, hypertension, lung disease, and osteoporosis. Summed scores were calculated. Primary Objective Caregiving Stressors Caregivers reported on the number of activities of daily living (ADLs) and instrumental activities of daily living (IADLs) during which they assisted their partner. ADLs included bathing, dressing, eating, toileting, getting in/out of bed, and mobility inside/outside the house. IADLs included laundry, shopping, preparing meals, banking, and managing money. Separate summed scores were created for assistance with ADLs (range = 0–7) and IADLs (range = 0–5). We also examined two types of medical care activities. Health system interaction tasks included ordering medication, scheduling appointments, speaking to providers, helping to change/add insurance or drug plans, and other medical insurance matters. Medical/nursing tasks included giving shots/injections, wound/sore care, teeth/denture care, foot care (e.g., clipping nails), managing medical tasks (e.g., ostomy care, IVs, testing blood), monitoring medications, helping with prescribed exercises, and assisting with a special diet. Spouses reported whether they helped with each task (1 = yes, 0 = no). Separate summed scores were determined for health system interaction tasks (range = 0–5) and medical/nursing tasks (range = 0–8). Lastly, we evaluated two indicators of care recipients’ health and functioning based on care recipient (n = 315) or proxy (n = 50 spouses; n = 1 adult daughter; n = 1 adult son) reports. Dementia status was determined from reports of a physician diagnosis (1 = dementia, 0 = no dementia). Care recipients (or proxies) also reported whether they had been diagnosed with eight chronic health conditions (1 = yes, 0 = no): arthritis, cancer, diabetes, heart disease, hypertension, lung disease, osteoporosis, and stroke. Summed scores were calculated. Primary Subjective Caregiving Stressors Emotional caregiving difficulties were assessed with two items. Caregivers were asked whether caregiving was emotionally difficult for them (1 = yes, 0 = no). Caregivers who reported emotional caregiving difficulties were then asked to rate their extent from 1 (a little difficult) to 5 (very difficult). We combined these items such that caregivers who did not report difficulties were given a zero and caregivers who reported difficulties were assigned a score based on their degree (0 = no difficulty to 5 = high difficulty). As a measure of role overload (Pearlin et al., 1990), we also considered three items assessing how much caregivers feel they (a) are exhausted when they go to bed at night; (b) have more things to do than they can handle; and (c) do not have time for themselves (1 = very much, 2 = somewhat, 3 = not so much). Items were reverse coded and averaged so that higher scores represented greater role overload (α = .75). Secondary Role Strains Care-related valued activity restriction was ascertained from three items. First, caregivers reported whether they participated in four activities in the last month (1 = yes, 0 = no): (a) visiting in person with friends or family; (b) attending religious services; (c) participating in club meetings, classes, or group activities; and (d) going out for enjoyment. Caregivers were then asked whether their care tasks kept them from participating in these activities in the last month (1 = yes, 0 = no), along with the importance they placed on each activity (1 = very important, 2 = somewhat important, 3 = not so important). Most caregivers (73.8%) reported no care-related restriction in their valued activities (i.e., activities rated as “somewhat” or “very” important). As such, we created a binary variable to examine whether caregivers were kept from participating in one or more activities they valued (1 = yes, 0 = no). Negative caregiving relationship quality included two items. Caregivers reported how much the care recipient (a) argues with them; and (b) gets on their nerves from 1 (a lot) to 4 (not at all). Items were reverse coded and averaged such that higher scores reflected greater negative relationship quality. The Spearman–Brown estimate, which is recommended as the appropriate reliability estimate for two-item scales (Eisinga, Grotenhuis, & Pelzer, 2013), was .63. Care-related family disagreements were measured with one item assessing, in general, how much the caregiver’s family disagreed over the care recipient’s care (1 = very much, 2 = somewhat, 3 = not so much). Most caregivers (86.6%) reported that their family disagreed “not so much,” and so we created a binary variable to assess whether family disagreements occurred “somewhat” or “very much” (1 = yes, 0 = no). Statistical Analysis Hierarchical regressions were conducted to evaluate how the addition of each component of our conceptual model contributed to explained variance in the outcome. Regression analyses were performed in SAS 9.4 using the NSOC analytic weight and statistical procedures to account for the complex survey design (Kasper, Freedman, & Spillman, 2013). Linear regressions were estimated to examine how predictors were associated with levels of negative caregiving relationship quality, whereas logistic regressions were estimated to determine their associations with the likelihood of experiencing care-related valued activity restriction or family disagreements. Caregiver gender and background characteristics (age, educational attainment, employment status, chronic health conditions) were entered in Model 1, primary objective caregiving stressors (assistance with ADLs and IADLs, health system interaction tasks, and medical/nursing tasks; care recipients’ dementia status and chronic health conditions) were included in Model 2, and primary subjective caregiving stressors (emotional caregiving difficulties and role overload) were added in Model 3. The first step of the models estimated main effects of the predictors and the second step evaluated gender differences with two-way interaction terms (e.g., Caregiver Age × Caregiver Gender) for the set of predictors added to each model. To test the mediating role of primary subjective stressors in the links between primary objective stressors and secondary role strains, linear regressions were also estimated with caregiver background characteristics as covariates, primary objective stressors as predictors, and emotional caregiving difficulties and role overload as outcomes. The first step examined main effects, whereas the second step tested gender differences with two-way interaction terms between each primary objective stressor and caregiver gender. The nature of significant interactions was explored by estimating simple slopes for wives and husbands. Continuous predictors were mean centered. In accordance with the procedures outlined by Baron and Kenny (1986), mediation occurred if in the aforementioned models: (a) primary objective stressors were significantly linked to secondary role strains before primary subjective stressors were added as predictors; (b) primary objective stressors were significantly associated with primary subjective stressors; (c) primary subjective stressors were significantly linked to secondary role strains with primary objective stressors included as predictors; and (d) associations between primary objective stressors and secondary role strains were reduced to nonsignificance when primary subjective stressors were entered as predictors. Partial mediation resulted when the first three criteria were met, whereas full mediation was achieved through meeting all four criteria. To evaluate whether the mediating process differed for wives and husbands, mediated moderation was examined using the guidelines proposed by Muller, Judd, and Yzerbyt (2005). Mediated moderation occurred if mediation was present, and the direct path along with one or both indirect paths from primary objective stressors to secondary role strains (i.e., through primary subjective stressors) were moderated by gender. Results Independent t-tests and Pearson chi-square tests were examined in preliminary analyses to consider gender differences in study variables (Table 1). Compared with caregiving husbands, caregiving wives were younger, less educated, and cared for older spouses. Wives were more likely to care for a partner with dementia and provided more help with IADLs, health system interaction tasks, and medical/nursing tasks. In addition, wives reported greater emotional caregiving difficulties and role overload. Finally, wives perceived more negative caregiving relationship quality and were more likely to report care-related valued activity restriction. Regression models are presented in Tables 2–4 to ascertain how caregiver background characteristics and primary stressors were associated with secondary role strains, and in Table 5 to explore primary subjective stressors as potential mediators of the associations between primary objective stressors and secondary role strains. Unstandardized coefficients, standard errors, confidence intervals (CIs), and accounted variance are shown from the first step of each model. As measures of effect size, standardized betas are reported for the linear regressions and odds ratios (ORs) are reported for the logistic regressions. Table 2. Predictors of Care-Related Valued Activity Restriction for Spousal Caregivers Predictors  Model 1  Model 2  Model 3  B (SE)  OR  95% CI  B (SE)  OR  95% CI  B (SE)  OR  95% CI  Background characteristics   CG gender (female)  0.85 (0.38)*  2.35  1.09, 5.04  0.79 (0.42)  2.19  0.94, 5.11  0.31 (0.43)  1.37  0.58, 3.24   CG age  0.03 (0.02)  1.03  0.99, 1.06  0.04 (0.02)  1.04  1.00, 1.09  0.05 (0.02)*  1.05  1.00, 1.10   CG educational attainment  0.18 (0.08)*  1.20  1.02, 1.41  0.19 (0.09)*  1.21  1.00, 1.46  0.15 (0.10)  1.17  0.96, 1.41   CG employment status  −0.15 (0.41)  0.86  0.38, 1.97  0.18 (0.45)  1.20  0.48, 2.97  −0.01 (0.43)  0.99  0.42, 2.37   CG chronic conditions  0.21 (0.14)  1.23  0.94, 1.61  0.23 (0.13)  1.26  0.97, 1.63  0.18 (0.13)  1.20  0.92, 1.55  Primary objective stressors   CR chronic conditions  —  —  —  −0.004 (0.11)  1.00  0.81, 1.23  −0.05 (0.11)  0.95  0.76, 1.18   CR dementia status  —  —  —  0.01 (0.37)  1.01  0.49, 2.12  −0.22 (0.41)  0.81  0.35, 1.85   CG ADL assistance  —  —  —  0.25 (0.09)**  1.28  1.07, 1.54  0.19 (0.09)*  1.21  1.00, 1.46   CG IADL assistance  —  —  —  −0.23 (0.14)  0.80  0.61, 1.05  −0.20 (0.15)  0.82  0.61, 1.10   CG health system tasks  —  —  —  0.62 (0.14)***  1.85  1.39, 2.47  0.57 (0.15)***  1.76  1.30, 2.39   CG medical/nursing tasks  —  —  —  0.07 (0.09)  1.07  0.89, 1.29  0.02 (0.09)  1.02  0.85, 1.23  Primary subjective stressors   CG emotional difficulties  –  –  –  –  –  –  0.17 (0.13)  1.18  0.92, 1.52   CG role overload  –  –  –  –  –  –  1.13 (0.36)**  3.08  1.50, 6.32    Total R2  .06  .23  .29  Predictors  Model 1  Model 2  Model 3  B (SE)  OR  95% CI  B (SE)  OR  95% CI  B (SE)  OR  95% CI  Background characteristics   CG gender (female)  0.85 (0.38)*  2.35  1.09, 5.04  0.79 (0.42)  2.19  0.94, 5.11  0.31 (0.43)  1.37  0.58, 3.24   CG age  0.03 (0.02)  1.03  0.99, 1.06  0.04 (0.02)  1.04  1.00, 1.09  0.05 (0.02)*  1.05  1.00, 1.10   CG educational attainment  0.18 (0.08)*  1.20  1.02, 1.41  0.19 (0.09)*  1.21  1.00, 1.46  0.15 (0.10)  1.17  0.96, 1.41   CG employment status  −0.15 (0.41)  0.86  0.38, 1.97  0.18 (0.45)  1.20  0.48, 2.97  −0.01 (0.43)  0.99  0.42, 2.37   CG chronic conditions  0.21 (0.14)  1.23  0.94, 1.61  0.23 (0.13)  1.26  0.97, 1.63  0.18 (0.13)  1.20  0.92, 1.55  Primary objective stressors   CR chronic conditions  —  —  —  −0.004 (0.11)  1.00  0.81, 1.23  −0.05 (0.11)  0.95  0.76, 1.18   CR dementia status  —  —  —  0.01 (0.37)  1.01  0.49, 2.12  −0.22 (0.41)  0.81  0.35, 1.85   CG ADL assistance  —  —  —  0.25 (0.09)**  1.28  1.07, 1.54  0.19 (0.09)*  1.21  1.00, 1.46   CG IADL assistance  —  —  —  −0.23 (0.14)  0.80  0.61, 1.05  −0.20 (0.15)  0.82  0.61, 1.10   CG health system tasks  —  —  —  0.62 (0.14)***  1.85  1.39, 2.47  0.57 (0.15)***  1.76  1.30, 2.39   CG medical/nursing tasks  —  —  —  0.07 (0.09)  1.07  0.89, 1.29  0.02 (0.09)  1.02  0.85, 1.23  Primary subjective stressors   CG emotional difficulties  –  –  –  –  –  –  0.17 (0.13)  1.18  0.92, 1.52   CG role overload  –  –  –  –  –  –  1.13 (0.36)**  3.08  1.50, 6.32    Total R2  .06  .23  .29  Note. ADL = activities of daily living; CG = caregiver; CI = confidence interval; CR = care recipient; IADL = independent activities of daily living; OR = odds ratio; SE = standard error. Estimates are from Step 1 of each model. Step 2 (not shown in table) included gender interactions with background characteristics in Model 1, gender interactions with primary objective stressors in Model 2, and gender interactions with primary subjective stressors in Model 3. There were no significant gender interactions in Step 2. N = 367 spousal caregivers. *p ≤ .05. **p ≤ .01. ***p ≤ .001. View Large Table 2. Predictors of Care-Related Valued Activity Restriction for Spousal Caregivers Predictors  Model 1  Model 2  Model 3  B (SE)  OR  95% CI  B (SE)  OR  95% CI  B (SE)  OR  95% CI  Background characteristics   CG gender (female)  0.85 (0.38)*  2.35  1.09, 5.04  0.79 (0.42)  2.19  0.94, 5.11  0.31 (0.43)  1.37  0.58, 3.24   CG age  0.03 (0.02)  1.03  0.99, 1.06  0.04 (0.02)  1.04  1.00, 1.09  0.05 (0.02)*  1.05  1.00, 1.10   CG educational attainment  0.18 (0.08)*  1.20  1.02, 1.41  0.19 (0.09)*  1.21  1.00, 1.46  0.15 (0.10)  1.17  0.96, 1.41   CG employment status  −0.15 (0.41)  0.86  0.38, 1.97  0.18 (0.45)  1.20  0.48, 2.97  −0.01 (0.43)  0.99  0.42, 2.37   CG chronic conditions  0.21 (0.14)  1.23  0.94, 1.61  0.23 (0.13)  1.26  0.97, 1.63  0.18 (0.13)  1.20  0.92, 1.55  Primary objective stressors   CR chronic conditions  —  —  —  −0.004 (0.11)  1.00  0.81, 1.23  −0.05 (0.11)  0.95  0.76, 1.18   CR dementia status  —  —  —  0.01 (0.37)  1.01  0.49, 2.12  −0.22 (0.41)  0.81  0.35, 1.85   CG ADL assistance  —  —  —  0.25 (0.09)**  1.28  1.07, 1.54  0.19 (0.09)*  1.21  1.00, 1.46   CG IADL assistance  —  —  —  −0.23 (0.14)  0.80  0.61, 1.05  −0.20 (0.15)  0.82  0.61, 1.10   CG health system tasks  —  —  —  0.62 (0.14)***  1.85  1.39, 2.47  0.57 (0.15)***  1.76  1.30, 2.39   CG medical/nursing tasks  —  —  —  0.07 (0.09)  1.07  0.89, 1.29  0.02 (0.09)  1.02  0.85, 1.23  Primary subjective stressors   CG emotional difficulties  –  –  –  –  –  –  0.17 (0.13)  1.18  0.92, 1.52   CG role overload  –  –  –  –  –  –  1.13 (0.36)**  3.08  1.50, 6.32    Total R2  .06  .23  .29  Predictors  Model 1  Model 2  Model 3  B (SE)  OR  95% CI  B (SE)  OR  95% CI  B (SE)  OR  95% CI  Background characteristics   CG gender (female)  0.85 (0.38)*  2.35  1.09, 5.04  0.79 (0.42)  2.19  0.94, 5.11  0.31 (0.43)  1.37  0.58, 3.24   CG age  0.03 (0.02)  1.03  0.99, 1.06  0.04 (0.02)  1.04  1.00, 1.09  0.05 (0.02)*  1.05  1.00, 1.10   CG educational attainment  0.18 (0.08)*  1.20  1.02, 1.41  0.19 (0.09)*  1.21  1.00, 1.46  0.15 (0.10)  1.17  0.96, 1.41   CG employment status  −0.15 (0.41)  0.86  0.38, 1.97  0.18 (0.45)  1.20  0.48, 2.97  −0.01 (0.43)  0.99  0.42, 2.37   CG chronic conditions  0.21 (0.14)  1.23  0.94, 1.61  0.23 (0.13)  1.26  0.97, 1.63  0.18 (0.13)  1.20  0.92, 1.55  Primary objective stressors   CR chronic conditions  —  —  —  −0.004 (0.11)  1.00  0.81, 1.23  −0.05 (0.11)  0.95  0.76, 1.18   CR dementia status  —  —  —  0.01 (0.37)  1.01  0.49, 2.12  −0.22 (0.41)  0.81  0.35, 1.85   CG ADL assistance  —  —  —  0.25 (0.09)**  1.28  1.07, 1.54  0.19 (0.09)*  1.21  1.00, 1.46   CG IADL assistance  —  —  —  −0.23 (0.14)  0.80  0.61, 1.05  −0.20 (0.15)  0.82  0.61, 1.10   CG health system tasks  —  —  —  0.62 (0.14)***  1.85  1.39, 2.47  0.57 (0.15)***  1.76  1.30, 2.39   CG medical/nursing tasks  —  —  —  0.07 (0.09)  1.07  0.89, 1.29  0.02 (0.09)  1.02  0.85, 1.23  Primary subjective stressors   CG emotional difficulties  –  –  –  –  –  –  0.17 (0.13)  1.18  0.92, 1.52   CG role overload  –  –  –  –  –  –  1.13 (0.36)**  3.08  1.50, 6.32    Total R2  .06  .23  .29  Note. ADL = activities of daily living; CG = caregiver; CI = confidence interval; CR = care recipient; IADL = independent activities of daily living; OR = odds ratio; SE = standard error. Estimates are from Step 1 of each model. Step 2 (not shown in table) included gender interactions with background characteristics in Model 1, gender interactions with primary objective stressors in Model 2, and gender interactions with primary subjective stressors in Model 3. There were no significant gender interactions in Step 2. N = 367 spousal caregivers. *p ≤ .05. **p ≤ .01. ***p ≤ .001. View Large Table 4. Predictors of Care-Related Family Disagreements for Spousal Caregivers Predictors  Model 1  Model 2  Model 3  B (SE)  OR  95% CI  B (SE)  OR  95% CI  B (SE)  OR  95% CI  Background characteristics   CG gender (female)  0.14 (0.34)  1.15  0.58, 2.27  −0.02 (0.47)  0.99  0.39, 2.51  −0.19 (0.45)  0.83  0.34, 2.04   CG age  −0.02 (0.03)  0.98  0.93, 1.03  −0.03 (0.03)  0.97  0.92, 1.02  −0.04 (0.03)  0.97  0.92, 1.02   CG educational attainment  −0.10 (0.11)  0.91  0.73, 1.13  −0.10 (0.11)  0.90  0.72, 1.13  −0.09 (0.11)  0.91  0.74, 1.14   CG employment status  −0.94 (0.71)  0.39  0.09, 1.62  −0.88 (0.70)  0.41  0.10, 1.68  −1.01 (0.73)  0.36  0.08, 1.58   CG chronic conditions  −0.03 (0.14)  0.97  0.74, 1.28  −0.06 (0.13)  0.95  0.73, 1.22  −0.03 (0.13)  0.97  0.75, 1.24  Primary objective stressors   CR chronic conditions  —  —  —  −0.06 (0.15)  0.94  0.70, 1.25  −0.09(0.15)  0.92  0.68, 1.23   CR dementia statusa  —  —  —  0.89 (0.54)  2.43  0.83, 7.11  0.92 (0.53)  2.51  0.87, 7.23   CG ADL assistance  —  —  —  0.01 (0.12)  1.01  0.79, 1.29  −0.004 (0.12)  0.98  0.78, 1.27   CG IADL assistance  —  —  —  −0.26 (0.19)  0.77  0.52, 1.14  −0.23 (0.19)  0.79  0.54, 1.16   CG health system tasks  —  —  —  0.11 (0.14)  1.12  0.85, 1.47  0.09 (0.15)  1.09  0.82, 1.46   CG medical/nursing tasks  —  —  —  0.18 (0.12)  1.20  0.94, 1.54  0.14 (0.12)  1.15  0.91, 1.47  Primary subjective stressors   CG emotional difficulties  —  —  —  —  —  —  −0.20 (0.14)  0.82  0.62, 1.07   CG role overload  —  —  —  —  —  —  0.88 (0.36)*  2.40  1.17, 4.94    Total R2  .02  .06  .09  Predictors  Model 1  Model 2  Model 3  B (SE)  OR  95% CI  B (SE)  OR  95% CI  B (SE)  OR  95% CI  Background characteristics   CG gender (female)  0.14 (0.34)  1.15  0.58, 2.27  −0.02 (0.47)  0.99  0.39, 2.51  −0.19 (0.45)  0.83  0.34, 2.04   CG age  −0.02 (0.03)  0.98  0.93, 1.03  −0.03 (0.03)  0.97  0.92, 1.02  −0.04 (0.03)  0.97  0.92, 1.02   CG educational attainment  −0.10 (0.11)  0.91  0.73, 1.13  −0.10 (0.11)  0.90  0.72, 1.13  −0.09 (0.11)  0.91  0.74, 1.14   CG employment status  −0.94 (0.71)  0.39  0.09, 1.62  −0.88 (0.70)  0.41  0.10, 1.68  −1.01 (0.73)  0.36  0.08, 1.58   CG chronic conditions  −0.03 (0.14)  0.97  0.74, 1.28  −0.06 (0.13)  0.95  0.73, 1.22  −0.03 (0.13)  0.97  0.75, 1.24  Primary objective stressors   CR chronic conditions  —  —  —  −0.06 (0.15)  0.94  0.70, 1.25  −0.09(0.15)  0.92  0.68, 1.23   CR dementia statusa  —  —  —  0.89 (0.54)  2.43  0.83, 7.11  0.92 (0.53)  2.51  0.87, 7.23   CG ADL assistance  —  —  —  0.01 (0.12)  1.01  0.79, 1.29  −0.004 (0.12)  0.98  0.78, 1.27   CG IADL assistance  —  —  —  −0.26 (0.19)  0.77  0.52, 1.14  −0.23 (0.19)  0.79  0.54, 1.16   CG health system tasks  —  —  —  0.11 (0.14)  1.12  0.85, 1.47  0.09 (0.15)  1.09  0.82, 1.46   CG medical/nursing tasks  —  —  —  0.18 (0.12)  1.20  0.94, 1.54  0.14 (0.12)  1.15  0.91, 1.47  Primary subjective stressors   CG emotional difficulties  —  —  —  —  —  —  −0.20 (0.14)  0.82  0.62, 1.07   CG role overload  —  —  —  —  —  —  0.88 (0.36)*  2.40  1.17, 4.94    Total R2  .02  .06  .09  Note. ADL = activities of daily living; CG = caregiver; CI = confidence interval; CR = care recipient; IADL = independent activities of daily living; OR = odds ratio; SE = standard error. Estimates are from Step 1 of each model. Step 2 (not shown in table) included gender interactions with background characteristics in Model 1, gender interactions with primary objective stressors in Model 2, and gender interactions with primary subjective stressors in Model 3. N = 367 spousal caregivers. aDenotes significant gender interaction found in Step 2. *p ≤ .05. **p ≤ .01. ***p ≤ .001. View Large Table 4. Predictors of Care-Related Family Disagreements for Spousal Caregivers Predictors  Model 1  Model 2  Model 3  B (SE)  OR  95% CI  B (SE)  OR  95% CI  B (SE)  OR  95% CI  Background characteristics   CG gender (female)  0.14 (0.34)  1.15  0.58, 2.27  −0.02 (0.47)  0.99  0.39, 2.51  −0.19 (0.45)  0.83  0.34, 2.04   CG age  −0.02 (0.03)  0.98  0.93, 1.03  −0.03 (0.03)  0.97  0.92, 1.02  −0.04 (0.03)  0.97  0.92, 1.02   CG educational attainment  −0.10 (0.11)  0.91  0.73, 1.13  −0.10 (0.11)  0.90  0.72, 1.13  −0.09 (0.11)  0.91  0.74, 1.14   CG employment status  −0.94 (0.71)  0.39  0.09, 1.62  −0.88 (0.70)  0.41  0.10, 1.68  −1.01 (0.73)  0.36  0.08, 1.58   CG chronic conditions  −0.03 (0.14)  0.97  0.74, 1.28  −0.06 (0.13)  0.95  0.73, 1.22  −0.03 (0.13)  0.97  0.75, 1.24  Primary objective stressors   CR chronic conditions  —  —  —  −0.06 (0.15)  0.94  0.70, 1.25  −0.09(0.15)  0.92  0.68, 1.23   CR dementia statusa  —  —  —  0.89 (0.54)  2.43  0.83, 7.11  0.92 (0.53)  2.51  0.87, 7.23   CG ADL assistance  —  —  —  0.01 (0.12)  1.01  0.79, 1.29  −0.004 (0.12)  0.98  0.78, 1.27   CG IADL assistance  —  —  —  −0.26 (0.19)  0.77  0.52, 1.14  −0.23 (0.19)  0.79  0.54, 1.16   CG health system tasks  —  —  —  0.11 (0.14)  1.12  0.85, 1.47  0.09 (0.15)  1.09  0.82, 1.46   CG medical/nursing tasks  —  —  —  0.18 (0.12)  1.20  0.94, 1.54  0.14 (0.12)  1.15  0.91, 1.47  Primary subjective stressors   CG emotional difficulties  —  —  —  —  —  —  −0.20 (0.14)  0.82  0.62, 1.07   CG role overload  —  —  —  —  —  —  0.88 (0.36)*  2.40  1.17, 4.94    Total R2  .02  .06  .09  Predictors  Model 1  Model 2  Model 3  B (SE)  OR  95% CI  B (SE)  OR  95% CI  B (SE)  OR  95% CI  Background characteristics   CG gender (female)  0.14 (0.34)  1.15  0.58, 2.27  −0.02 (0.47)  0.99  0.39, 2.51  −0.19 (0.45)  0.83  0.34, 2.04   CG age  −0.02 (0.03)  0.98  0.93, 1.03  −0.03 (0.03)  0.97  0.92, 1.02  −0.04 (0.03)  0.97  0.92, 1.02   CG educational attainment  −0.10 (0.11)  0.91  0.73, 1.13  −0.10 (0.11)  0.90  0.72, 1.13  −0.09 (0.11)  0.91  0.74, 1.14   CG employment status  −0.94 (0.71)  0.39  0.09, 1.62  −0.88 (0.70)  0.41  0.10, 1.68  −1.01 (0.73)  0.36  0.08, 1.58   CG chronic conditions  −0.03 (0.14)  0.97  0.74, 1.28  −0.06 (0.13)  0.95  0.73, 1.22  −0.03 (0.13)  0.97  0.75, 1.24  Primary objective stressors   CR chronic conditions  —  —  —  −0.06 (0.15)  0.94  0.70, 1.25  −0.09(0.15)  0.92  0.68, 1.23   CR dementia statusa  —  —  —  0.89 (0.54)  2.43  0.83, 7.11  0.92 (0.53)  2.51  0.87, 7.23   CG ADL assistance  —  —  —  0.01 (0.12)  1.01  0.79, 1.29  −0.004 (0.12)  0.98  0.78, 1.27   CG IADL assistance  —  —  —  −0.26 (0.19)  0.77  0.52, 1.14  −0.23 (0.19)  0.79  0.54, 1.16   CG health system tasks  —  —  —  0.11 (0.14)  1.12  0.85, 1.47  0.09 (0.15)  1.09  0.82, 1.46   CG medical/nursing tasks  —  —  —  0.18 (0.12)  1.20  0.94, 1.54  0.14 (0.12)  1.15  0.91, 1.47  Primary subjective stressors   CG emotional difficulties  —  —  —  —  —  —  −0.20 (0.14)  0.82  0.62, 1.07   CG role overload  —  —  —  —  —  —  0.88 (0.36)*  2.40  1.17, 4.94    Total R2  .02  .06  .09  Note. ADL = activities of daily living; CG = caregiver; CI = confidence interval; CR = care recipient; IADL = independent activities of daily living; OR = odds ratio; SE = standard error. Estimates are from Step 1 of each model. Step 2 (not shown in table) included gender interactions with background characteristics in Model 1, gender interactions with primary objective stressors in Model 2, and gender interactions with primary subjective stressors in Model 3. N = 367 spousal caregivers. aDenotes significant gender interaction found in Step 2. *p ≤ .05. **p ≤ .01. ***p ≤ .001. View Large Table 5. Predictors of Primary Subjective Stressors for Spousal Caregivers Predictors  Emotional difficulties  Role overload  B (SE)  β  95% CI  B (SE)  β  95% CI  Background characteristics   CG gender (female)  0.52 (0.19)**  0.16  0.14, 0.91  0.29 (0.07)***  0.24  0.16, 0.43   CG age  −0.01 (0.01)  −0.07  −0.04, 0.02  −0.002 (0.01)  −0.03  −0.01, 0.01   CG educational attainment  0.14 (0.05)**  0.17  0.04, 0.23  0.03 (0.02)  0.09  −0.01, 0.06   CG employment status  0.16 (0.25)  0.04  −0.35, 0.66  0.08 (0.09)  0.05  −0.10, 0.26   CG chronic conditions  0.18 (0.06)**  0.16  0.06, 0.31  0.02 (0.02)  0.05  −0.03, 0.06  Primary objective stressors   CR chronic conditions  0.09 (0.05)  0.08  −0.02, 0.19  0.04 (0.02)*  0.09  0.001, 0.08   CR dementia status  0.74 (0.26)**  0.15  0.23, 1.26  0.10 (0.10)  0.06  −0.10, 0.30   CG ADL assistance  0.10 (0.06)  0.11  −0.02, 0.23  0.05 (0.02)*  0.14  0.01, 0.09   CG IADL assistance  −0.04 (0.08)  −0.03  −0.19, 0.12  −0.03 (0.03)  −0.06  −0.08, 0.03   CG health system tasks  0.19 (0.05)***  0.19  0.09, 0.29  0.07 (0.02)***  0.19  0.03, 0.11   CG medical/nursing tasks  0.002 (0.04)  0.002  −0.08, 0.09  0.05 (0.02)**  0.16  0.01, 0.09    Total R2  .22  .26  Predictors  Emotional difficulties  Role overload  B (SE)  β  95% CI  B (SE)  β  95% CI  Background characteristics   CG gender (female)  0.52 (0.19)**  0.16  0.14, 0.91  0.29 (0.07)***  0.24  0.16, 0.43   CG age  −0.01 (0.01)  −0.07  −0.04, 0.02  −0.002 (0.01)  −0.03  −0.01, 0.01   CG educational attainment  0.14 (0.05)**  0.17  0.04, 0.23  0.03 (0.02)  0.09  −0.01, 0.06   CG employment status  0.16 (0.25)  0.04  −0.35, 0.66  0.08 (0.09)  0.05  −0.10, 0.26   CG chronic conditions  0.18 (0.06)**  0.16  0.06, 0.31  0.02 (0.02)  0.05  −0.03, 0.06  Primary objective stressors   CR chronic conditions  0.09 (0.05)  0.08  −0.02, 0.19  0.04 (0.02)*  0.09  0.001, 0.08   CR dementia status  0.74 (0.26)**  0.15  0.23, 1.26  0.10 (0.10)  0.06  −0.10, 0.30   CG ADL assistance  0.10 (0.06)  0.11  −0.02, 0.23  0.05 (0.02)*  0.14  0.01, 0.09   CG IADL assistance  −0.04 (0.08)  −0.03  −0.19, 0.12  −0.03 (0.03)  −0.06  −0.08, 0.03   CG health system tasks  0.19 (0.05)***  0.19  0.09, 0.29  0.07 (0.02)***  0.19  0.03, 0.11   CG medical/nursing tasks  0.002 (0.04)  0.002  −0.08, 0.09  0.05 (0.02)**  0.16  0.01, 0.09    Total R2  .22  .26  Note. ADL = activities of daily living; CG = caregiver; CI = confidence interval; CR = care recipient; IADL = independent activities of daily living; SE = standard error. Estimates are from Step 1 of each model. Step 2 (not shown in table) included gender interactions with each primary objective stressor. There were no significant gender interactions in Step 2. N = 367 spousal caregivers. *p ≤ .05. **p ≤ .01. ***p ≤ .001. View Large Table 5. Predictors of Primary Subjective Stressors for Spousal Caregivers Predictors  Emotional difficulties  Role overload  B (SE)  β  95% CI  B (SE)  β  95% CI  Background characteristics   CG gender (female)  0.52 (0.19)**  0.16  0.14, 0.91  0.29 (0.07)***  0.24  0.16, 0.43   CG age  −0.01 (0.01)  −0.07  −0.04, 0.02  −0.002 (0.01)  −0.03  −0.01, 0.01   CG educational attainment  0.14 (0.05)**  0.17  0.04, 0.23  0.03 (0.02)  0.09  −0.01, 0.06   CG employment status  0.16 (0.25)  0.04  −0.35, 0.66  0.08 (0.09)  0.05  −0.10, 0.26   CG chronic conditions  0.18 (0.06)**  0.16  0.06, 0.31  0.02 (0.02)  0.05  −0.03, 0.06  Primary objective stressors   CR chronic conditions  0.09 (0.05)  0.08  −0.02, 0.19  0.04 (0.02)*  0.09  0.001, 0.08   CR dementia status  0.74 (0.26)**  0.15  0.23, 1.26  0.10 (0.10)  0.06  −0.10, 0.30   CG ADL assistance  0.10 (0.06)  0.11  −0.02, 0.23  0.05 (0.02)*  0.14  0.01, 0.09   CG IADL assistance  −0.04 (0.08)  −0.03  −0.19, 0.12  −0.03 (0.03)  −0.06  −0.08, 0.03   CG health system tasks  0.19 (0.05)***  0.19  0.09, 0.29  0.07 (0.02)***  0.19  0.03, 0.11   CG medical/nursing tasks  0.002 (0.04)  0.002  −0.08, 0.09  0.05 (0.02)**  0.16  0.01, 0.09    Total R2  .22  .26  Predictors  Emotional difficulties  Role overload  B (SE)  β  95% CI  B (SE)  β  95% CI  Background characteristics   CG gender (female)  0.52 (0.19)**  0.16  0.14, 0.91  0.29 (0.07)***  0.24  0.16, 0.43   CG age  −0.01 (0.01)  −0.07  −0.04, 0.02  −0.002 (0.01)  −0.03  −0.01, 0.01   CG educational attainment  0.14 (0.05)**  0.17  0.04, 0.23  0.03 (0.02)  0.09  −0.01, 0.06   CG employment status  0.16 (0.25)  0.04  −0.35, 0.66  0.08 (0.09)  0.05  −0.10, 0.26   CG chronic conditions  0.18 (0.06)**  0.16  0.06, 0.31  0.02 (0.02)  0.05  −0.03, 0.06  Primary objective stressors   CR chronic conditions  0.09 (0.05)  0.08  −0.02, 0.19  0.04 (0.02)*  0.09  0.001, 0.08   CR dementia status  0.74 (0.26)**  0.15  0.23, 1.26  0.10 (0.10)  0.06  −0.10, 0.30   CG ADL assistance  0.10 (0.06)  0.11  −0.02, 0.23  0.05 (0.02)*  0.14  0.01, 0.09   CG IADL assistance  −0.04 (0.08)  −0.03  −0.19, 0.12  −0.03 (0.03)  −0.06  −0.08, 0.03   CG health system tasks  0.19 (0.05)***  0.19  0.09, 0.29  0.07 (0.02)***  0.19  0.03, 0.11   CG medical/nursing tasks  0.002 (0.04)  0.002  −0.08, 0.09  0.05 (0.02)**  0.16  0.01, 0.09    Total R2  .22  .26  Note. ADL = activities of daily living; CG = caregiver; CI = confidence interval; CR = care recipient; IADL = independent activities of daily living; SE = standard error. Estimates are from Step 1 of each model. Step 2 (not shown in table) included gender interactions with each primary objective stressor. There were no significant gender interactions in Step 2. N = 367 spousal caregivers. *p ≤ .05. **p ≤ .01. ***p ≤ .001. View Large Predictors of Secondary Role Strains Care-Related Valued Activity Restriction Table 2 shows that caregivers’ female gender (Model 1) and higher educational attainment (Models 1 and 2) were linked to a greater likelihood of care-related valued activity restriction. In Models 2 and 3, caregivers who provided more assistance with ADLs and health system interaction tasks were also more likely to report valued activity restriction. Beyond the aforementioned predictors, Model 3 reveals that older caregivers and those with higher role overload reported a greater likelihood of valued activity restriction. There were no significant gender differences in these associations (not shown in table). As presented in Table 5, assistance with ADLs was positively linked to role overload. Similarly, health system interaction tasks were positively linked to role overload and emotional caregiving difficulties. Accordingly, role overload was associated with indirect paths for both care tasks (Table 5) but did not reduce their associations with the outcome to nonsignificance after accounting for the link between role overload and the outcome (Model 3 in Table 2). Thus, role overload partially mediated the links between these tasks and valued activity restriction. Negative Caregiving Relationship Quality Table 3 shows that caregivers’ female gender and higher education were linked to greater negative caregiving relationship quality in Models 1 and 2. Caring for a partner with dementia was also associated with higher negative relationship quality in Model 2. This association was reduced to nonsignificance with the addition of primary subjective stressors in Model 3, demonstrating that caregivers with greater emotional caregiving difficulties and role overload reported higher negative caregiving relationship quality. Table 3. Predictors of Negative Caregiving Relationship Quality for Spousal Caregivers Predictors  Model 1  Model 2  Model 3  B (SE)  β  95% CI  B (SE)  β  95% CI  B (SE)  β  95% CI  Background characteristics   CG gender (female)  0.25 (0.10)*  0.16  0.05, 0.45  0.27 (0.10)**  0.17  0.07, 0.47  0.11 (0.10)  0.07  −0.08, 0.31   CG age  −0.01 (0.01)  −0.08  −0.02, 0.01  −0.01 (0.01)  −0.08  −0.02, 0.01  −0.01 (0.01)  −0.06  −0.02, 0.01   CG educational attainment  0.05 (0.02)*  0.11  0.01, 0.09  0.05 (0.02)*  0.12  0.01, 0.09  0.02 (0.02)  0.05  −0.02, 0.06   CG employment status  0.05 (0.19)  0.02  −0.33, 0.43  0.08 (0.18)  0.04  −0.28, 0.43  0.03 (0.15)  0.02  −0.27, 0.33   CG chronic conditions  0.06 (0.03)  0.10  −0.01, 0.13  0.05 (0.03)  0.09  −0.02, 0.12  0.02 (0.03)  0.03  −0.04, 0.07  Primary objective stressors   CR chronic conditionsa  —  —  —  0.05 (0.03)  0.09  −0.02, 0.11  0.03 (0.03)  0.05  −0.03, 0.08   CR dementia status  —  —  —  0.30 (0.13)*  0.13  0.04, 0.57  0.16 (0.12)  0.07  −0.08, 0.39   CG ADL assistance  —  —  —  −0.002 (0.03)  −0.004  −0.06, 0.06  −0.03 (0.03)  −0.07  −0.09, 0.03   CG IADL assistance  —  —  —  −0.06 (0.05)  −0.08  −0.15, 0.04  −0.04 (0.04)  −0.07  −0.12, 0.04   CG health system tasks  —  —  —  0.03 (0.04)  0.05  −0.05, 0.10  −0.02 (0.04)  −0.05  −0.10, 0.05   CG medical/nursing tasks  —  —  —  −0.001 (0.03)  −0.003  −0.06, 0.05  −0.01 (0.03)  −0.03  −0.06, 0.04  Primary subjective stressors   CG emotional difficulties  —  —  —  —  —  —  0.16 (0.03)***  0.33  0.10, 0.23   CG role overload  —  —  —  —  —  —  0.25 (0.10)**  0.19  0.05, 0.44    Total R2  .06  .09  .24  Predictors  Model 1  Model 2  Model 3  B (SE)  β  95% CI  B (SE)  β  95% CI  B (SE)  β  95% CI  Background characteristics   CG gender (female)  0.25 (0.10)*  0.16  0.05, 0.45  0.27 (0.10)**  0.17  0.07, 0.47  0.11 (0.10)  0.07  −0.08, 0.31   CG age  −0.01 (0.01)  −0.08  −0.02, 0.01  −0.01 (0.01)  −0.08  −0.02, 0.01  −0.01 (0.01)  −0.06  −0.02, 0.01   CG educational attainment  0.05 (0.02)*  0.11  0.01, 0.09  0.05 (0.02)*  0.12  0.01, 0.09  0.02 (0.02)  0.05  −0.02, 0.06   CG employment status  0.05 (0.19)  0.02  −0.33, 0.43  0.08 (0.18)  0.04  −0.28, 0.43  0.03 (0.15)  0.02  −0.27, 0.33   CG chronic conditions  0.06 (0.03)  0.10  −0.01, 0.13  0.05 (0.03)  0.09  −0.02, 0.12  0.02 (0.03)  0.03  −0.04, 0.07  Primary objective stressors   CR chronic conditionsa  —  —  —  0.05 (0.03)  0.09  −0.02, 0.11  0.03 (0.03)  0.05  −0.03, 0.08   CR dementia status  —  —  —  0.30 (0.13)*  0.13  0.04, 0.57  0.16 (0.12)  0.07  −0.08, 0.39   CG ADL assistance  —  —  —  −0.002 (0.03)  −0.004  −0.06, 0.06  −0.03 (0.03)  −0.07  −0.09, 0.03   CG IADL assistance  —  —  —  −0.06 (0.05)  −0.08  −0.15, 0.04  −0.04 (0.04)  −0.07  −0.12, 0.04   CG health system tasks  —  —  —  0.03 (0.04)  0.05  −0.05, 0.10  −0.02 (0.04)  −0.05  −0.10, 0.05   CG medical/nursing tasks  —  —  —  −0.001 (0.03)  −0.003  −0.06, 0.05  −0.01 (0.03)  −0.03  −0.06, 0.04  Primary subjective stressors   CG emotional difficulties  —  —  —  —  —  —  0.16 (0.03)***  0.33  0.10, 0.23   CG role overload  —  —  —  —  —  —  0.25 (0.10)**  0.19  0.05, 0.44    Total R2  .06  .09  .24  Note. ADL = activities of daily living; CG = caregiver; CI = confidence interval; CR = care recipient; IADL = independent activities of daily living; SE = standard error. Estimates are from Step 1 of each model. Step 2 (not shown in table) included gender interactions with background characteristics in Model 1, gender interactions with primary objective stressors in Model 2, and gender interactions with primary subjective stressors in Model 3. N = 367 spousal caregivers. aDenotes significant gender interaction found in Step 2. *p ≤ .05. **p ≤ .01. ***p ≤ .001. View Large Table 3. Predictors of Negative Caregiving Relationship Quality for Spousal Caregivers Predictors  Model 1  Model 2  Model 3  B (SE)  β  95% CI  B (SE)  β  95% CI  B (SE)  β  95% CI  Background characteristics   CG gender (female)  0.25 (0.10)*  0.16  0.05, 0.45  0.27 (0.10)**  0.17  0.07, 0.47  0.11 (0.10)  0.07  −0.08, 0.31   CG age  −0.01 (0.01)  −0.08  −0.02, 0.01  −0.01 (0.01)  −0.08  −0.02, 0.01  −0.01 (0.01)  −0.06  −0.02, 0.01   CG educational attainment  0.05 (0.02)*  0.11  0.01, 0.09  0.05 (0.02)*  0.12  0.01, 0.09  0.02 (0.02)  0.05  −0.02, 0.06   CG employment status  0.05 (0.19)  0.02  −0.33, 0.43  0.08 (0.18)  0.04  −0.28, 0.43  0.03 (0.15)  0.02  −0.27, 0.33   CG chronic conditions  0.06 (0.03)  0.10  −0.01, 0.13  0.05 (0.03)  0.09  −0.02, 0.12  0.02 (0.03)  0.03  −0.04, 0.07  Primary objective stressors   CR chronic conditionsa  —  —  —  0.05 (0.03)  0.09  −0.02, 0.11  0.03 (0.03)  0.05  −0.03, 0.08   CR dementia status  —  —  —  0.30 (0.13)*  0.13  0.04, 0.57  0.16 (0.12)  0.07  −0.08, 0.39   CG ADL assistance  —  —  —  −0.002 (0.03)  −0.004  −0.06, 0.06  −0.03 (0.03)  −0.07  −0.09, 0.03   CG IADL assistance  —  —  —  −0.06 (0.05)  −0.08  −0.15, 0.04  −0.04 (0.04)  −0.07  −0.12, 0.04   CG health system tasks  —  —  —  0.03 (0.04)  0.05  −0.05, 0.10  −0.02 (0.04)  −0.05  −0.10, 0.05   CG medical/nursing tasks  —  —  —  −0.001 (0.03)  −0.003  −0.06, 0.05  −0.01 (0.03)  −0.03  −0.06, 0.04  Primary subjective stressors   CG emotional difficulties  —  —  —  —  —  —  0.16 (0.03)***  0.33  0.10, 0.23   CG role overload  —  —  —  —  —  —  0.25 (0.10)**  0.19  0.05, 0.44    Total R2  .06  .09  .24  Predictors  Model 1  Model 2  Model 3  B (SE)  β  95% CI  B (SE)  β  95% CI  B (SE)  β  95% CI  Background characteristics   CG gender (female)  0.25 (0.10)*  0.16  0.05, 0.45  0.27 (0.10)**  0.17  0.07, 0.47  0.11 (0.10)  0.07  −0.08, 0.31   CG age  −0.01 (0.01)  −0.08  −0.02, 0.01  −0.01 (0.01)  −0.08  −0.02, 0.01  −0.01 (0.01)  −0.06  −0.02, 0.01   CG educational attainment  0.05 (0.02)*  0.11  0.01, 0.09  0.05 (0.02)*  0.12  0.01, 0.09  0.02 (0.02)  0.05  −0.02, 0.06   CG employment status  0.05 (0.19)  0.02  −0.33, 0.43  0.08 (0.18)  0.04  −0.28, 0.43  0.03 (0.15)  0.02  −0.27, 0.33   CG chronic conditions  0.06 (0.03)  0.10  −0.01, 0.13  0.05 (0.03)  0.09  −0.02, 0.12  0.02 (0.03)  0.03  −0.04, 0.07  Primary objective stressors   CR chronic conditionsa  —  —  —  0.05 (0.03)  0.09  −0.02, 0.11  0.03 (0.03)  0.05  −0.03, 0.08   CR dementia status  —  —  —  0.30 (0.13)*  0.13  0.04, 0.57  0.16 (0.12)  0.07  −0.08, 0.39   CG ADL assistance  —  —  —  −0.002 (0.03)  −0.004  −0.06, 0.06  −0.03 (0.03)  −0.07  −0.09, 0.03   CG IADL assistance  —  —  —  −0.06 (0.05)  −0.08  −0.15, 0.04  −0.04 (0.04)  −0.07  −0.12, 0.04   CG health system tasks  —  —  —  0.03 (0.04)  0.05  −0.05, 0.10  −0.02 (0.04)  −0.05  −0.10, 0.05   CG medical/nursing tasks  —  —  —  −0.001 (0.03)  −0.003  −0.06, 0.05  −0.01 (0.03)  −0.03  −0.06, 0.04  Primary subjective stressors   CG emotional difficulties  —  —  —  —  —  —  0.16 (0.03)***  0.33  0.10, 0.23   CG role overload  —  —  —  —  —  —  0.25 (0.10)**  0.19  0.05, 0.44    Total R2  .06  .09  .24  Note. ADL = activities of daily living; CG = caregiver; CI = confidence interval; CR = care recipient; IADL = independent activities of daily living; SE = standard error. Estimates are from Step 1 of each model. Step 2 (not shown in table) included gender interactions with background characteristics in Model 1, gender interactions with primary objective stressors in Model 2, and gender interactions with primary subjective stressors in Model 3. N = 367 spousal caregivers. aDenotes significant gender interaction found in Step 2. *p ≤ .05. **p ≤ .01. ***p ≤ .001. View Large Table 5 reveals that caring for a partner with dementia was associated with greater emotional caregiving difficulties. Emotional difficulties were therefore linked to the indirect path between dementia status and negative caregiving relationship quality (Table 5), and reduced this association to nonsignificance when accounting for the link between emotional difficulties and the outcome (Model 3 in Table 3). This provides evidence that emotional difficulties fully mediated the association between dementia status and negative caregiving relationship quality. A significant interaction emerged between care recipients’ chronic health conditions and caregiver gender in Model 2 (B = −0.17, β = −0.24, p = .02, 95% CI [−0.31, −0.03]). Husbands reported more negative caregiving relationship quality when their partner had a higher number of chronic conditions (B = 0.15, β = 0.27, p = .002, 95% CI [0.06, 0.25]); but this link was not found for wives (B = −0.02, β = −0.03, p = .70, 95% CI [−0.11, 0.07]) (not shown in table). Although the care recipient’s number of chronic conditions was positively associated with role overload (Table 5), caregiver gender did not moderate this association or the link between role overload and negative caregiving relationship quality (not shown in tables). As a consequence, this interaction was not mediated by gender differences pertaining to subjective stressors. Care-Related Family Disagreements As depicted in Table 4, caregivers were more likely to report care-related family disagreements when they perceived higher role overload (Model 3). None of the background characteristics or primary objective stressors were directly associated with family disagreements. There was a significant interaction between care recipients’ dementia status and caregiver gender in Model 2 (B = −3.22, p = .002). Husbands were more likely to report care-related family disagreements when they cared for a partner with dementia (B = 2.77, OR = 15.98, p < .001, 95% CI [4.02, 63.49]); however, this link was not present for wives (B = −0.45, OR = 0.64, p = .49, 95% CI [0.18, 2.28]) (not shown in table). Dementia status was positively linked to emotional caregiving difficulties (Table 5); but caregiver gender did not moderate this association or the link between emotional difficulties and care-related family disagreements (not shown in tables). As such, this interaction was not mediated by gender differences relating to subjective stressors. Post Hoc Tests To test the stability of the findings regarding gender differences, we estimated a full model for each outcome including all predictors and their gender interaction terms. The findings on gender differences from the main analysis remained significant, confirming their stability. Discussion and Implications The present study demonstrates that a number of factors may contribute to secondary role strains among spouses caring for a partner with functional disability. Wives generally reported more primary stressors and secondary role strains than husbands on average. These findings are consistent with prior work showing that, relative to men, women typically face more intense and stressful care situations (Pinquart & Sörenson, 2006; Swinkels et al., 2017). Nonetheless, this study reveals few gender differences in caregiving stress proliferation, which implies that a given level of primary care-related stressors impacts wives and husbands in similar ways. Overall, time-intensive care tasks and role overload seem to be most consequential for secondary role strains among wives and husbands. But husbands appear uniquely susceptible to marital and family discord when their partner has more complex care needs. Secondary strains play a key part in explaining gender differences in spouses’ caregiver burden (Swinkels et al., 2017). Consequently, wives and husbands may be roughly equal in their experience of burden and related adverse health effects when comparable levels of caregiving stress spread to other life areas. The findings therefore yield critical insights for clinicians and policymakers concerned with spousal caregivers’ well-being and their ability to provide quality care. Predictors of Care-Related Valued Activity Restriction In line with research showing that women experience more family barriers to self-care than men (Rosland et al., 2010; Thomeer et al., 2015), wives were more likely than husbands to report care-related valued activity restriction, independent of their background characteristics (e.g., age, educational attainment). But this association was not observed after accounting for primary objective stressors, which suggests that it may be partially explained by wives’ greater exposure to these stressors on average (Pinquart & Sörensen, 2006; Swinkels et al., 2017). Contrary to prediction, caregivers with more education were more likely to report care-related valued activity restriction. This finding may be attributed in part to the type of activities assessed in this study, some of which require financial resources that may limit participation among caregivers with less education (e.g., going out for enjoyment). Of note, however, this link was no longer present after taking primary subjective stressors into account, indicating that education is relatively inconsequential when considering how caregivers appraise their care role. Caregivers were also more likely to report care-related valued activity restriction when they provided more assistance with ADLs (e.g., mobility) and health system interaction tasks (e.g., scheduling appointments). Helping with ADLs requires a substantial time commitment and has been associated with decreased self-care behaviors among caregiving spouses (Gallant & Connell, 1997), which could extend to reduced participation in activities they enjoy. Interacting with health systems similarly involves complicated decision-making (e.g., selecting a drug insurance plan) and coordinating care across multiple providers that can be challenging and time-consuming (Giovannetti et al., 2012). Supporting these points, caregivers’ greater perceptions of role overload partially accounted for the associations between both care tasks and care-related valued activity restriction. Over and above primary objective stressors, older caregivers and those who perceived higher role overload were more likely to report valued activity restriction. Caregivers who feel overloaded may forgo their valued activities to manage other noncare tasks (e.g., housework), and could benefit from respite services along with guidance in how to optimize time away from caregiving to incorporate activities that restore their well-being (Lund, Utz, Caserta, & Wright, 2009). The finding for age was present only after accounting for primary subjective stressors, implying that caregivers’ age-related vulnerabilities may increase their likelihood of valued activity restriction when considering their negative appraisals of the care role. Research has shown that older adults are impaired in their capacity to cope with negative experiences in circumstances they cannot easily avoid such as spousal caregiving (Charles, 2010), which may in turn deplete their energy to engage in valued activities. Predictors of Negative Caregiving Relationship Quality Wives reported more negative caregiving relationship quality than husbands, independent of their background characteristics and primary objective stressors. This finding is consistent with the broader literature demonstrating more negative views of marital quality among wives than husbands in middle and later life (e.g., Umberson & Williams, 2005). Contrary to prediction, caregivers with higher levels of education also perceived more negative caregiving relationship quality. It is plausible that the challenges of spousal caregiving lead to more marital tension for highly educated individuals because they have had less experience in handling adversity than those with lower educational attainment (Seery, 2011). Beyond gender and education, caring for a partner with dementia was linked to greater perceptions of negative caregiving relationship quality. This association was fully mediated by emotional care-related difficulties, indicating that such difficulties may escalate negative aspects of the caregiving relationship in the context of dementia. Diminished marital communication and intimacy as dementia progresses could further play a part in explaining these findings (Quinn et al., 2009). Greater emotional caregiving difficulties and role overload emerged as the sole correlates of higher negative caregiving relationship quality when all predictors were considered. This implies that negative appraisals of the care role may have a particularly detrimental impact on spousal interactions. Alternatively, these links could be bidirectional. Caregivers who are highly distressed, for example, tend to be more critical or hostile toward the care recipient (Smith et al., 2011). In turn, poor marital quality may amplify caregivers’ distress (Baucom et al., 2012; Berg & Upchurch, 2007). Caring for a partner with a higher number of chronic health conditions was associated with more negative caregiving relationship quality for husbands only, even when controlling for primary subjective stressors. Compared with women, men rely more heavily on their partner to satisfy emotional support needs (Antonucci, 2001). Thus, one possibility is that greater medical morbidity renders wives less able to meet these needs, thereby contributing to marital discord. Relatedly, men are often unable or unwilling to provide emotional support to their wife regarding her own health issues (Thomeer et al., 2015), which may intensify marital problems. Predictors of Care-Related Family Disagreements Caregivers were more prone to report care-related family disagreements when they felt higher role overload. In accord with family systems theory (Bowen, 1978), this finding may reflect reciprocal exchanges such that caregivers’ feelings of overload exacerbate disagreements, while disagreements magnify perceptions of being overloaded. The absence of a significant gender difference suggests that caregiving wives and husbands who experience role overload are both vulnerable to potentially harmful family dynamics. For husbands only, caring for a partner with dementia was associated with a higher likelihood of care-related family disagreements. Men are traditionally viewed as less competent in the care role than women (Calasanti, 2010; Thompson, 2002), and so family strife may partly stem from doubts about husbands’ ability to provide care in more complicated caregiving situations. Alternatively, in complex circumstances, husbands may be more likely to serve as a caregiver when there is family discord because other potential helpers are unavailable or unwilling to contribute. Strengths and Limitations Strengths of this study include a nationally representative sample, emphasis on caregiving stress proliferation among spousal caregivers, and the statistical comparison of differences between caregiving wives and husbands. Models accounted for caregivers’ sociodemographics and health conditions, along with multiple primary objective and subjective caregiving stressors. Hence, the present study identifies an array of factors that are independently linked to secondary role strains and could eventually erode each care partner’s well-being. We acknowledge several limitations. First, cross-sectional analyses preclude the determination of causal associations. Second, on average, spouses reported low levels of primary care-related stressors and secondary role strains. Although these levels are in line with other population-based studies demonstrating considerable resilience among caregivers as a whole (Roth et al., 2015), the findings may not generalize to more distressed spousal caregivers. Third, we were limited by the base study measures. The NSOC measure of activity restriction, for instance, may not have captured all valued activities that are impacted by caregiving. Additionally, care-related family disagreements were assessed with one item. We also used binary measures of care-related valued activity restriction and family disagreements due to a lack of variability in both outcomes, which limits understanding of the degree to which these secondary role strains are associated with characteristics of the spousal care situation. Finally, NSOC does not include a pre-caregiving measure of relationship quality, limiting the ability to disentangle whether the findings mirror continued marital patterns. The finding that wives report more negative caregiving relationship quality than husbands in the multivariate analyses, for example, may be partly attributed to fewer benefits of marriage derived by women (Antonucci, 2001; Umberson & Williams, 2005). Fourth, the findings may not translate to specific care contexts (e.g., dementia caregiving). Nevertheless, this study provides the foundation for subsequent work to elucidate knowledge of caregiving stress proliferation among aging adults caring for partners with functional disability. Future Directions The present study brings forth a number of practical considerations that should be explored in future research. First, this study underscores the importance of examining how family relationships both influence and are influenced by the care situation. Prior work has placed a near-exclusive focus on primary caregivers, neglecting an understanding of the family system (Schulz & Eden, 2016). Yet, caregiving emerges from a past and present family narrative that shapes caregivers’ experience of stressors as well as their available resources to combat stress (Zarit et al., in press). Determining strategies for caregiving spouses to leverage their positive social ties (e.g., finding other helpers, sharing emotional support) while minimizing the effects of negative relationships (e.g., marital or family conflict) could be crucial intervention targets. Family meetings to guide the negotiation of care responsibilities and the resolution of disagreements may also attenuate caregiving stress proliferation (Joling et al., 2012). Second, counter to the common view that men’s managerial approach to providing care shields them from negative care-related outcomes (Calasanti, 2010; Thompson, 2002), this study indicates that husbands may be highly susceptible to caregiving stress proliferation under certain conditions. In line with previous studies (e.g., DePasquale et al., 2016; Swinkels et al., 2017), these findings indicate that clinicians and policymakers should recognize the variability in men’s caregiving experiences. Increased attention to the needs and concerns of caregiving husbands in more complex care situations may be especially warranted. Lastly, this study illuminates the need to consider long-term consequences of spouses’ caregiving stress proliferation. Previous research has demonstrated, for example, that reduced scheduled activities and negative marital interactions are each linked to spousal caregivers’ lower positive affect and higher negative affect across time (Savla et al., 2011), raising the question of whether these secondary strains have lasting detrimental implications for well-being. Balancing competing roles and maintaining self-care activities have been pinpointed as unmet support needs among spousal caregivers (Badr et al., 2016), and so the proliferation of caregiving stress could present a risk of detrimental outcomes for both care partners. Conclusions In summary, this study highlights the multifaceted ways that background characteristics and primary care-related stressors are linked to secondary role strains among spouses caring for a partner with functional disability. Notably, the proliferation of caregiving stress may develop through both common and distinct pathways for wives and husbands, identifying those who could require support in managing their care role along with noncare roles and activities. The findings call for a more comprehensive approach to caregiver interventions and policies that consider how caregiving may influence and be influenced by other important life domains. Funding This work was supported by the National Institute of Mental Health (T32 MH073553-11 to S. J. Bartels) and the National Institute on Aging (F31 AG050385 to N. DePasquale). The National Health and Aging Trends Study (NHATS) and National Study of Caregiving (NSOC) are sponsored by the National Institute on Aging (NIA U01AG032947) through a cooperative agreement with the Johns Hopkins Bloomberg School of Public Health. Conflict of Interest None reported. References Aneshensel, C.S., Pearlin, L.I., Mullan, J.T., Zarit, S.H., & Whitlatch, C.J. ( 1995). Profiles in caregiving. The unexpected career . San Diego, CA: Academic Press. Antonucci, T.C. ( 2001). Social relations: An examination of social networks, social support and sense of control. In J. E. Birren, K. W. Schaie, & R. P. Abeles (Eds.), Handbook of the psychology of aging (pp. 427–453) . San Diego, CA: Academic Press. Badr, H., Herbert, K., Reckson, B., Rainey, H., Sallam, A., & Gupta, V. ( 2016). Unmet needs and relationship challenges of head and neck cancer patients and their spouses. Journal of Psychosocial Oncology , 34, 336– 346. doi: 10.1080/07347332.2016.1195901 Google Scholar CrossRef Search ADS PubMed  Baron, R.M., & Kenny, D.A. ( 1986). The moderator-mediator variable distinction in social psychological research: Conceptual, strategic, and statistical considerations. Journal of Personality and Social Psychology , 51, 1173– 1182. doi: 10.1037/0022-3514.51.6.1173 Google Scholar CrossRef Search ADS PubMed  Baucom, D.H., Porter, L.S., Kirby, J.S., & Hudepohl, J. ( 2012). Couple-based interventions for medical problems. Behavior Therapy , 43, 61– 76. doi: 10.1016/j.beth.2011.01.008. Google Scholar CrossRef Search ADS PubMed  Berg, C.A., & Upchurch, R. ( 2007). A developmental-contextual model of couples coping with chronic illness across the adult life span. Psychological Bulletin , 133, 920– 954. doi: 10.1037/0033-2909.133.6.920 Google Scholar CrossRef Search ADS PubMed  Bowen, M. ( 1978). Family therapy in clinical practice . New York: Jason Aronson. Calasanti, T. ( 2010). Gender relations and applied research on aging. The Gerontologist , 50, 720– 734. doi: 10.1093/geront/gnq085 Google Scholar CrossRef Search ADS PubMed  Chappell, N.L., Dujela, C., & Smith, A. ( 2015). Caregiver well-being: Intersections of relationship and gender. Research on Aging , 37, 623– 645. doi: 10.1177/0164027514549258 Google Scholar CrossRef Search ADS PubMed  Charles, S.T. ( 2010). Strength and vulnerability integration (SAVI): A model of emotional well-being across adulthood. Psychological Bulletin , 136, 1068– 1091. doi: 10.1037/a0021232 Google Scholar CrossRef Search ADS PubMed  Chattillion, E.A., Ceglowski, J., Roepke, S.K., von Känel, R., Losada, A., Mills, P.J., … Mausbach, B.T. ( 2013). Pleasant events, activity restriction, and blood pressure in dementia caregivers. Health Psychology , 32, 793– 801. doi: 10.1037/a0029412 Google Scholar CrossRef Search ADS PubMed  DePasquale, N., Polenick, C.A., Hinde, J., Bray, J., Zarit, S.H., Moen, P., … Almeida, D.M. ( 2016). Health behavior among men with multiple family roles: The moderating effects of perceived partner relationship quality. American Journal of Men’s Health . Advance online publication. doi: 10.1177/1557988316660088 Eisinga, R., Grotenhuis, M., & Pelzer, B. ( 2013). The reliability of a two-item scale: Pearson, Cronbach, or Spearman-Brown? International Journal of Public Health , 58, 637– 642. doi: 10.1007/s00038-012-0416-3 Google Scholar CrossRef Search ADS PubMed  Gallant, M. P., & Connell, C. M. ( 1997). Predictors of decreased self-care among spouse caregivers of older adults with dementing illnesses. Journal of Aging and Health , 9, 373– 395. doi: 10.1177/089826439700900306 Google Scholar CrossRef Search ADS PubMed  Gaugler, J.E., Linder, J., Given, C.W., Kataria, R., Tucker, G., & Regine, W.F. ( 2008). The proliferation of primary cancer caregiving stress to secondary stress. Cancer Nursing , 31, 116– 123. doi: 10.1097/01.NCC.0000305700.05250.9d. Google Scholar CrossRef Search ADS PubMed  Gaugler, J.E., Mendiondo, M., Smith, C.D., & Schmitt, F.A. ( 2003). Secondary dementia caregiving and its consequences. American Journal of Alzheimer’s Disease and Other Dementia , 18, 300– 308. doi: 10.1177/153331750301800505 Google Scholar CrossRef Search ADS   Giovannetti, E.R., Wolff, J.L., Xue, Q., Weiss, C.O., Leff, B., Boult, C., … Boyd, C. ( 2012). Difficulty assisting with health care tasks among caregivers of multimorbid older adults. Journal of General Internal Medicine , 27, 37– 44. doi: 10.1007/s11606-011-1831-5. Google Scholar CrossRef Search ADS PubMed  Gwyther, L.P., & Matchar, B.G. ( 2015). The Duke employee elder care consultation: Meeting employees where they are. Generations , 39, 105– 108. Holt-Lunstad, J., Smith, T.B., Baker, M., Harris, T., & Stephenson, D. ( 2015). Loneliness and social isolation as risk factors for mortality: A meta-analytic review. Perspectives on Psychological Science , 10, 227– 37. doi: 10.1177/1745691614568352 Google Scholar CrossRef Search ADS PubMed  Hsu, T., Loscalzo, M., Ramani, R., Forman, S., Popplewell, L., Clark, K., … Smith, D. ( 2014). Factors associated with high burden in caregivers of older adults with cancer. Cancer , 120, 2927– 2935. doi: 10.1002/cncr.28765 Google Scholar CrossRef Search ADS PubMed  Joling, K.J., van Marwijk, H.W.J., Smit, F., van der Horst, H.E., Scheltens, P., van de Ven, P. M., … & van Hout, H.P.J. ( 2012). Does a family meeting intervention prevent depression and anxiety in family caregivers of dementia patients? A randomized trial. PLoS ONE , 7, e30936. doi: 10.1371/journal.pone.0030936 Google Scholar CrossRef Search ADS PubMed  Kasper, J.D., Freedman, V.A., & Spillman, B. ( 2013). National Study of Caregiving user guide . Baltimore, MD: Johns Hopkins University School of Public Health. Lund, D.A., Utz, R.. Caserta, M.S., & Wright, S.D. ( 2009). Examining what caregivers do during respite time to make respite more effective. Journal of Applied Gerontology , 28, 109– 131. doi: 10.1177/0733464808323448 Google Scholar CrossRef Search ADS   Lyons, K.S., Zarit, S.H., Sayer, A.G., & Whitlatch, C.J. ( 2002). Caregiving as a dyadic process: Perspectives from caregiver and care receiver. Journal of Gerontology, Series B: Psychological Sciences and Social Sciences , 57, 195– 204. doi: 10.1093/geronb/57.3.P195 Google Scholar CrossRef Search ADS   Mausbach, B.T., Chattillion, E.A., Moore, R.C., Roepke, S.K., Depp, C.A., & Roesch, S. ( 2011). Activity restriction and depression in medical patients and their caregivers: A meta-analysis. Clinical Psychology Review , 31, 900– 908. doi: 10.1016/j.cpr.2011.04.004 Google Scholar CrossRef Search ADS PubMed  Moore, R.C., Harmell, A.L., Chattillion, E., Ancoli-Isreal, S., Grant, I., & Mausbach, B.T. ( 2011). PEAR model and sleep outcomes in dementia caregivers: Influence of activity restriction and pleasant events on sleep disturbances. International Psychogeriatrics , 23, 1462– 1469. doi: 10.1017/S1041610211000512 Google Scholar CrossRef Search ADS PubMed  Muller, D., Judd, C.M., & Yzerbyt, V.Y. ( 2005). When moderation is mediated and mediation is moderated. Journal of Personal and Social Psychology , 89, 852– 863. doi: 10.1037/0022-3514.89.6.852 Google Scholar CrossRef Search ADS   Navaie-Waliser, M., Feldman P.H., Gould, D.A., Levine, C., Kuerbis, A.N. & Donelan, K. ( 2002). When the caregiver needs care: The plight of vulnerable caregivers. Journal of Public Health , 92, 409– 413. doi: 10.2105/AJPH.92.3.409 Oldenkamp, M., Hagedoorn, M., Slaets, J., Stolk, R., Wittek, R., & Smidt, N. ( 2016). Subjective burden among spousal and adult-child informal caregivers of older adults: Results from a longitudinal cohort study. BMC Geriatrics , 16, 208. doi: 10.1186/s12877-016-0387-y Google Scholar CrossRef Search ADS PubMed  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, 583– 594. doi: 10.1093/geront/30.5.583 Google Scholar CrossRef Search ADS PubMed  Piette, J.D., Rosland, A.M., Silveira, M., Kabeto, M., & Langa, K.M. ( 2010). The case for involving adult children outside the household in the self-management support of older adults with chronic illnesses. Chronic Illness , 6, 34– 45. doi: 10.1177/1742395309347804. Google Scholar CrossRef Search ADS PubMed  Pinquart, M., & Sörenson, S. ( 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, 33– 45. doi: 10.1093/geronb/61.1.P33 Google Scholar CrossRef Search ADS   Pinquart, M., & Sörenson, S. ( 2007). Correlates of physical health of informal caregivers: A meta-analysis. The Journals of Gerontology, Series B: Psychological Sciences and Social Sciences , 62, P126– P137. doi: 10.1093/geronb/62.2.P126 Google Scholar CrossRef Search ADS   Pinquart, M., & Sörenson, S. ( 2011). Spouses, adult children, and children in-law as caregivers of older adults: A meta-analytic comparison. Psychology and Aging , 26, 1– 14. doi: 10.1037/a0021863 Google Scholar CrossRef Search ADS PubMed  Quinn, C., Claire, L., & Woods, B. ( 2009). The impact of the quality of relationship on the experiences and wellbeing of caregivers of people with dementia: A systematic review. Aging and Mental Health , 13, 143– 154. doi: 10.1080/13607860802459799. Google Scholar CrossRef Search ADS PubMed  Reinhard, S.C., Levine, C., & Samis, S. ( 2014). Family caregivers providing complex chronic care to their spouses . Washington, DC: AARP and United Hospital Fund. Rosland, A., Heisler, M., Choi, H., Silveira, M.J., & Piette, J.D. ( 2010). Family influences on self-management among functionally independent adults with diabetes or heart failure: Do family members hinder as much as they help? Chronic Illness , 6, 22– 33. doi: 10.1177/1742395309354608 Google Scholar CrossRef Search ADS PubMed  Roth, D.L., Fredman, L., & Haley, W.E. ( 2015). Informal caregiving and its impact on health: A reappraisal from population-based studies. The Gerontologist , 55, 309– 319. doi: 10.1093/geront/gnu177 Google Scholar CrossRef Search ADS PubMed  Rudman, D.L., Hebert, D., & Reid, D. ( 2006). Living in a restricted occupational world: The occupational experiences of stroke survivors who are wheelchair users and their caregivers. Canadian Journal of Occupational Therapy , 73, 141– 152. doi: 10.2182/cjot.05.0014 Google Scholar CrossRef Search ADS   Savla, J., Roberto, K.A., Blieszner, R., Cox, M., & Gwazdauskas, F. ( 2011). Effects of daily stressors on the psychological and biological well-being of spouses of persons with mild cognitive impairment. The Journal of Gerontology, Series B: Psychological Sciences and Social Sciences , 66, 653– 664. doi: 10.1093/geronb/gbr041 Google Scholar CrossRef Search ADS   Scharlach, A., Li, W., & Dalvi, T.B. ( 2006). Family conflict as a mediator of caregiver strain. Family Relations , 55, 625– 635. doi: 10.1111/j.1741-3729.2006.00431.x Google Scholar CrossRef Search ADS   Schulz, R., & Eden, J. ( 2016). Families caring for an aging America . Washington, DC: The National Academies Press. Retrieved from http://www.nap.edu/23606 Google Scholar CrossRef Search ADS   Seery, M. D. ( 2011). Resilience: A silver lining to experiencing adverse life events? Current Directions in Psychological Science , 20, 390– 394. doi: 10.1177/0963721411424740 Google Scholar CrossRef Search ADS   Semple, S.J. ( 1992). Conflict in Alzheimer’s caregiving families: Its dimensions and consequences. The Gerontologist , 32, 648– 655. doi: 10.1093/geront/32.5.648 Google Scholar CrossRef Search ADS PubMed  Smith, G.R., Williamson, G.M., Miller, L.S., & Schulz, R. ( 2011). Depression and quality of informal care: A longitudinal investigation of caregiving stressors. Psychology and Aging , 26, 584– 591. doi: 10.1037/a0022263 Google Scholar CrossRef Search ADS PubMed  Swinkels, J., van Tilburg, T., Verbakel, E., & Broese van Groenou, M. ( 2017). Explaining the gender gap in the caregiving burden of partner caregivers. The Journals of Gerontology, Series B: Psychological Sciences and Social Sciences . Advance online publication. doi: 10.1093/geronb/gbx036 Thomeer, M.B., Reczek, C., & Umberson, D. ( 2015). Gendered emotion work around physical health problems in mid- and later-life marriages. Journal of Aging Studies , 32, 12– 22. doi: 10.1016/j.jaging.2014.12.001 Google Scholar CrossRef Search ADS PubMed  Thompson, E.H. ( 2002). What’s unique about men’s caregiving? Men as caregivers: Theory, research, and service implications. (pp. 20–47) . New York, NY: Springer Publishing. Umberson, D., & Williams, K. ( 2005). Marital quality, health, and aging: Gender equity? The Journals of Gerontology. Series B: Psychological Sciences and Social Sciences , 60 (SPEC), 109– 113. doi: 10.1093/geronb/60.Special_Issue_2.S109 Google Scholar CrossRef Search ADS   Valtorta, N.K., Kanaan, M., Gilbody, S., Ronzi, S., & Hanratty, B. ( 2016). Loneliness and social isolation as risk factors for coronary heart disease and stroke: Systematic review and meta-analysis of longitudinal observational studies. Heart , 102, 1009– 1016. doi: 10.1136/heartjnl-2015–308790 Google Scholar CrossRef Search ADS PubMed  Zarit, S.H., Polenick, C.A, DePasquale, N., Liu, Y., & Bangerter, L.R. ( in press). Family support and caregiving in middle and late life. In B. H. Fiese (Ed.), APA Handbook of Contemporary Family Psychology (Vol. 2) . Washington, DC: American Psychological Association. © The Author(s) 2018. Published by Oxford University Press on behalf of The Gerontological Society of America. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com. http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png The Gerontologist Oxford University Press

Predictors of Secondary Role Strains Among Spousal Caregivers of Older Adults With Functional Disability

Loading next page...
 
/lp/ou_press/predictors-of-secondary-role-strains-among-spousal-caregivers-of-older-0J3MFufd2Y
Publisher
Oxford University Press
Copyright
© The Author(s) 2018. Published by Oxford University Press on behalf of The Gerontological Society of America. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com.
ISSN
0016-9013
eISSN
1758-5341
D.O.I.
10.1093/geront/gnx204
Publisher site
See Article on Publisher Site

Abstract

Abstract Background and Objectives Aging spouses commonly care for a partner with functional disability, but little is known about how spousal caregiving may impact different life domains. This study evaluated how caregiving characteristics are associated with secondary role strains among spousal caregivers. Research Design and Methods This cross-sectional study examined 367 spousal caregivers and their partners from the 2011 National Health and Aging Trends Study and National Study of Caregiving. Hierarchical regressions were estimated to determine how caregiver background factors (sociodemographics, health conditions) along with primary objective (care activities, care recipient health conditions, and dementia status) and subjective (emotional caregiving difficulties, role overload) stressors are linked to care-related valued activity restriction, negative caregiving relationship quality, and care-related family disagreements. Gender differences were considered. Results After accounting for all predictors, older caregivers and caregivers providing more help with activities of daily living and health system interactions (e.g., scheduling appointments) were more likely to report activity restriction, whereas caregivers with more emotional difficulties reported higher negative caregiving relationship quality. Role overload was positively associated with all three secondary strains. For husbands only, caring for a partner with more chronic conditions was linked to higher negative caregiving relationship quality and caring for a partner with dementia was associated with a greater likelihood of family disagreements. Discussion and Implications Secondary role strains may develop through similar and unique pathways for caregiving wives and husbands. Further research is needed to identify those who could benefit from support in managing their care responsibilities alongside other life areas. Caregiving-informal, Caregiver stress, Disabilities, Marriage, Family issues, Gender Introduction Nearly 18 million U.S. adults provide family care for a person aged 65 or older due to limitations in physical, mental, or cognitive capacities (Schulz & Eden, 2016). When spouses or cohabiting partners are available, they often assist with a variety of activities including personal care, household chores, and medical tasks (e.g., managing medication; Reinhard, Levine, & Samis, 2014; Schulz & Eden, 2016). Although it may be perceived as a normative part of marriage, spousal caregiving can take a heavy toll on well-being. Relative to adult child caregivers, caregiving spouses report higher depressive symptoms and worse physical health (Pinquart & Sörenson, 2007, 2011). Spousal caregivers are also more likely to experience chronic conditions and declines in physical function that diminish their ability to provide care (Navaie-Waliser et al., 2002; Oldenkamp et al., 2016; Piette, Rosland, Silveira, Kabeto, & Langa, 2010). Yet, relatively little is known about which aspects of spousal caregiving may be most consequential for other important life areas, such as caregivers’ valued activity engagement and the quality of their marital and family relationships (Schulz & Eden, 2016). Understanding the broader implications of the spousal care role for aging wives and husbands would enhance efforts to sustain the long-term well-being of a particularly vulnerable caregiving subgroup. Guided by stress process models of caregiving, this study examined how spousal caregivers’ background characteristics and caregiving stressors are linked to care-related valued activity restriction, negative caregiving relationship quality, and care-related family disagreements. Stress Process Perspectives on Spousal Caregiving According to stress process models, caring for an older spouse with functional disability is a chronic source of stress that can lead to emotional and physical problems (Aneshensel et al., 1995; Pearlin, Mullan, Semple, & Skaff, 1990). As we describe below, however, the extent to which caregivers experience adverse outcomes depends in part on several key components of the care situation. Caregiver Background Characteristics Background characteristics reflect caregivers’ exposure to various stressors as well as their resources to manage stress (Aneshensel et al., 1995; Pearlin et al., 1990). As such, these factors help to identify caregivers who may be especially likely to suffer negative health consequences. Research has shown, for example, that caregivers who are older, less educated, and employed tend to report greater emotional distress, higher burden, and worse physical health (e.g., Chappell, Dujela, & Smith, 2015; Hsu et al., 2014; Pinquart & Sörensen, 2007; Schulz & Eden, 2016). Likewise, caregivers’ own health problems are linked to greater caregiving burden and may hinder their ability to cope with care-related stress (Navaie-Waliser et al., 2002; Pinquart & Sörensen, 2011; Roth, Fredman, & Haley, 2015). Primary Stressors Primary stressors are difficulties directly related to providing care, and involve objective and subjective elements. Primary objective stressors such as care tasks and care recipient impairments are considered to be the starting point of the caregiving stress process (Aneshensel et al., 1995). Primary subjective stressors represent caregivers’ emotional and psychological responses to these stressors, including self-appraisals of caregiving as difficult or too much to handle. Both types of primary stressors have been linked to adverse mental and physical health outcomes (Pinquart & Sörensen, 2006, 2007, 2011; Zarit, Polenick, DePasquale, Liu, & Bangerter, in press). Differentiating between these stressors aids in understanding how the caregiving stress process unfolds, which is valuable in determining targets for interventions to mitigate or prevent care-related distress. Moreover, stress process models posit that primary objective stressors lead to negative outcomes through increases in primary subjective stressors (Aneshensel et al., 1995), forming the basis for exploring potential mechanistic pathways. Secondary Role Strains Caregivers commonly encounter secondary role strains, or hardships experienced in roles and activities indirectly related to caregiving that emanate from primary care-related stressors (Aneshensel et al., 1995; Pearlin et al., 1990). That is, primary caregiving stress proliferates or “spreads” to other life areas. The time, energy, and resources devoted to providing care, for instance, can restrict caregivers’ participation in valued noncare activities (Gaugler et al., 2008; Schulz & Eden, 2016). Reduced participation is also partly a result of caregivers’ feelings of guilt about engaging in noncare activities and/or their reluctance to partake in activities without their partner (Badr et al., 2016; Rudman, Hebert, & Reid, 2006). Cross-sectional studies show that caregivers’ activity restriction is associated with poor health indicators such as higher blood pressure and worse sleep quality (Chattillion et al., 2013; Moore et al., 2011). These associations are likely dynamic and bidirectional, as evidenced by the reciprocal links between care-related activity restriction and depressive symptoms over time (Mausbach et al., 2011; Smith, Williamson, Miller, & Schulz, 2011). Social isolation could also result from activity restriction and has been associated with an increased risk of detrimental outcomes including coronary heart disease, stroke, and mortality (Holt-Lunstad, Smith, Baker, Harris, & Stephenson, 2015; Valtorta, Kanaan, Gilbody, Ronzi, & Hanratty, 2016). Therefore, it is essential to consider how caregiving stress proliferation may restrict spouses’ valued activities. Caregiving stress proliferation may also have reverberating adverse consequences for family relationships. Within the care dyad, a partner’s chronic illness or disability is a shared stressor that can negatively impact marital interactions (Baucom, Porter, Kirby, & Hudepohl, 2012; Berg & Upchurch, 2007). The stress of giving and receiving care, for example, may aggravate preexisting relationship issues and/or contribute to spousal discord that stems from shifting roles and activities in the marriage (Lyons, Zarit, Sayer, & Whitlatch, 2002; Quinn, Clare, & Woods, 2009; Savla, Roberto, Blieszner, Cox, & Gwazdauskas, 2011). Caregiving spouses may also resent their care role because they view their partner as controlling or demanding (Smith et al., 2011). Such negative relationship qualities can amplify caregivers’ emotional distress and result in potentially harmful caregiving behaviors (e.g., yelling, hitting) that pose a serious threat to care recipients’ long-term health and well-being (Smith et al., 2011). Beyond the care dyad, care-related conflict can arise within the larger family system. Although few studies have considered this topic, existing research indicates that caregiving stress is diffused throughout the family and not just limited to active helpers (Schulz & Eden, 2016; Zarit et al., in press). Consequently, families may disagree about multiple aspects of the care situation, including the amount and quality of care, decisions about medical treatment, and the adequacy of one another’s contributions (Aneshensel et al., 1995; Gwyther & Matchar, 2015; Pearlin et al., 1990; Semple, 1992). These disagreements can be an enduring source of stress for caregivers (Aneshensel et al., 1995; Scharlach, Li, & Dalvi, 2006) and reduce the support they receive from their family networks (Gaugler, Mendiondo, Smith, & Schmitt, 2003). As a whole, the literature on marital and family stress in the caregiving context underscores the need for a better understanding of factors linked to these secondary role strains for spousal caregivers. Caregiver Gender as a Moderator of Caregiving Stress Proliferation The nature and consequences of spousal caregiving often vary by gender (e.g., Aneshensel et al., 1995; Calasanti, 2010; Chappell et al., 2015; Pinquart & Sörensen, 2006; Swinkels, Verbakel, & Broese van Groenou, 2017; Thomeer, Reczek, & Umberson, 2015). Caregiving wives and husbands may therefore differ in how their background characteristics and primary care-related stressors are associated with secondary role strains. Compared with men, women perceive more societal expectation to place family members’ needs above their own as well as more family barriers to self-care (Calasanti, 2010; Rosland, Heisler, Choi, Silveira, & Piette, 2010; Thomeer et al., 2015). Women also provide more hours of caregiving than men, help with a higher number and wider range of care tasks, and experience more secondary stressors (Pinquart & Sörensen, 2006; Swinkels et al., 2017). Furthermore, whereas women tend to be emotionally entrenched in the care role, men commonly adopt a managerial caregiving style that emphasizes delegating tasks to other helpers and preserves their own well-being (Calasanti, 2010; Thompson, 2002). Perhaps in part as a result, caregiving women report higher caregiver burden and depressive symptoms, lower subjective well-being, and worse overall physical health than their male counterparts (Pinquart & Sörensen, 2006). In order to more effectively address the needs of spousal caregivers in managing their care role along with other central life domains, it is imperative to evaluate whether and how the experiences of caregiving wives and husbands differ. Considering women’s greater care demands in tandem with their heightened vulnerability to care-related stress, we propose that caregivers’ background characteristics and primary caregiving stressors may have stronger associations with secondary role strains for wives than for husbands. The Present Study Drawing from a nationally representative U.S. sample of spousal caregivers and their partners with functional disability, we examined the associations among caregivers’ background characteristics, primary caregiving stressors, and secondary role strains (care-related valued activity restriction, negative caregiving relationship quality, and care-related family disagreements). We predicted the following: 1) Caregivers’ older age, lower educational attainment, employment, and higher number of chronic health conditions will be positively associated with secondary role strains. 2) Caregivers’ higher levels of primary objective and subjective care-related stressors will be positively associated with secondary role strains. 3) Primary subjective stressors will mediate associations between primary objective stressors and secondary role strains, accounting for background characteristics. 4) The above associations will be significantly stronger for wives than for husbands. Design and Methods Sample and Procedures The sample for this cross-sectional study included 367 community-dwelling adults aged 65 and older with functional disability and their spousal caregivers drawn from the 2011 National Health and Aging Trends Study (NHATS) and the National Study of Caregiving (NSOC). Participants were eligible for NHATS if they were Medicare enrollees aged 65 and older, lived in the contiguous United States, and received health-related assistance in the last month for one or more of the following: mobility, personal care, and/or household chores. Participants were recruited from a Medicare enrollment database using a stratified three-stage sampling design. NHATS participants were eligible for NSOC if they had at least one family or unpaid nonfamily helper who provided health-related assistance with mobility, self-care, household chores, and/or medical care activities. The 2,423 eligible NHATS participants had 4,935 eligible caregivers. Contact information was provided for 3,362 (68.1%) caregivers, from which 2,007 (59.7%) completed a 30-minute telephone interview. In total, 519 (37.9%) of the 1,368 NHATS participants with one or more caregivers who participated in NSOC were married. Of these, 422 (81.3%) spousal caregivers were interviewed. We were interested in spouses of community-dwelling older adults who lived together, and so we removed 32 spouses whose care recipients lived in residential care and two spouses who did not live with their care recipient. A total of 388 spousal caregivers resided with their partner in the community. Of these, 21 were removed due to missing data on study variables. Hence, the analytic sample included 367 spousal caregivers and care recipients with complete data. On average, caregivers had been married to the care recipient for 42.97 years (SD = 18.07, range = 0–72). Table 1 shows caregivers’ background characteristics and scores on study variables. Table 1. Background Characteristics and Key Variables for Caregiving Wives and Husbands Characteristic  Wives (n = 218)  Husbands (n = 149)  t  df  M  SD  M  SD  CG age in years  71.81  9.63  76.22  7.06  −4.89***  357.40  CR age in years  78.00  7.44  75.58  7.13  3.12**  365.00  CG chronic health conditions  2.22  1.43  1.96  1.39  1.73  365.00  CR chronic health conditions  3.04  1.51  3.09  1.39  −0.37  365.00  CG ADL assistance  1.84  2.03  1.49  1.57  1.83  359.45  CG IADL assistance  3.73  0.99  2.98  1.27  6.11***  265.95  CG health system tasks  2.82  1.60  2.11  1.61  4.14***  365.00  CG medical/nursing tasks  2.84  2.06  1.91  1.66  4.79***  355.07  CG emotional difficulties  1.69  1.80  0.80  1.34  5.44***  362.42  CG role overload  1.78  0.65  1.40  0.49  6.27***  361.40  CG negative caregiving relationship quality  2.41  0.84  2.08  0.70  4.10***  350.72    Proportions    χ2  df  CG educational attainment   High school graduate  0.34  0.24  4.40*  1   College graduate  0.08  0.13  0.78  1   Post graduate  0.06  0.09  2.38  1  CG employed for pay  0.16  0.14  0.26  1  CR dementia status  0.22  0.11  6.84**  1  CG valued activity restriction  0.32  0.17  9.85**  1  CG family disagreements  0.13  0.13  0.001  1  Characteristic  Wives (n = 218)  Husbands (n = 149)  t  df  M  SD  M  SD  CG age in years  71.81  9.63  76.22  7.06  −4.89***  357.40  CR age in years  78.00  7.44  75.58  7.13  3.12**  365.00  CG chronic health conditions  2.22  1.43  1.96  1.39  1.73  365.00  CR chronic health conditions  3.04  1.51  3.09  1.39  −0.37  365.00  CG ADL assistance  1.84  2.03  1.49  1.57  1.83  359.45  CG IADL assistance  3.73  0.99  2.98  1.27  6.11***  265.95  CG health system tasks  2.82  1.60  2.11  1.61  4.14***  365.00  CG medical/nursing tasks  2.84  2.06  1.91  1.66  4.79***  355.07  CG emotional difficulties  1.69  1.80  0.80  1.34  5.44***  362.42  CG role overload  1.78  0.65  1.40  0.49  6.27***  361.40  CG negative caregiving relationship quality  2.41  0.84  2.08  0.70  4.10***  350.72    Proportions    χ2  df  CG educational attainment   High school graduate  0.34  0.24  4.40*  1   College graduate  0.08  0.13  0.78  1   Post graduate  0.06  0.09  2.38  1  CG employed for pay  0.16  0.14  0.26  1  CR dementia status  0.22  0.11  6.84**  1  CG valued activity restriction  0.32  0.17  9.85**  1  CG family disagreements  0.13  0.13  0.001  1  Notes. ADL = activities of daily living; CG = caregiver; CR = care recipient; df = degree of freedom; IADL = instrumental activities of daily living; M = mean; SD = standard deviation. N = 367 spousal caregivers. *p ≤ .05. **p ≤ .01. ***p ≤ .001. View Large Table 1. Background Characteristics and Key Variables for Caregiving Wives and Husbands Characteristic  Wives (n = 218)  Husbands (n = 149)  t  df  M  SD  M  SD  CG age in years  71.81  9.63  76.22  7.06  −4.89***  357.40  CR age in years  78.00  7.44  75.58  7.13  3.12**  365.00  CG chronic health conditions  2.22  1.43  1.96  1.39  1.73  365.00  CR chronic health conditions  3.04  1.51  3.09  1.39  −0.37  365.00  CG ADL assistance  1.84  2.03  1.49  1.57  1.83  359.45  CG IADL assistance  3.73  0.99  2.98  1.27  6.11***  265.95  CG health system tasks  2.82  1.60  2.11  1.61  4.14***  365.00  CG medical/nursing tasks  2.84  2.06  1.91  1.66  4.79***  355.07  CG emotional difficulties  1.69  1.80  0.80  1.34  5.44***  362.42  CG role overload  1.78  0.65  1.40  0.49  6.27***  361.40  CG negative caregiving relationship quality  2.41  0.84  2.08  0.70  4.10***  350.72    Proportions    χ2  df  CG educational attainment   High school graduate  0.34  0.24  4.40*  1   College graduate  0.08  0.13  0.78  1   Post graduate  0.06  0.09  2.38  1  CG employed for pay  0.16  0.14  0.26  1  CR dementia status  0.22  0.11  6.84**  1  CG valued activity restriction  0.32  0.17  9.85**  1  CG family disagreements  0.13  0.13  0.001  1  Characteristic  Wives (n = 218)  Husbands (n = 149)  t  df  M  SD  M  SD  CG age in years  71.81  9.63  76.22  7.06  −4.89***  357.40  CR age in years  78.00  7.44  75.58  7.13  3.12**  365.00  CG chronic health conditions  2.22  1.43  1.96  1.39  1.73  365.00  CR chronic health conditions  3.04  1.51  3.09  1.39  −0.37  365.00  CG ADL assistance  1.84  2.03  1.49  1.57  1.83  359.45  CG IADL assistance  3.73  0.99  2.98  1.27  6.11***  265.95  CG health system tasks  2.82  1.60  2.11  1.61  4.14***  365.00  CG medical/nursing tasks  2.84  2.06  1.91  1.66  4.79***  355.07  CG emotional difficulties  1.69  1.80  0.80  1.34  5.44***  362.42  CG role overload  1.78  0.65  1.40  0.49  6.27***  361.40  CG negative caregiving relationship quality  2.41  0.84  2.08  0.70  4.10***  350.72    Proportions    χ2  df  CG educational attainment   High school graduate  0.34  0.24  4.40*  1   College graduate  0.08  0.13  0.78  1   Post graduate  0.06  0.09  2.38  1  CG employed for pay  0.16  0.14  0.26  1  CR dementia status  0.22  0.11  6.84**  1  CG valued activity restriction  0.32  0.17  9.85**  1  CG family disagreements  0.13  0.13  0.001  1  Notes. ADL = activities of daily living; CG = caregiver; CR = care recipient; df = degree of freedom; IADL = instrumental activities of daily living; M = mean; SD = standard deviation. N = 367 spousal caregivers. *p ≤ .05. **p ≤ .01. ***p ≤ .001. View Large Measures Caregiver Gender Gender (1 = female, 0 = male) was examined as a moderator to consider whether associations significantly differed for caregiving wives and husbands. Caregiver Background Characteristics Sociodemographics included caregivers’ age, educational attainment (1 = no schooling completed to 9 = masters, professional, or doctoral degree), and employment status (1 = worked for pay in the last week, 0 = did not work for pay in the last week). Caregivers reported whether they had been diagnosed with seven chronic health conditions (1 = yes, 0 = no): arthritis, cancer, diabetes, heart disease, hypertension, lung disease, and osteoporosis. Summed scores were calculated. Primary Objective Caregiving Stressors Caregivers reported on the number of activities of daily living (ADLs) and instrumental activities of daily living (IADLs) during which they assisted their partner. ADLs included bathing, dressing, eating, toileting, getting in/out of bed, and mobility inside/outside the house. IADLs included laundry, shopping, preparing meals, banking, and managing money. Separate summed scores were created for assistance with ADLs (range = 0–7) and IADLs (range = 0–5). We also examined two types of medical care activities. Health system interaction tasks included ordering medication, scheduling appointments, speaking to providers, helping to change/add insurance or drug plans, and other medical insurance matters. Medical/nursing tasks included giving shots/injections, wound/sore care, teeth/denture care, foot care (e.g., clipping nails), managing medical tasks (e.g., ostomy care, IVs, testing blood), monitoring medications, helping with prescribed exercises, and assisting with a special diet. Spouses reported whether they helped with each task (1 = yes, 0 = no). Separate summed scores were determined for health system interaction tasks (range = 0–5) and medical/nursing tasks (range = 0–8). Lastly, we evaluated two indicators of care recipients’ health and functioning based on care recipient (n = 315) or proxy (n = 50 spouses; n = 1 adult daughter; n = 1 adult son) reports. Dementia status was determined from reports of a physician diagnosis (1 = dementia, 0 = no dementia). Care recipients (or proxies) also reported whether they had been diagnosed with eight chronic health conditions (1 = yes, 0 = no): arthritis, cancer, diabetes, heart disease, hypertension, lung disease, osteoporosis, and stroke. Summed scores were calculated. Primary Subjective Caregiving Stressors Emotional caregiving difficulties were assessed with two items. Caregivers were asked whether caregiving was emotionally difficult for them (1 = yes, 0 = no). Caregivers who reported emotional caregiving difficulties were then asked to rate their extent from 1 (a little difficult) to 5 (very difficult). We combined these items such that caregivers who did not report difficulties were given a zero and caregivers who reported difficulties were assigned a score based on their degree (0 = no difficulty to 5 = high difficulty). As a measure of role overload (Pearlin et al., 1990), we also considered three items assessing how much caregivers feel they (a) are exhausted when they go to bed at night; (b) have more things to do than they can handle; and (c) do not have time for themselves (1 = very much, 2 = somewhat, 3 = not so much). Items were reverse coded and averaged so that higher scores represented greater role overload (α = .75). Secondary Role Strains Care-related valued activity restriction was ascertained from three items. First, caregivers reported whether they participated in four activities in the last month (1 = yes, 0 = no): (a) visiting in person with friends or family; (b) attending religious services; (c) participating in club meetings, classes, or group activities; and (d) going out for enjoyment. Caregivers were then asked whether their care tasks kept them from participating in these activities in the last month (1 = yes, 0 = no), along with the importance they placed on each activity (1 = very important, 2 = somewhat important, 3 = not so important). Most caregivers (73.8%) reported no care-related restriction in their valued activities (i.e., activities rated as “somewhat” or “very” important). As such, we created a binary variable to examine whether caregivers were kept from participating in one or more activities they valued (1 = yes, 0 = no). Negative caregiving relationship quality included two items. Caregivers reported how much the care recipient (a) argues with them; and (b) gets on their nerves from 1 (a lot) to 4 (not at all). Items were reverse coded and averaged such that higher scores reflected greater negative relationship quality. The Spearman–Brown estimate, which is recommended as the appropriate reliability estimate for two-item scales (Eisinga, Grotenhuis, & Pelzer, 2013), was .63. Care-related family disagreements were measured with one item assessing, in general, how much the caregiver’s family disagreed over the care recipient’s care (1 = very much, 2 = somewhat, 3 = not so much). Most caregivers (86.6%) reported that their family disagreed “not so much,” and so we created a binary variable to assess whether family disagreements occurred “somewhat” or “very much” (1 = yes, 0 = no). Statistical Analysis Hierarchical regressions were conducted to evaluate how the addition of each component of our conceptual model contributed to explained variance in the outcome. Regression analyses were performed in SAS 9.4 using the NSOC analytic weight and statistical procedures to account for the complex survey design (Kasper, Freedman, & Spillman, 2013). Linear regressions were estimated to examine how predictors were associated with levels of negative caregiving relationship quality, whereas logistic regressions were estimated to determine their associations with the likelihood of experiencing care-related valued activity restriction or family disagreements. Caregiver gender and background characteristics (age, educational attainment, employment status, chronic health conditions) were entered in Model 1, primary objective caregiving stressors (assistance with ADLs and IADLs, health system interaction tasks, and medical/nursing tasks; care recipients’ dementia status and chronic health conditions) were included in Model 2, and primary subjective caregiving stressors (emotional caregiving difficulties and role overload) were added in Model 3. The first step of the models estimated main effects of the predictors and the second step evaluated gender differences with two-way interaction terms (e.g., Caregiver Age × Caregiver Gender) for the set of predictors added to each model. To test the mediating role of primary subjective stressors in the links between primary objective stressors and secondary role strains, linear regressions were also estimated with caregiver background characteristics as covariates, primary objective stressors as predictors, and emotional caregiving difficulties and role overload as outcomes. The first step examined main effects, whereas the second step tested gender differences with two-way interaction terms between each primary objective stressor and caregiver gender. The nature of significant interactions was explored by estimating simple slopes for wives and husbands. Continuous predictors were mean centered. In accordance with the procedures outlined by Baron and Kenny (1986), mediation occurred if in the aforementioned models: (a) primary objective stressors were significantly linked to secondary role strains before primary subjective stressors were added as predictors; (b) primary objective stressors were significantly associated with primary subjective stressors; (c) primary subjective stressors were significantly linked to secondary role strains with primary objective stressors included as predictors; and (d) associations between primary objective stressors and secondary role strains were reduced to nonsignificance when primary subjective stressors were entered as predictors. Partial mediation resulted when the first three criteria were met, whereas full mediation was achieved through meeting all four criteria. To evaluate whether the mediating process differed for wives and husbands, mediated moderation was examined using the guidelines proposed by Muller, Judd, and Yzerbyt (2005). Mediated moderation occurred if mediation was present, and the direct path along with one or both indirect paths from primary objective stressors to secondary role strains (i.e., through primary subjective stressors) were moderated by gender. Results Independent t-tests and Pearson chi-square tests were examined in preliminary analyses to consider gender differences in study variables (Table 1). Compared with caregiving husbands, caregiving wives were younger, less educated, and cared for older spouses. Wives were more likely to care for a partner with dementia and provided more help with IADLs, health system interaction tasks, and medical/nursing tasks. In addition, wives reported greater emotional caregiving difficulties and role overload. Finally, wives perceived more negative caregiving relationship quality and were more likely to report care-related valued activity restriction. Regression models are presented in Tables 2–4 to ascertain how caregiver background characteristics and primary stressors were associated with secondary role strains, and in Table 5 to explore primary subjective stressors as potential mediators of the associations between primary objective stressors and secondary role strains. Unstandardized coefficients, standard errors, confidence intervals (CIs), and accounted variance are shown from the first step of each model. As measures of effect size, standardized betas are reported for the linear regressions and odds ratios (ORs) are reported for the logistic regressions. Table 2. Predictors of Care-Related Valued Activity Restriction for Spousal Caregivers Predictors  Model 1  Model 2  Model 3  B (SE)  OR  95% CI  B (SE)  OR  95% CI  B (SE)  OR  95% CI  Background characteristics   CG gender (female)  0.85 (0.38)*  2.35  1.09, 5.04  0.79 (0.42)  2.19  0.94, 5.11  0.31 (0.43)  1.37  0.58, 3.24   CG age  0.03 (0.02)  1.03  0.99, 1.06  0.04 (0.02)  1.04  1.00, 1.09  0.05 (0.02)*  1.05  1.00, 1.10   CG educational attainment  0.18 (0.08)*  1.20  1.02, 1.41  0.19 (0.09)*  1.21  1.00, 1.46  0.15 (0.10)  1.17  0.96, 1.41   CG employment status  −0.15 (0.41)  0.86  0.38, 1.97  0.18 (0.45)  1.20  0.48, 2.97  −0.01 (0.43)  0.99  0.42, 2.37   CG chronic conditions  0.21 (0.14)  1.23  0.94, 1.61  0.23 (0.13)  1.26  0.97, 1.63  0.18 (0.13)  1.20  0.92, 1.55  Primary objective stressors   CR chronic conditions  —  —  —  −0.004 (0.11)  1.00  0.81, 1.23  −0.05 (0.11)  0.95  0.76, 1.18   CR dementia status  —  —  —  0.01 (0.37)  1.01  0.49, 2.12  −0.22 (0.41)  0.81  0.35, 1.85   CG ADL assistance  —  —  —  0.25 (0.09)**  1.28  1.07, 1.54  0.19 (0.09)*  1.21  1.00, 1.46   CG IADL assistance  —  —  —  −0.23 (0.14)  0.80  0.61, 1.05  −0.20 (0.15)  0.82  0.61, 1.10   CG health system tasks  —  —  —  0.62 (0.14)***  1.85  1.39, 2.47  0.57 (0.15)***  1.76  1.30, 2.39   CG medical/nursing tasks  —  —  —  0.07 (0.09)  1.07  0.89, 1.29  0.02 (0.09)  1.02  0.85, 1.23  Primary subjective stressors   CG emotional difficulties  –  –  –  –  –  –  0.17 (0.13)  1.18  0.92, 1.52   CG role overload  –  –  –  –  –  –  1.13 (0.36)**  3.08  1.50, 6.32    Total R2  .06  .23  .29  Predictors  Model 1  Model 2  Model 3  B (SE)  OR  95% CI  B (SE)  OR  95% CI  B (SE)  OR  95% CI  Background characteristics   CG gender (female)  0.85 (0.38)*  2.35  1.09, 5.04  0.79 (0.42)  2.19  0.94, 5.11  0.31 (0.43)  1.37  0.58, 3.24   CG age  0.03 (0.02)  1.03  0.99, 1.06  0.04 (0.02)  1.04  1.00, 1.09  0.05 (0.02)*  1.05  1.00, 1.10   CG educational attainment  0.18 (0.08)*  1.20  1.02, 1.41  0.19 (0.09)*  1.21  1.00, 1.46  0.15 (0.10)  1.17  0.96, 1.41   CG employment status  −0.15 (0.41)  0.86  0.38, 1.97  0.18 (0.45)  1.20  0.48, 2.97  −0.01 (0.43)  0.99  0.42, 2.37   CG chronic conditions  0.21 (0.14)  1.23  0.94, 1.61  0.23 (0.13)  1.26  0.97, 1.63  0.18 (0.13)  1.20  0.92, 1.55  Primary objective stressors   CR chronic conditions  —  —  —  −0.004 (0.11)  1.00  0.81, 1.23  −0.05 (0.11)  0.95  0.76, 1.18   CR dementia status  —  —  —  0.01 (0.37)  1.01  0.49, 2.12  −0.22 (0.41)  0.81  0.35, 1.85   CG ADL assistance  —  —  —  0.25 (0.09)**  1.28  1.07, 1.54  0.19 (0.09)*  1.21  1.00, 1.46   CG IADL assistance  —  —  —  −0.23 (0.14)  0.80  0.61, 1.05  −0.20 (0.15)  0.82  0.61, 1.10   CG health system tasks  —  —  —  0.62 (0.14)***  1.85  1.39, 2.47  0.57 (0.15)***  1.76  1.30, 2.39   CG medical/nursing tasks  —  —  —  0.07 (0.09)  1.07  0.89, 1.29  0.02 (0.09)  1.02  0.85, 1.23  Primary subjective stressors   CG emotional difficulties  –  –  –  –  –  –  0.17 (0.13)  1.18  0.92, 1.52   CG role overload  –  –  –  –  –  –  1.13 (0.36)**  3.08  1.50, 6.32    Total R2  .06  .23  .29  Note. ADL = activities of daily living; CG = caregiver; CI = confidence interval; CR = care recipient; IADL = independent activities of daily living; OR = odds ratio; SE = standard error. Estimates are from Step 1 of each model. Step 2 (not shown in table) included gender interactions with background characteristics in Model 1, gender interactions with primary objective stressors in Model 2, and gender interactions with primary subjective stressors in Model 3. There were no significant gender interactions in Step 2. N = 367 spousal caregivers. *p ≤ .05. **p ≤ .01. ***p ≤ .001. View Large Table 2. Predictors of Care-Related Valued Activity Restriction for Spousal Caregivers Predictors  Model 1  Model 2  Model 3  B (SE)  OR  95% CI  B (SE)  OR  95% CI  B (SE)  OR  95% CI  Background characteristics   CG gender (female)  0.85 (0.38)*  2.35  1.09, 5.04  0.79 (0.42)  2.19  0.94, 5.11  0.31 (0.43)  1.37  0.58, 3.24   CG age  0.03 (0.02)  1.03  0.99, 1.06  0.04 (0.02)  1.04  1.00, 1.09  0.05 (0.02)*  1.05  1.00, 1.10   CG educational attainment  0.18 (0.08)*  1.20  1.02, 1.41  0.19 (0.09)*  1.21  1.00, 1.46  0.15 (0.10)  1.17  0.96, 1.41   CG employment status  −0.15 (0.41)  0.86  0.38, 1.97  0.18 (0.45)  1.20  0.48, 2.97  −0.01 (0.43)  0.99  0.42, 2.37   CG chronic conditions  0.21 (0.14)  1.23  0.94, 1.61  0.23 (0.13)  1.26  0.97, 1.63  0.18 (0.13)  1.20  0.92, 1.55  Primary objective stressors   CR chronic conditions  —  —  —  −0.004 (0.11)  1.00  0.81, 1.23  −0.05 (0.11)  0.95  0.76, 1.18   CR dementia status  —  —  —  0.01 (0.37)  1.01  0.49, 2.12  −0.22 (0.41)  0.81  0.35, 1.85   CG ADL assistance  —  —  —  0.25 (0.09)**  1.28  1.07, 1.54  0.19 (0.09)*  1.21  1.00, 1.46   CG IADL assistance  —  —  —  −0.23 (0.14)  0.80  0.61, 1.05  −0.20 (0.15)  0.82  0.61, 1.10   CG health system tasks  —  —  —  0.62 (0.14)***  1.85  1.39, 2.47  0.57 (0.15)***  1.76  1.30, 2.39   CG medical/nursing tasks  —  —  —  0.07 (0.09)  1.07  0.89, 1.29  0.02 (0.09)  1.02  0.85, 1.23  Primary subjective stressors   CG emotional difficulties  –  –  –  –  –  –  0.17 (0.13)  1.18  0.92, 1.52   CG role overload  –  –  –  –  –  –  1.13 (0.36)**  3.08  1.50, 6.32    Total R2  .06  .23  .29  Predictors  Model 1  Model 2  Model 3  B (SE)  OR  95% CI  B (SE)  OR  95% CI  B (SE)  OR  95% CI  Background characteristics   CG gender (female)  0.85 (0.38)*  2.35  1.09, 5.04  0.79 (0.42)  2.19  0.94, 5.11  0.31 (0.43)  1.37  0.58, 3.24   CG age  0.03 (0.02)  1.03  0.99, 1.06  0.04 (0.02)  1.04  1.00, 1.09  0.05 (0.02)*  1.05  1.00, 1.10   CG educational attainment  0.18 (0.08)*  1.20  1.02, 1.41  0.19 (0.09)*  1.21  1.00, 1.46  0.15 (0.10)  1.17  0.96, 1.41   CG employment status  −0.15 (0.41)  0.86  0.38, 1.97  0.18 (0.45)  1.20  0.48, 2.97  −0.01 (0.43)  0.99  0.42, 2.37   CG chronic conditions  0.21 (0.14)  1.23  0.94, 1.61  0.23 (0.13)  1.26  0.97, 1.63  0.18 (0.13)  1.20  0.92, 1.55  Primary objective stressors   CR chronic conditions  —  —  —  −0.004 (0.11)  1.00  0.81, 1.23  −0.05 (0.11)  0.95  0.76, 1.18   CR dementia status  —  —  —  0.01 (0.37)  1.01  0.49, 2.12  −0.22 (0.41)  0.81  0.35, 1.85   CG ADL assistance  —  —  —  0.25 (0.09)**  1.28  1.07, 1.54  0.19 (0.09)*  1.21  1.00, 1.46   CG IADL assistance  —  —  —  −0.23 (0.14)  0.80  0.61, 1.05  −0.20 (0.15)  0.82  0.61, 1.10   CG health system tasks  —  —  —  0.62 (0.14)***  1.85  1.39, 2.47  0.57 (0.15)***  1.76  1.30, 2.39   CG medical/nursing tasks  —  —  —  0.07 (0.09)  1.07  0.89, 1.29  0.02 (0.09)  1.02  0.85, 1.23  Primary subjective stressors   CG emotional difficulties  –  –  –  –  –  –  0.17 (0.13)  1.18  0.92, 1.52   CG role overload  –  –  –  –  –  –  1.13 (0.36)**  3.08  1.50, 6.32    Total R2  .06  .23  .29  Note. ADL = activities of daily living; CG = caregiver; CI = confidence interval; CR = care recipient; IADL = independent activities of daily living; OR = odds ratio; SE = standard error. Estimates are from Step 1 of each model. Step 2 (not shown in table) included gender interactions with background characteristics in Model 1, gender interactions with primary objective stressors in Model 2, and gender interactions with primary subjective stressors in Model 3. There were no significant gender interactions in Step 2. N = 367 spousal caregivers. *p ≤ .05. **p ≤ .01. ***p ≤ .001. View Large Table 4. Predictors of Care-Related Family Disagreements for Spousal Caregivers Predictors  Model 1  Model 2  Model 3  B (SE)  OR  95% CI  B (SE)  OR  95% CI  B (SE)  OR  95% CI  Background characteristics   CG gender (female)  0.14 (0.34)  1.15  0.58, 2.27  −0.02 (0.47)  0.99  0.39, 2.51  −0.19 (0.45)  0.83  0.34, 2.04   CG age  −0.02 (0.03)  0.98  0.93, 1.03  −0.03 (0.03)  0.97  0.92, 1.02  −0.04 (0.03)  0.97  0.92, 1.02   CG educational attainment  −0.10 (0.11)  0.91  0.73, 1.13  −0.10 (0.11)  0.90  0.72, 1.13  −0.09 (0.11)  0.91  0.74, 1.14   CG employment status  −0.94 (0.71)  0.39  0.09, 1.62  −0.88 (0.70)  0.41  0.10, 1.68  −1.01 (0.73)  0.36  0.08, 1.58   CG chronic conditions  −0.03 (0.14)  0.97  0.74, 1.28  −0.06 (0.13)  0.95  0.73, 1.22  −0.03 (0.13)  0.97  0.75, 1.24  Primary objective stressors   CR chronic conditions  —  —  —  −0.06 (0.15)  0.94  0.70, 1.25  −0.09(0.15)  0.92  0.68, 1.23   CR dementia statusa  —  —  —  0.89 (0.54)  2.43  0.83, 7.11  0.92 (0.53)  2.51  0.87, 7.23   CG ADL assistance  —  —  —  0.01 (0.12)  1.01  0.79, 1.29  −0.004 (0.12)  0.98  0.78, 1.27   CG IADL assistance  —  —  —  −0.26 (0.19)  0.77  0.52, 1.14  −0.23 (0.19)  0.79  0.54, 1.16   CG health system tasks  —  —  —  0.11 (0.14)  1.12  0.85, 1.47  0.09 (0.15)  1.09  0.82, 1.46   CG medical/nursing tasks  —  —  —  0.18 (0.12)  1.20  0.94, 1.54  0.14 (0.12)  1.15  0.91, 1.47  Primary subjective stressors   CG emotional difficulties  —  —  —  —  —  —  −0.20 (0.14)  0.82  0.62, 1.07   CG role overload  —  —  —  —  —  —  0.88 (0.36)*  2.40  1.17, 4.94    Total R2  .02  .06  .09  Predictors  Model 1  Model 2  Model 3  B (SE)  OR  95% CI  B (SE)  OR  95% CI  B (SE)  OR  95% CI  Background characteristics   CG gender (female)  0.14 (0.34)  1.15  0.58, 2.27  −0.02 (0.47)  0.99  0.39, 2.51  −0.19 (0.45)  0.83  0.34, 2.04   CG age  −0.02 (0.03)  0.98  0.93, 1.03  −0.03 (0.03)  0.97  0.92, 1.02  −0.04 (0.03)  0.97  0.92, 1.02   CG educational attainment  −0.10 (0.11)  0.91  0.73, 1.13  −0.10 (0.11)  0.90  0.72, 1.13  −0.09 (0.11)  0.91  0.74, 1.14   CG employment status  −0.94 (0.71)  0.39  0.09, 1.62  −0.88 (0.70)  0.41  0.10, 1.68  −1.01 (0.73)  0.36  0.08, 1.58   CG chronic conditions  −0.03 (0.14)  0.97  0.74, 1.28  −0.06 (0.13)  0.95  0.73, 1.22  −0.03 (0.13)  0.97  0.75, 1.24  Primary objective stressors   CR chronic conditions  —  —  —  −0.06 (0.15)  0.94  0.70, 1.25  −0.09(0.15)  0.92  0.68, 1.23   CR dementia statusa  —  —  —  0.89 (0.54)  2.43  0.83, 7.11  0.92 (0.53)  2.51  0.87, 7.23   CG ADL assistance  —  —  —  0.01 (0.12)  1.01  0.79, 1.29  −0.004 (0.12)  0.98  0.78, 1.27   CG IADL assistance  —  —  —  −0.26 (0.19)  0.77  0.52, 1.14  −0.23 (0.19)  0.79  0.54, 1.16   CG health system tasks  —  —  —  0.11 (0.14)  1.12  0.85, 1.47  0.09 (0.15)  1.09  0.82, 1.46   CG medical/nursing tasks  —  —  —  0.18 (0.12)  1.20  0.94, 1.54  0.14 (0.12)  1.15  0.91, 1.47  Primary subjective stressors   CG emotional difficulties  —  —  —  —  —  —  −0.20 (0.14)  0.82  0.62, 1.07   CG role overload  —  —  —  —  —  —  0.88 (0.36)*  2.40  1.17, 4.94    Total R2  .02  .06  .09  Note. ADL = activities of daily living; CG = caregiver; CI = confidence interval; CR = care recipient; IADL = independent activities of daily living; OR = odds ratio; SE = standard error. Estimates are from Step 1 of each model. Step 2 (not shown in table) included gender interactions with background characteristics in Model 1, gender interactions with primary objective stressors in Model 2, and gender interactions with primary subjective stressors in Model 3. N = 367 spousal caregivers. aDenotes significant gender interaction found in Step 2. *p ≤ .05. **p ≤ .01. ***p ≤ .001. View Large Table 4. Predictors of Care-Related Family Disagreements for Spousal Caregivers Predictors  Model 1  Model 2  Model 3  B (SE)  OR  95% CI  B (SE)  OR  95% CI  B (SE)  OR  95% CI  Background characteristics   CG gender (female)  0.14 (0.34)  1.15  0.58, 2.27  −0.02 (0.47)  0.99  0.39, 2.51  −0.19 (0.45)  0.83  0.34, 2.04   CG age  −0.02 (0.03)  0.98  0.93, 1.03  −0.03 (0.03)  0.97  0.92, 1.02  −0.04 (0.03)  0.97  0.92, 1.02   CG educational attainment  −0.10 (0.11)  0.91  0.73, 1.13  −0.10 (0.11)  0.90  0.72, 1.13  −0.09 (0.11)  0.91  0.74, 1.14   CG employment status  −0.94 (0.71)  0.39  0.09, 1.62  −0.88 (0.70)  0.41  0.10, 1.68  −1.01 (0.73)  0.36  0.08, 1.58   CG chronic conditions  −0.03 (0.14)  0.97  0.74, 1.28  −0.06 (0.13)  0.95  0.73, 1.22  −0.03 (0.13)  0.97  0.75, 1.24  Primary objective stressors   CR chronic conditions  —  —  —  −0.06 (0.15)  0.94  0.70, 1.25  −0.09(0.15)  0.92  0.68, 1.23   CR dementia statusa  —  —  —  0.89 (0.54)  2.43  0.83, 7.11  0.92 (0.53)  2.51  0.87, 7.23   CG ADL assistance  —  —  —  0.01 (0.12)  1.01  0.79, 1.29  −0.004 (0.12)  0.98  0.78, 1.27   CG IADL assistance  —  —  —  −0.26 (0.19)  0.77  0.52, 1.14  −0.23 (0.19)  0.79  0.54, 1.16   CG health system tasks  —  —  —  0.11 (0.14)  1.12  0.85, 1.47  0.09 (0.15)  1.09  0.82, 1.46   CG medical/nursing tasks  —  —  —  0.18 (0.12)  1.20  0.94, 1.54  0.14 (0.12)  1.15  0.91, 1.47  Primary subjective stressors   CG emotional difficulties  —  —  —  —  —  —  −0.20 (0.14)  0.82  0.62, 1.07   CG role overload  —  —  —  —  —  —  0.88 (0.36)*  2.40  1.17, 4.94    Total R2  .02  .06  .09  Predictors  Model 1  Model 2  Model 3  B (SE)  OR  95% CI  B (SE)  OR  95% CI  B (SE)  OR  95% CI  Background characteristics   CG gender (female)  0.14 (0.34)  1.15  0.58, 2.27  −0.02 (0.47)  0.99  0.39, 2.51  −0.19 (0.45)  0.83  0.34, 2.04   CG age  −0.02 (0.03)  0.98  0.93, 1.03  −0.03 (0.03)  0.97  0.92, 1.02  −0.04 (0.03)  0.97  0.92, 1.02   CG educational attainment  −0.10 (0.11)  0.91  0.73, 1.13  −0.10 (0.11)  0.90  0.72, 1.13  −0.09 (0.11)  0.91  0.74, 1.14   CG employment status  −0.94 (0.71)  0.39  0.09, 1.62  −0.88 (0.70)  0.41  0.10, 1.68  −1.01 (0.73)  0.36  0.08, 1.58   CG chronic conditions  −0.03 (0.14)  0.97  0.74, 1.28  −0.06 (0.13)  0.95  0.73, 1.22  −0.03 (0.13)  0.97  0.75, 1.24  Primary objective stressors   CR chronic conditions  —  —  —  −0.06 (0.15)  0.94  0.70, 1.25  −0.09(0.15)  0.92  0.68, 1.23   CR dementia statusa  —  —  —  0.89 (0.54)  2.43  0.83, 7.11  0.92 (0.53)  2.51  0.87, 7.23   CG ADL assistance  —  —  —  0.01 (0.12)  1.01  0.79, 1.29  −0.004 (0.12)  0.98  0.78, 1.27   CG IADL assistance  —  —  —  −0.26 (0.19)  0.77  0.52, 1.14  −0.23 (0.19)  0.79  0.54, 1.16   CG health system tasks  —  —  —  0.11 (0.14)  1.12  0.85, 1.47  0.09 (0.15)  1.09  0.82, 1.46   CG medical/nursing tasks  —  —  —  0.18 (0.12)  1.20  0.94, 1.54  0.14 (0.12)  1.15  0.91, 1.47  Primary subjective stressors   CG emotional difficulties  —  —  —  —  —  —  −0.20 (0.14)  0.82  0.62, 1.07   CG role overload  —  —  —  —  —  —  0.88 (0.36)*  2.40  1.17, 4.94    Total R2  .02  .06  .09  Note. ADL = activities of daily living; CG = caregiver; CI = confidence interval; CR = care recipient; IADL = independent activities of daily living; OR = odds ratio; SE = standard error. Estimates are from Step 1 of each model. Step 2 (not shown in table) included gender interactions with background characteristics in Model 1, gender interactions with primary objective stressors in Model 2, and gender interactions with primary subjective stressors in Model 3. N = 367 spousal caregivers. aDenotes significant gender interaction found in Step 2. *p ≤ .05. **p ≤ .01. ***p ≤ .001. View Large Table 5. Predictors of Primary Subjective Stressors for Spousal Caregivers Predictors  Emotional difficulties  Role overload  B (SE)  β  95% CI  B (SE)  β  95% CI  Background characteristics   CG gender (female)  0.52 (0.19)**  0.16  0.14, 0.91  0.29 (0.07)***  0.24  0.16, 0.43   CG age  −0.01 (0.01)  −0.07  −0.04, 0.02  −0.002 (0.01)  −0.03  −0.01, 0.01   CG educational attainment  0.14 (0.05)**  0.17  0.04, 0.23  0.03 (0.02)  0.09  −0.01, 0.06   CG employment status  0.16 (0.25)  0.04  −0.35, 0.66  0.08 (0.09)  0.05  −0.10, 0.26   CG chronic conditions  0.18 (0.06)**  0.16  0.06, 0.31  0.02 (0.02)  0.05  −0.03, 0.06  Primary objective stressors   CR chronic conditions  0.09 (0.05)  0.08  −0.02, 0.19  0.04 (0.02)*  0.09  0.001, 0.08   CR dementia status  0.74 (0.26)**  0.15  0.23, 1.26  0.10 (0.10)  0.06  −0.10, 0.30   CG ADL assistance  0.10 (0.06)  0.11  −0.02, 0.23  0.05 (0.02)*  0.14  0.01, 0.09   CG IADL assistance  −0.04 (0.08)  −0.03  −0.19, 0.12  −0.03 (0.03)  −0.06  −0.08, 0.03   CG health system tasks  0.19 (0.05)***  0.19  0.09, 0.29  0.07 (0.02)***  0.19  0.03, 0.11   CG medical/nursing tasks  0.002 (0.04)  0.002  −0.08, 0.09  0.05 (0.02)**  0.16  0.01, 0.09    Total R2  .22  .26  Predictors  Emotional difficulties  Role overload  B (SE)  β  95% CI  B (SE)  β  95% CI  Background characteristics   CG gender (female)  0.52 (0.19)**  0.16  0.14, 0.91  0.29 (0.07)***  0.24  0.16, 0.43   CG age  −0.01 (0.01)  −0.07  −0.04, 0.02  −0.002 (0.01)  −0.03  −0.01, 0.01   CG educational attainment  0.14 (0.05)**  0.17  0.04, 0.23  0.03 (0.02)  0.09  −0.01, 0.06   CG employment status  0.16 (0.25)  0.04  −0.35, 0.66  0.08 (0.09)  0.05  −0.10, 0.26   CG chronic conditions  0.18 (0.06)**  0.16  0.06, 0.31  0.02 (0.02)  0.05  −0.03, 0.06  Primary objective stressors   CR chronic conditions  0.09 (0.05)  0.08  −0.02, 0.19  0.04 (0.02)*  0.09  0.001, 0.08   CR dementia status  0.74 (0.26)**  0.15  0.23, 1.26  0.10 (0.10)  0.06  −0.10, 0.30   CG ADL assistance  0.10 (0.06)  0.11  −0.02, 0.23  0.05 (0.02)*  0.14  0.01, 0.09   CG IADL assistance  −0.04 (0.08)  −0.03  −0.19, 0.12  −0.03 (0.03)  −0.06  −0.08, 0.03   CG health system tasks  0.19 (0.05)***  0.19  0.09, 0.29  0.07 (0.02)***  0.19  0.03, 0.11   CG medical/nursing tasks  0.002 (0.04)  0.002  −0.08, 0.09  0.05 (0.02)**  0.16  0.01, 0.09    Total R2  .22  .26  Note. ADL = activities of daily living; CG = caregiver; CI = confidence interval; CR = care recipient; IADL = independent activities of daily living; SE = standard error. Estimates are from Step 1 of each model. Step 2 (not shown in table) included gender interactions with each primary objective stressor. There were no significant gender interactions in Step 2. N = 367 spousal caregivers. *p ≤ .05. **p ≤ .01. ***p ≤ .001. View Large Table 5. Predictors of Primary Subjective Stressors for Spousal Caregivers Predictors  Emotional difficulties  Role overload  B (SE)  β  95% CI  B (SE)  β  95% CI  Background characteristics   CG gender (female)  0.52 (0.19)**  0.16  0.14, 0.91  0.29 (0.07)***  0.24  0.16, 0.43   CG age  −0.01 (0.01)  −0.07  −0.04, 0.02  −0.002 (0.01)  −0.03  −0.01, 0.01   CG educational attainment  0.14 (0.05)**  0.17  0.04, 0.23  0.03 (0.02)  0.09  −0.01, 0.06   CG employment status  0.16 (0.25)  0.04  −0.35, 0.66  0.08 (0.09)  0.05  −0.10, 0.26   CG chronic conditions  0.18 (0.06)**  0.16  0.06, 0.31  0.02 (0.02)  0.05  −0.03, 0.06  Primary objective stressors   CR chronic conditions  0.09 (0.05)  0.08  −0.02, 0.19  0.04 (0.02)*  0.09  0.001, 0.08   CR dementia status  0.74 (0.26)**  0.15  0.23, 1.26  0.10 (0.10)  0.06  −0.10, 0.30   CG ADL assistance  0.10 (0.06)  0.11  −0.02, 0.23  0.05 (0.02)*  0.14  0.01, 0.09   CG IADL assistance  −0.04 (0.08)  −0.03  −0.19, 0.12  −0.03 (0.03)  −0.06  −0.08, 0.03   CG health system tasks  0.19 (0.05)***  0.19  0.09, 0.29  0.07 (0.02)***  0.19  0.03, 0.11   CG medical/nursing tasks  0.002 (0.04)  0.002  −0.08, 0.09  0.05 (0.02)**  0.16  0.01, 0.09    Total R2  .22  .26  Predictors  Emotional difficulties  Role overload  B (SE)  β  95% CI  B (SE)  β  95% CI  Background characteristics   CG gender (female)  0.52 (0.19)**  0.16  0.14, 0.91  0.29 (0.07)***  0.24  0.16, 0.43   CG age  −0.01 (0.01)  −0.07  −0.04, 0.02  −0.002 (0.01)  −0.03  −0.01, 0.01   CG educational attainment  0.14 (0.05)**  0.17  0.04, 0.23  0.03 (0.02)  0.09  −0.01, 0.06   CG employment status  0.16 (0.25)  0.04  −0.35, 0.66  0.08 (0.09)  0.05  −0.10, 0.26   CG chronic conditions  0.18 (0.06)**  0.16  0.06, 0.31  0.02 (0.02)  0.05  −0.03, 0.06  Primary objective stressors   CR chronic conditions  0.09 (0.05)  0.08  −0.02, 0.19  0.04 (0.02)*  0.09  0.001, 0.08   CR dementia status  0.74 (0.26)**  0.15  0.23, 1.26  0.10 (0.10)  0.06  −0.10, 0.30   CG ADL assistance  0.10 (0.06)  0.11  −0.02, 0.23  0.05 (0.02)*  0.14  0.01, 0.09   CG IADL assistance  −0.04 (0.08)  −0.03  −0.19, 0.12  −0.03 (0.03)  −0.06  −0.08, 0.03   CG health system tasks  0.19 (0.05)***  0.19  0.09, 0.29  0.07 (0.02)***  0.19  0.03, 0.11   CG medical/nursing tasks  0.002 (0.04)  0.002  −0.08, 0.09  0.05 (0.02)**  0.16  0.01, 0.09    Total R2  .22  .26  Note. ADL = activities of daily living; CG = caregiver; CI = confidence interval; CR = care recipient; IADL = independent activities of daily living; SE = standard error. Estimates are from Step 1 of each model. Step 2 (not shown in table) included gender interactions with each primary objective stressor. There were no significant gender interactions in Step 2. N = 367 spousal caregivers. *p ≤ .05. **p ≤ .01. ***p ≤ .001. View Large Predictors of Secondary Role Strains Care-Related Valued Activity Restriction Table 2 shows that caregivers’ female gender (Model 1) and higher educational attainment (Models 1 and 2) were linked to a greater likelihood of care-related valued activity restriction. In Models 2 and 3, caregivers who provided more assistance with ADLs and health system interaction tasks were also more likely to report valued activity restriction. Beyond the aforementioned predictors, Model 3 reveals that older caregivers and those with higher role overload reported a greater likelihood of valued activity restriction. There were no significant gender differences in these associations (not shown in table). As presented in Table 5, assistance with ADLs was positively linked to role overload. Similarly, health system interaction tasks were positively linked to role overload and emotional caregiving difficulties. Accordingly, role overload was associated with indirect paths for both care tasks (Table 5) but did not reduce their associations with the outcome to nonsignificance after accounting for the link between role overload and the outcome (Model 3 in Table 2). Thus, role overload partially mediated the links between these tasks and valued activity restriction. Negative Caregiving Relationship Quality Table 3 shows that caregivers’ female gender and higher education were linked to greater negative caregiving relationship quality in Models 1 and 2. Caring for a partner with dementia was also associated with higher negative relationship quality in Model 2. This association was reduced to nonsignificance with the addition of primary subjective stressors in Model 3, demonstrating that caregivers with greater emotional caregiving difficulties and role overload reported higher negative caregiving relationship quality. Table 3. Predictors of Negative Caregiving Relationship Quality for Spousal Caregivers Predictors  Model 1  Model 2  Model 3  B (SE)  β  95% CI  B (SE)  β  95% CI  B (SE)  β  95% CI  Background characteristics   CG gender (female)  0.25 (0.10)*  0.16  0.05, 0.45  0.27 (0.10)**  0.17  0.07, 0.47  0.11 (0.10)  0.07  −0.08, 0.31   CG age  −0.01 (0.01)  −0.08  −0.02, 0.01  −0.01 (0.01)  −0.08  −0.02, 0.01  −0.01 (0.01)  −0.06  −0.02, 0.01   CG educational attainment  0.05 (0.02)*  0.11  0.01, 0.09  0.05 (0.02)*  0.12  0.01, 0.09  0.02 (0.02)  0.05  −0.02, 0.06   CG employment status  0.05 (0.19)  0.02  −0.33, 0.43  0.08 (0.18)  0.04  −0.28, 0.43  0.03 (0.15)  0.02  −0.27, 0.33   CG chronic conditions  0.06 (0.03)  0.10  −0.01, 0.13  0.05 (0.03)  0.09  −0.02, 0.12  0.02 (0.03)  0.03  −0.04, 0.07  Primary objective stressors   CR chronic conditionsa  —  —  —  0.05 (0.03)  0.09  −0.02, 0.11  0.03 (0.03)  0.05  −0.03, 0.08   CR dementia status  —  —  —  0.30 (0.13)*  0.13  0.04, 0.57  0.16 (0.12)  0.07  −0.08, 0.39   CG ADL assistance  —  —  —  −0.002 (0.03)  −0.004  −0.06, 0.06  −0.03 (0.03)  −0.07  −0.09, 0.03   CG IADL assistance  —  —  —  −0.06 (0.05)  −0.08  −0.15, 0.04  −0.04 (0.04)  −0.07  −0.12, 0.04   CG health system tasks  —  —  —  0.03 (0.04)  0.05  −0.05, 0.10  −0.02 (0.04)  −0.05  −0.10, 0.05   CG medical/nursing tasks  —  —  —  −0.001 (0.03)  −0.003  −0.06, 0.05  −0.01 (0.03)  −0.03  −0.06, 0.04  Primary subjective stressors   CG emotional difficulties  —  —  —  —  —  —  0.16 (0.03)***  0.33  0.10, 0.23   CG role overload  —  —  —  —  —  —  0.25 (0.10)**  0.19  0.05, 0.44    Total R2  .06  .09  .24  Predictors  Model 1  Model 2  Model 3  B (SE)  β  95% CI  B (SE)  β  95% CI  B (SE)  β  95% CI  Background characteristics   CG gender (female)  0.25 (0.10)*  0.16  0.05, 0.45  0.27 (0.10)**  0.17  0.07, 0.47  0.11 (0.10)  0.07  −0.08, 0.31   CG age  −0.01 (0.01)  −0.08  −0.02, 0.01  −0.01 (0.01)  −0.08  −0.02, 0.01  −0.01 (0.01)  −0.06  −0.02, 0.01   CG educational attainment  0.05 (0.02)*  0.11  0.01, 0.09  0.05 (0.02)*  0.12  0.01, 0.09  0.02 (0.02)  0.05  −0.02, 0.06   CG employment status  0.05 (0.19)  0.02  −0.33, 0.43  0.08 (0.18)  0.04  −0.28, 0.43  0.03 (0.15)  0.02  −0.27, 0.33   CG chronic conditions  0.06 (0.03)  0.10  −0.01, 0.13  0.05 (0.03)  0.09  −0.02, 0.12  0.02 (0.03)  0.03  −0.04, 0.07  Primary objective stressors   CR chronic conditionsa  —  —  —  0.05 (0.03)  0.09  −0.02, 0.11  0.03 (0.03)  0.05  −0.03, 0.08   CR dementia status  —  —  —  0.30 (0.13)*  0.13  0.04, 0.57  0.16 (0.12)  0.07  −0.08, 0.39   CG ADL assistance  —  —  —  −0.002 (0.03)  −0.004  −0.06, 0.06  −0.03 (0.03)  −0.07  −0.09, 0.03   CG IADL assistance  —  —  —  −0.06 (0.05)  −0.08  −0.15, 0.04  −0.04 (0.04)  −0.07  −0.12, 0.04   CG health system tasks  —  —  —  0.03 (0.04)  0.05  −0.05, 0.10  −0.02 (0.04)  −0.05  −0.10, 0.05   CG medical/nursing tasks  —  —  —  −0.001 (0.03)  −0.003  −0.06, 0.05  −0.01 (0.03)  −0.03  −0.06, 0.04  Primary subjective stressors   CG emotional difficulties  —  —  —  —  —  —  0.16 (0.03)***  0.33  0.10, 0.23   CG role overload  —  —  —  —  —  —  0.25 (0.10)**  0.19  0.05, 0.44    Total R2  .06  .09  .24  Note. ADL = activities of daily living; CG = caregiver; CI = confidence interval; CR = care recipient; IADL = independent activities of daily living; SE = standard error. Estimates are from Step 1 of each model. Step 2 (not shown in table) included gender interactions with background characteristics in Model 1, gender interactions with primary objective stressors in Model 2, and gender interactions with primary subjective stressors in Model 3. N = 367 spousal caregivers. aDenotes significant gender interaction found in Step 2. *p ≤ .05. **p ≤ .01. ***p ≤ .001. View Large Table 3. Predictors of Negative Caregiving Relationship Quality for Spousal Caregivers Predictors  Model 1  Model 2  Model 3  B (SE)  β  95% CI  B (SE)  β  95% CI  B (SE)  β  95% CI  Background characteristics   CG gender (female)  0.25 (0.10)*  0.16  0.05, 0.45  0.27 (0.10)**  0.17  0.07, 0.47  0.11 (0.10)  0.07  −0.08, 0.31   CG age  −0.01 (0.01)  −0.08  −0.02, 0.01  −0.01 (0.01)  −0.08  −0.02, 0.01  −0.01 (0.01)  −0.06  −0.02, 0.01   CG educational attainment  0.05 (0.02)*  0.11  0.01, 0.09  0.05 (0.02)*  0.12  0.01, 0.09  0.02 (0.02)  0.05  −0.02, 0.06   CG employment status  0.05 (0.19)  0.02  −0.33, 0.43  0.08 (0.18)  0.04  −0.28, 0.43  0.03 (0.15)  0.02  −0.27, 0.33   CG chronic conditions  0.06 (0.03)  0.10  −0.01, 0.13  0.05 (0.03)  0.09  −0.02, 0.12  0.02 (0.03)  0.03  −0.04, 0.07  Primary objective stressors   CR chronic conditionsa  —  —  —  0.05 (0.03)  0.09  −0.02, 0.11  0.03 (0.03)  0.05  −0.03, 0.08   CR dementia status  —  —  —  0.30 (0.13)*  0.13  0.04, 0.57  0.16 (0.12)  0.07  −0.08, 0.39   CG ADL assistance  —  —  —  −0.002 (0.03)  −0.004  −0.06, 0.06  −0.03 (0.03)  −0.07  −0.09, 0.03   CG IADL assistance  —  —  —  −0.06 (0.05)  −0.08  −0.15, 0.04  −0.04 (0.04)  −0.07  −0.12, 0.04   CG health system tasks  —  —  —  0.03 (0.04)  0.05  −0.05, 0.10  −0.02 (0.04)  −0.05  −0.10, 0.05   CG medical/nursing tasks  —  —  —  −0.001 (0.03)  −0.003  −0.06, 0.05  −0.01 (0.03)  −0.03  −0.06, 0.04  Primary subjective stressors   CG emotional difficulties  —  —  —  —  —  —  0.16 (0.03)***  0.33  0.10, 0.23   CG role overload  —  —  —  —  —  —  0.25 (0.10)**  0.19  0.05, 0.44    Total R2  .06  .09  .24  Predictors  Model 1  Model 2  Model 3  B (SE)  β  95% CI  B (SE)  β  95% CI  B (SE)  β  95% CI  Background characteristics   CG gender (female)  0.25 (0.10)*  0.16  0.05, 0.45  0.27 (0.10)**  0.17  0.07, 0.47  0.11 (0.10)  0.07  −0.08, 0.31   CG age  −0.01 (0.01)  −0.08  −0.02, 0.01  −0.01 (0.01)  −0.08  −0.02, 0.01  −0.01 (0.01)  −0.06  −0.02, 0.01   CG educational attainment  0.05 (0.02)*  0.11  0.01, 0.09  0.05 (0.02)*  0.12  0.01, 0.09  0.02 (0.02)  0.05  −0.02, 0.06   CG employment status  0.05 (0.19)  0.02  −0.33, 0.43  0.08 (0.18)  0.04  −0.28, 0.43  0.03 (0.15)  0.02  −0.27, 0.33   CG chronic conditions  0.06 (0.03)  0.10  −0.01, 0.13  0.05 (0.03)  0.09  −0.02, 0.12  0.02 (0.03)  0.03  −0.04, 0.07  Primary objective stressors   CR chronic conditionsa  —  —  —  0.05 (0.03)  0.09  −0.02, 0.11  0.03 (0.03)  0.05  −0.03, 0.08   CR dementia status  —  —  —  0.30 (0.13)*  0.13  0.04, 0.57  0.16 (0.12)  0.07  −0.08, 0.39   CG ADL assistance  —  —  —  −0.002 (0.03)  −0.004  −0.06, 0.06  −0.03 (0.03)  −0.07  −0.09, 0.03   CG IADL assistance  —  —  —  −0.06 (0.05)  −0.08  −0.15, 0.04  −0.04 (0.04)  −0.07  −0.12, 0.04   CG health system tasks  —  —  —  0.03 (0.04)  0.05  −0.05, 0.10  −0.02 (0.04)  −0.05  −0.10, 0.05   CG medical/nursing tasks  —  —  —  −0.001 (0.03)  −0.003  −0.06, 0.05  −0.01 (0.03)  −0.03  −0.06, 0.04  Primary subjective stressors   CG emotional difficulties  —  —  —  —  —  —  0.16 (0.03)***  0.33  0.10, 0.23   CG role overload  —  —  —  —  —  —  0.25 (0.10)**  0.19  0.05, 0.44    Total R2  .06  .09  .24  Note. ADL = activities of daily living; CG = caregiver; CI = confidence interval; CR = care recipient; IADL = independent activities of daily living; SE = standard error. Estimates are from Step 1 of each model. Step 2 (not shown in table) included gender interactions with background characteristics in Model 1, gender interactions with primary objective stressors in Model 2, and gender interactions with primary subjective stressors in Model 3. N = 367 spousal caregivers. aDenotes significant gender interaction found in Step 2. *p ≤ .05. **p ≤ .01. ***p ≤ .001. View Large Table 5 reveals that caring for a partner with dementia was associated with greater emotional caregiving difficulties. Emotional difficulties were therefore linked to the indirect path between dementia status and negative caregiving relationship quality (Table 5), and reduced this association to nonsignificance when accounting for the link between emotional difficulties and the outcome (Model 3 in Table 3). This provides evidence that emotional difficulties fully mediated the association between dementia status and negative caregiving relationship quality. A significant interaction emerged between care recipients’ chronic health conditions and caregiver gender in Model 2 (B = −0.17, β = −0.24, p = .02, 95% CI [−0.31, −0.03]). Husbands reported more negative caregiving relationship quality when their partner had a higher number of chronic conditions (B = 0.15, β = 0.27, p = .002, 95% CI [0.06, 0.25]); but this link was not found for wives (B = −0.02, β = −0.03, p = .70, 95% CI [−0.11, 0.07]) (not shown in table). Although the care recipient’s number of chronic conditions was positively associated with role overload (Table 5), caregiver gender did not moderate this association or the link between role overload and negative caregiving relationship quality (not shown in tables). As a consequence, this interaction was not mediated by gender differences pertaining to subjective stressors. Care-Related Family Disagreements As depicted in Table 4, caregivers were more likely to report care-related family disagreements when they perceived higher role overload (Model 3). None of the background characteristics or primary objective stressors were directly associated with family disagreements. There was a significant interaction between care recipients’ dementia status and caregiver gender in Model 2 (B = −3.22, p = .002). Husbands were more likely to report care-related family disagreements when they cared for a partner with dementia (B = 2.77, OR = 15.98, p < .001, 95% CI [4.02, 63.49]); however, this link was not present for wives (B = −0.45, OR = 0.64, p = .49, 95% CI [0.18, 2.28]) (not shown in table). Dementia status was positively linked to emotional caregiving difficulties (Table 5); but caregiver gender did not moderate this association or the link between emotional difficulties and care-related family disagreements (not shown in tables). As such, this interaction was not mediated by gender differences relating to subjective stressors. Post Hoc Tests To test the stability of the findings regarding gender differences, we estimated a full model for each outcome including all predictors and their gender interaction terms. The findings on gender differences from the main analysis remained significant, confirming their stability. Discussion and Implications The present study demonstrates that a number of factors may contribute to secondary role strains among spouses caring for a partner with functional disability. Wives generally reported more primary stressors and secondary role strains than husbands on average. These findings are consistent with prior work showing that, relative to men, women typically face more intense and stressful care situations (Pinquart & Sörenson, 2006; Swinkels et al., 2017). Nonetheless, this study reveals few gender differences in caregiving stress proliferation, which implies that a given level of primary care-related stressors impacts wives and husbands in similar ways. Overall, time-intensive care tasks and role overload seem to be most consequential for secondary role strains among wives and husbands. But husbands appear uniquely susceptible to marital and family discord when their partner has more complex care needs. Secondary strains play a key part in explaining gender differences in spouses’ caregiver burden (Swinkels et al., 2017). Consequently, wives and husbands may be roughly equal in their experience of burden and related adverse health effects when comparable levels of caregiving stress spread to other life areas. The findings therefore yield critical insights for clinicians and policymakers concerned with spousal caregivers’ well-being and their ability to provide quality care. Predictors of Care-Related Valued Activity Restriction In line with research showing that women experience more family barriers to self-care than men (Rosland et al., 2010; Thomeer et al., 2015), wives were more likely than husbands to report care-related valued activity restriction, independent of their background characteristics (e.g., age, educational attainment). But this association was not observed after accounting for primary objective stressors, which suggests that it may be partially explained by wives’ greater exposure to these stressors on average (Pinquart & Sörensen, 2006; Swinkels et al., 2017). Contrary to prediction, caregivers with more education were more likely to report care-related valued activity restriction. This finding may be attributed in part to the type of activities assessed in this study, some of which require financial resources that may limit participation among caregivers with less education (e.g., going out for enjoyment). Of note, however, this link was no longer present after taking primary subjective stressors into account, indicating that education is relatively inconsequential when considering how caregivers appraise their care role. Caregivers were also more likely to report care-related valued activity restriction when they provided more assistance with ADLs (e.g., mobility) and health system interaction tasks (e.g., scheduling appointments). Helping with ADLs requires a substantial time commitment and has been associated with decreased self-care behaviors among caregiving spouses (Gallant & Connell, 1997), which could extend to reduced participation in activities they enjoy. Interacting with health systems similarly involves complicated decision-making (e.g., selecting a drug insurance plan) and coordinating care across multiple providers that can be challenging and time-consuming (Giovannetti et al., 2012). Supporting these points, caregivers’ greater perceptions of role overload partially accounted for the associations between both care tasks and care-related valued activity restriction. Over and above primary objective stressors, older caregivers and those who perceived higher role overload were more likely to report valued activity restriction. Caregivers who feel overloaded may forgo their valued activities to manage other noncare tasks (e.g., housework), and could benefit from respite services along with guidance in how to optimize time away from caregiving to incorporate activities that restore their well-being (Lund, Utz, Caserta, & Wright, 2009). The finding for age was present only after accounting for primary subjective stressors, implying that caregivers’ age-related vulnerabilities may increase their likelihood of valued activity restriction when considering their negative appraisals of the care role. Research has shown that older adults are impaired in their capacity to cope with negative experiences in circumstances they cannot easily avoid such as spousal caregiving (Charles, 2010), which may in turn deplete their energy to engage in valued activities. Predictors of Negative Caregiving Relationship Quality Wives reported more negative caregiving relationship quality than husbands, independent of their background characteristics and primary objective stressors. This finding is consistent with the broader literature demonstrating more negative views of marital quality among wives than husbands in middle and later life (e.g., Umberson & Williams, 2005). Contrary to prediction, caregivers with higher levels of education also perceived more negative caregiving relationship quality. It is plausible that the challenges of spousal caregiving lead to more marital tension for highly educated individuals because they have had less experience in handling adversity than those with lower educational attainment (Seery, 2011). Beyond gender and education, caring for a partner with dementia was linked to greater perceptions of negative caregiving relationship quality. This association was fully mediated by emotional care-related difficulties, indicating that such difficulties may escalate negative aspects of the caregiving relationship in the context of dementia. Diminished marital communication and intimacy as dementia progresses could further play a part in explaining these findings (Quinn et al., 2009). Greater emotional caregiving difficulties and role overload emerged as the sole correlates of higher negative caregiving relationship quality when all predictors were considered. This implies that negative appraisals of the care role may have a particularly detrimental impact on spousal interactions. Alternatively, these links could be bidirectional. Caregivers who are highly distressed, for example, tend to be more critical or hostile toward the care recipient (Smith et al., 2011). In turn, poor marital quality may amplify caregivers’ distress (Baucom et al., 2012; Berg & Upchurch, 2007). Caring for a partner with a higher number of chronic health conditions was associated with more negative caregiving relationship quality for husbands only, even when controlling for primary subjective stressors. Compared with women, men rely more heavily on their partner to satisfy emotional support needs (Antonucci, 2001). Thus, one possibility is that greater medical morbidity renders wives less able to meet these needs, thereby contributing to marital discord. Relatedly, men are often unable or unwilling to provide emotional support to their wife regarding her own health issues (Thomeer et al., 2015), which may intensify marital problems. Predictors of Care-Related Family Disagreements Caregivers were more prone to report care-related family disagreements when they felt higher role overload. In accord with family systems theory (Bowen, 1978), this finding may reflect reciprocal exchanges such that caregivers’ feelings of overload exacerbate disagreements, while disagreements magnify perceptions of being overloaded. The absence of a significant gender difference suggests that caregiving wives and husbands who experience role overload are both vulnerable to potentially harmful family dynamics. For husbands only, caring for a partner with dementia was associated with a higher likelihood of care-related family disagreements. Men are traditionally viewed as less competent in the care role than women (Calasanti, 2010; Thompson, 2002), and so family strife may partly stem from doubts about husbands’ ability to provide care in more complicated caregiving situations. Alternatively, in complex circumstances, husbands may be more likely to serve as a caregiver when there is family discord because other potential helpers are unavailable or unwilling to contribute. Strengths and Limitations Strengths of this study include a nationally representative sample, emphasis on caregiving stress proliferation among spousal caregivers, and the statistical comparison of differences between caregiving wives and husbands. Models accounted for caregivers’ sociodemographics and health conditions, along with multiple primary objective and subjective caregiving stressors. Hence, the present study identifies an array of factors that are independently linked to secondary role strains and could eventually erode each care partner’s well-being. We acknowledge several limitations. First, cross-sectional analyses preclude the determination of causal associations. Second, on average, spouses reported low levels of primary care-related stressors and secondary role strains. Although these levels are in line with other population-based studies demonstrating considerable resilience among caregivers as a whole (Roth et al., 2015), the findings may not generalize to more distressed spousal caregivers. Third, we were limited by the base study measures. The NSOC measure of activity restriction, for instance, may not have captured all valued activities that are impacted by caregiving. Additionally, care-related family disagreements were assessed with one item. We also used binary measures of care-related valued activity restriction and family disagreements due to a lack of variability in both outcomes, which limits understanding of the degree to which these secondary role strains are associated with characteristics of the spousal care situation. Finally, NSOC does not include a pre-caregiving measure of relationship quality, limiting the ability to disentangle whether the findings mirror continued marital patterns. The finding that wives report more negative caregiving relationship quality than husbands in the multivariate analyses, for example, may be partly attributed to fewer benefits of marriage derived by women (Antonucci, 2001; Umberson & Williams, 2005). Fourth, the findings may not translate to specific care contexts (e.g., dementia caregiving). Nevertheless, this study provides the foundation for subsequent work to elucidate knowledge of caregiving stress proliferation among aging adults caring for partners with functional disability. Future Directions The present study brings forth a number of practical considerations that should be explored in future research. First, this study underscores the importance of examining how family relationships both influence and are influenced by the care situation. Prior work has placed a near-exclusive focus on primary caregivers, neglecting an understanding of the family system (Schulz & Eden, 2016). Yet, caregiving emerges from a past and present family narrative that shapes caregivers’ experience of stressors as well as their available resources to combat stress (Zarit et al., in press). Determining strategies for caregiving spouses to leverage their positive social ties (e.g., finding other helpers, sharing emotional support) while minimizing the effects of negative relationships (e.g., marital or family conflict) could be crucial intervention targets. Family meetings to guide the negotiation of care responsibilities and the resolution of disagreements may also attenuate caregiving stress proliferation (Joling et al., 2012). Second, counter to the common view that men’s managerial approach to providing care shields them from negative care-related outcomes (Calasanti, 2010; Thompson, 2002), this study indicates that husbands may be highly susceptible to caregiving stress proliferation under certain conditions. In line with previous studies (e.g., DePasquale et al., 2016; Swinkels et al., 2017), these findings indicate that clinicians and policymakers should recognize the variability in men’s caregiving experiences. Increased attention to the needs and concerns of caregiving husbands in more complex care situations may be especially warranted. Lastly, this study illuminates the need to consider long-term consequences of spouses’ caregiving stress proliferation. Previous research has demonstrated, for example, that reduced scheduled activities and negative marital interactions are each linked to spousal caregivers’ lower positive affect and higher negative affect across time (Savla et al., 2011), raising the question of whether these secondary strains have lasting detrimental implications for well-being. Balancing competing roles and maintaining self-care activities have been pinpointed as unmet support needs among spousal caregivers (Badr et al., 2016), and so the proliferation of caregiving stress could present a risk of detrimental outcomes for both care partners. Conclusions In summary, this study highlights the multifaceted ways that background characteristics and primary care-related stressors are linked to secondary role strains among spouses caring for a partner with functional disability. Notably, the proliferation of caregiving stress may develop through both common and distinct pathways for wives and husbands, identifying those who could require support in managing their care role along with noncare roles and activities. The findings call for a more comprehensive approach to caregiver interventions and policies that consider how caregiving may influence and be influenced by other important life domains. Funding This work was supported by the National Institute of Mental Health (T32 MH073553-11 to S. J. Bartels) and the National Institute on Aging (F31 AG050385 to N. DePasquale). The National Health and Aging Trends Study (NHATS) and National Study of Caregiving (NSOC) are sponsored by the National Institute on Aging (NIA U01AG032947) through a cooperative agreement with the Johns Hopkins Bloomberg School of Public Health. Conflict of Interest None reported. References Aneshensel, C.S., Pearlin, L.I., Mullan, J.T., Zarit, S.H., & Whitlatch, C.J. ( 1995). Profiles in caregiving. The unexpected career . San Diego, CA: Academic Press. Antonucci, T.C. ( 2001). Social relations: An examination of social networks, social support and sense of control. In J. E. Birren, K. W. Schaie, & R. P. Abeles (Eds.), Handbook of the psychology of aging (pp. 427–453) . San Diego, CA: Academic Press. Badr, H., Herbert, K., Reckson, B., Rainey, H., Sallam, A., & Gupta, V. ( 2016). Unmet needs and relationship challenges of head and neck cancer patients and their spouses. Journal of Psychosocial Oncology , 34, 336– 346. doi: 10.1080/07347332.2016.1195901 Google Scholar CrossRef Search ADS PubMed  Baron, R.M., & Kenny, D.A. ( 1986). The moderator-mediator variable distinction in social psychological research: Conceptual, strategic, and statistical considerations. Journal of Personality and Social Psychology , 51, 1173– 1182. doi: 10.1037/0022-3514.51.6.1173 Google Scholar CrossRef Search ADS PubMed  Baucom, D.H., Porter, L.S., Kirby, J.S., & Hudepohl, J. ( 2012). Couple-based interventions for medical problems. Behavior Therapy , 43, 61– 76. doi: 10.1016/j.beth.2011.01.008. Google Scholar CrossRef Search ADS PubMed  Berg, C.A., & Upchurch, R. ( 2007). A developmental-contextual model of couples coping with chronic illness across the adult life span. Psychological Bulletin , 133, 920– 954. doi: 10.1037/0033-2909.133.6.920 Google Scholar CrossRef Search ADS PubMed  Bowen, M. ( 1978). Family therapy in clinical practice . New York: Jason Aronson. Calasanti, T. ( 2010). Gender relations and applied research on aging. The Gerontologist , 50, 720– 734. doi: 10.1093/geront/gnq085 Google Scholar CrossRef Search ADS PubMed  Chappell, N.L., Dujela, C., & Smith, A. ( 2015). Caregiver well-being: Intersections of relationship and gender. Research on Aging , 37, 623– 645. doi: 10.1177/0164027514549258 Google Scholar CrossRef Search ADS PubMed  Charles, S.T. ( 2010). Strength and vulnerability integration (SAVI): A model of emotional well-being across adulthood. Psychological Bulletin , 136, 1068– 1091. doi: 10.1037/a0021232 Google Scholar CrossRef Search ADS PubMed  Chattillion, E.A., Ceglowski, J., Roepke, S.K., von Känel, R., Losada, A., Mills, P.J., … Mausbach, B.T. ( 2013). Pleasant events, activity restriction, and blood pressure in dementia caregivers. Health Psychology , 32, 793– 801. doi: 10.1037/a0029412 Google Scholar CrossRef Search ADS PubMed  DePasquale, N., Polenick, C.A., Hinde, J., Bray, J., Zarit, S.H., Moen, P., … Almeida, D.M. ( 2016). Health behavior among men with multiple family roles: The moderating effects of perceived partner relationship quality. American Journal of Men’s Health . Advance online publication. doi: 10.1177/1557988316660088 Eisinga, R., Grotenhuis, M., & Pelzer, B. ( 2013). The reliability of a two-item scale: Pearson, Cronbach, or Spearman-Brown? International Journal of Public Health , 58, 637– 642. doi: 10.1007/s00038-012-0416-3 Google Scholar CrossRef Search ADS PubMed  Gallant, M. P., & Connell, C. M. ( 1997). Predictors of decreased self-care among spouse caregivers of older adults with dementing illnesses. Journal of Aging and Health , 9, 373– 395. doi: 10.1177/089826439700900306 Google Scholar CrossRef Search ADS PubMed  Gaugler, J.E., Linder, J., Given, C.W., Kataria, R., Tucker, G., & Regine, W.F. ( 2008). The proliferation of primary cancer caregiving stress to secondary stress. Cancer Nursing , 31, 116– 123. doi: 10.1097/01.NCC.0000305700.05250.9d. Google Scholar CrossRef Search ADS PubMed  Gaugler, J.E., Mendiondo, M., Smith, C.D., & Schmitt, F.A. ( 2003). Secondary dementia caregiving and its consequences. American Journal of Alzheimer’s Disease and Other Dementia , 18, 300– 308. doi: 10.1177/153331750301800505 Google Scholar CrossRef Search ADS   Giovannetti, E.R., Wolff, J.L., Xue, Q., Weiss, C.O., Leff, B., Boult, C., … Boyd, C. ( 2012). Difficulty assisting with health care tasks among caregivers of multimorbid older adults. Journal of General Internal Medicine , 27, 37– 44. doi: 10.1007/s11606-011-1831-5. Google Scholar CrossRef Search ADS PubMed  Gwyther, L.P., & Matchar, B.G. ( 2015). The Duke employee elder care consultation: Meeting employees where they are. Generations , 39, 105– 108. Holt-Lunstad, J., Smith, T.B., Baker, M., Harris, T., & Stephenson, D. ( 2015). Loneliness and social isolation as risk factors for mortality: A meta-analytic review. Perspectives on Psychological Science , 10, 227– 37. doi: 10.1177/1745691614568352 Google Scholar CrossRef Search ADS PubMed  Hsu, T., Loscalzo, M., Ramani, R., Forman, S., Popplewell, L., Clark, K., … Smith, D. ( 2014). Factors associated with high burden in caregivers of older adults with cancer. Cancer , 120, 2927– 2935. doi: 10.1002/cncr.28765 Google Scholar CrossRef Search ADS PubMed  Joling, K.J., van Marwijk, H.W.J., Smit, F., van der Horst, H.E., Scheltens, P., van de Ven, P. M., … & van Hout, H.P.J. ( 2012). Does a family meeting intervention prevent depression and anxiety in family caregivers of dementia patients? A randomized trial. PLoS ONE , 7, e30936. doi: 10.1371/journal.pone.0030936 Google Scholar CrossRef Search ADS PubMed  Kasper, J.D., Freedman, V.A., & Spillman, B. ( 2013). National Study of Caregiving user guide . Baltimore, MD: Johns Hopkins University School of Public Health. Lund, D.A., Utz, R.. Caserta, M.S., & Wright, S.D. ( 2009). Examining what caregivers do during respite time to make respite more effective. Journal of Applied Gerontology , 28, 109– 131. doi: 10.1177/0733464808323448 Google Scholar CrossRef Search ADS   Lyons, K.S., Zarit, S.H., Sayer, A.G., & Whitlatch, C.J. ( 2002). Caregiving as a dyadic process: Perspectives from caregiver and care receiver. Journal of Gerontology, Series B: Psychological Sciences and Social Sciences , 57, 195– 204. doi: 10.1093/geronb/57.3.P195 Google Scholar CrossRef Search ADS   Mausbach, B.T., Chattillion, E.A., Moore, R.C., Roepke, S.K., Depp, C.A., & Roesch, S. ( 2011). Activity restriction and depression in medical patients and their caregivers: A meta-analysis. Clinical Psychology Review , 31, 900– 908. doi: 10.1016/j.cpr.2011.04.004 Google Scholar CrossRef Search ADS PubMed  Moore, R.C., Harmell, A.L., Chattillion, E., Ancoli-Isreal, S., Grant, I., & Mausbach, B.T. ( 2011). PEAR model and sleep outcomes in dementia caregivers: Influence of activity restriction and pleasant events on sleep disturbances. International Psychogeriatrics , 23, 1462– 1469. doi: 10.1017/S1041610211000512 Google Scholar CrossRef Search ADS PubMed  Muller, D., Judd, C.M., & Yzerbyt, V.Y. ( 2005). When moderation is mediated and mediation is moderated. Journal of Personal and Social Psychology , 89, 852– 863. doi: 10.1037/0022-3514.89.6.852 Google Scholar CrossRef Search ADS   Navaie-Waliser, M., Feldman P.H., Gould, D.A., Levine, C., Kuerbis, A.N. & Donelan, K. ( 2002). When the caregiver needs care: The plight of vulnerable caregivers. Journal of Public Health , 92, 409– 413. doi: 10.2105/AJPH.92.3.409 Oldenkamp, M., Hagedoorn, M., Slaets, J., Stolk, R., Wittek, R., & Smidt, N. ( 2016). Subjective burden among spousal and adult-child informal caregivers of older adults: Results from a longitudinal cohort study. BMC Geriatrics , 16, 208. doi: 10.1186/s12877-016-0387-y Google Scholar CrossRef Search ADS PubMed  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, 583– 594. doi: 10.1093/geront/30.5.583 Google Scholar CrossRef Search ADS PubMed  Piette, J.D., Rosland, A.M., Silveira, M., Kabeto, M., & Langa, K.M. ( 2010). The case for involving adult children outside the household in the self-management support of older adults with chronic illnesses. Chronic Illness , 6, 34– 45. doi: 10.1177/1742395309347804. Google Scholar CrossRef Search ADS PubMed  Pinquart, M., & Sörenson, S. ( 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, 33– 45. doi: 10.1093/geronb/61.1.P33 Google Scholar CrossRef Search ADS   Pinquart, M., & Sörenson, S. ( 2007). Correlates of physical health of informal caregivers: A meta-analysis. The Journals of Gerontology, Series B: Psychological Sciences and Social Sciences , 62, P126– P137. doi: 10.1093/geronb/62.2.P126 Google Scholar CrossRef Search ADS   Pinquart, M., & Sörenson, S. ( 2011). Spouses, adult children, and children in-law as caregivers of older adults: A meta-analytic comparison. Psychology and Aging , 26, 1– 14. doi: 10.1037/a0021863 Google Scholar CrossRef Search ADS PubMed  Quinn, C., Claire, L., & Woods, B. ( 2009). The impact of the quality of relationship on the experiences and wellbeing of caregivers of people with dementia: A systematic review. Aging and Mental Health , 13, 143– 154. doi: 10.1080/13607860802459799. Google Scholar CrossRef Search ADS PubMed  Reinhard, S.C., Levine, C., & Samis, S. ( 2014). Family caregivers providing complex chronic care to their spouses . Washington, DC: AARP and United Hospital Fund. Rosland, A., Heisler, M., Choi, H., Silveira, M.J., & Piette, J.D. ( 2010). Family influences on self-management among functionally independent adults with diabetes or heart failure: Do family members hinder as much as they help? Chronic Illness , 6, 22– 33. doi: 10.1177/1742395309354608 Google Scholar CrossRef Search ADS PubMed  Roth, D.L., Fredman, L., & Haley, W.E. ( 2015). Informal caregiving and its impact on health: A reappraisal from population-based studies. The Gerontologist , 55, 309– 319. doi: 10.1093/geront/gnu177 Google Scholar CrossRef Search ADS PubMed  Rudman, D.L., Hebert, D., & Reid, D. ( 2006). Living in a restricted occupational world: The occupational experiences of stroke survivors who are wheelchair users and their caregivers. Canadian Journal of Occupational Therapy , 73, 141– 152. doi: 10.2182/cjot.05.0014 Google Scholar CrossRef Search ADS   Savla, J., Roberto, K.A., Blieszner, R., Cox, M., & Gwazdauskas, F. ( 2011). Effects of daily stressors on the psychological and biological well-being of spouses of persons with mild cognitive impairment. The Journal of Gerontology, Series B: Psychological Sciences and Social Sciences , 66, 653– 664. doi: 10.1093/geronb/gbr041 Google Scholar CrossRef Search ADS   Scharlach, A., Li, W., & Dalvi, T.B. ( 2006). Family conflict as a mediator of caregiver strain. Family Relations , 55, 625– 635. doi: 10.1111/j.1741-3729.2006.00431.x Google Scholar CrossRef Search ADS   Schulz, R., & Eden, J. ( 2016). Families caring for an aging America . Washington, DC: The National Academies Press. Retrieved from http://www.nap.edu/23606 Google Scholar CrossRef Search ADS   Seery, M. D. ( 2011). Resilience: A silver lining to experiencing adverse life events? Current Directions in Psychological Science , 20, 390– 394. doi: 10.1177/0963721411424740 Google Scholar CrossRef Search ADS   Semple, S.J. ( 1992). Conflict in Alzheimer’s caregiving families: Its dimensions and consequences. The Gerontologist , 32, 648– 655. doi: 10.1093/geront/32.5.648 Google Scholar CrossRef Search ADS PubMed  Smith, G.R., Williamson, G.M., Miller, L.S., & Schulz, R. ( 2011). Depression and quality of informal care: A longitudinal investigation of caregiving stressors. Psychology and Aging , 26, 584– 591. doi: 10.1037/a0022263 Google Scholar CrossRef Search ADS PubMed  Swinkels, J., van Tilburg, T., Verbakel, E., & Broese van Groenou, M. ( 2017). Explaining the gender gap in the caregiving burden of partner caregivers. The Journals of Gerontology, Series B: Psychological Sciences and Social Sciences . Advance online publication. doi: 10.1093/geronb/gbx036 Thomeer, M.B., Reczek, C., & Umberson, D. ( 2015). Gendered emotion work around physical health problems in mid- and later-life marriages. Journal of Aging Studies , 32, 12– 22. doi: 10.1016/j.jaging.2014.12.001 Google Scholar CrossRef Search ADS PubMed  Thompson, E.H. ( 2002). What’s unique about men’s caregiving? Men as caregivers: Theory, research, and service implications. (pp. 20–47) . New York, NY: Springer Publishing. Umberson, D., & Williams, K. ( 2005). Marital quality, health, and aging: Gender equity? The Journals of Gerontology. Series B: Psychological Sciences and Social Sciences , 60 (SPEC), 109– 113. doi: 10.1093/geronb/60.Special_Issue_2.S109 Google Scholar CrossRef Search ADS   Valtorta, N.K., Kanaan, M., Gilbody, S., Ronzi, S., & Hanratty, B. ( 2016). Loneliness and social isolation as risk factors for coronary heart disease and stroke: Systematic review and meta-analysis of longitudinal observational studies. Heart , 102, 1009– 1016. doi: 10.1136/heartjnl-2015–308790 Google Scholar CrossRef Search ADS PubMed  Zarit, S.H., Polenick, C.A, DePasquale, N., Liu, Y., & Bangerter, L.R. ( in press). Family support and caregiving in middle and late life. In B. H. Fiese (Ed.), APA Handbook of Contemporary Family Psychology (Vol. 2) . Washington, DC: American Psychological Association. © The Author(s) 2018. Published by Oxford University Press on behalf of The Gerontological Society of America. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com.

Journal

The GerontologistOxford University Press

Published: Jan 8, 2018

There are no references for this article.

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


DeepDyve is your
personal research library

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

Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.

All for just $49/month

Explore the DeepDyve Library

Search

Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly

Organize

Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.

Access

Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.

Your journals are on DeepDyve

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

All the latest content is available, no embargo periods.

See the journals in your area

DeepDyve

Freelancer

DeepDyve

Pro

Price

FREE

$49/month
$360/year

Save searches from
Google Scholar,
PubMed

Create lists to
organize your research

Export lists, citations

Read DeepDyve articles

Abstract access only

Unlimited access to over
18 million full-text articles

Print

20 pages / month

PDF Discount

20% off