Marital Quality and Well-Being Among Older Adults: A Typology of Supportive,Aversive,Indifferent,and Ambivalent Marriages

Abstract

A large body of work has linked marital quality to the health and well-being of older adults, but there is a lack of agreement on how to best measure dimensions of marital quality. Drawing on a stress-process life course perspective, we construct a typology of marriage type that captures the synergistic relationship between positive and negative marital qualities and health. Using data from Wave 1 (2005/2006) and Wave 2 (2010/2011) of the NSHAP survey from the United States, we examine the association between supportive, aversive, ambivalent, and indifferent marriages for older adults that remained married over the study period on multiple indicators of well-being (depression, happiness, and self-rated health; N = 769 males and 461 females). Results suggest that older adults in aversive marriages reported lower happiness (men and women) and physical health (men). There was less evidence that those in ambivalent and indifferent marriages reported worse well-being.

Keywords

health well-being marital quality gender

Despite overall signs indicating the relative advantages of marriage for health and well-being in later life (Liu & Waite, 2014; Thomas et al., 2017; Waite & Gallagher, 2000), a long research tradition has cautioned that marriage is not always beneficial for health. Indeed, while a happy marriage characterized by positive support and mutual respect between partners appears to be health protective, a poor marital relationship may be more harmful for health than not being married (Carr et al., 2014; Proulx et al., 2007), especially for older adults (Birmingham et al., 2015; Holt-Lunstad et al., 2008; Liu & Waite, 2014; Umberson et al., 2006).

While extensive and informative in nature, we argue that the current literature on marital quality and health is limited in three important ways. First, previous studies have offered narrow definitions of marital quality, tending to treat the effects of positive and negative aspects of marriage as a “competition” to determine which has the stronger influence on well-being (Fincham & Linfield, 1997; Fingerman et al., 2006; Uchino et al., 2001). In the current study, we expand on the conceptualization of marital quality and following recent research by Hseih and Hawkley (2018), construct a typology of marriage type that offers a unique look into the synergistic relationship between positive and negative marital qualities and health. Second, the majority of past research on marital quality and well-being in later life has been cross-sectional in nature (Galinsky & Waite, 2014; Hseih & Hawkley, 2018; though see Liu & Waite, 2014, and Umberson et al., 2006, for notable exceptions). We adopt a stress-process life course perspective and rely on two waves of panel data from the National Social Life, Health, and Aging Project (NSHAP) to test our hypotheses. The use of longitudinal data allows us to examine whether marital quality has associations with health and well-being that persist over time, net of baseline measures of health and well-being, a first of many steps in documenting a causal relationship.

Finally, previous research has suggested that the relative effects of positive and negative marital qualities on well-being may vary across domains of health and well-being. While most studies of later life couples reveal that marital support and marital strain are positively or inversely related to physical health, respectively (see Thomas et al., 2017, for a review), the relationship between marital quality and mental well-being tends to be more nuanced. For instance, there is some evidence that marital support is more strongly associated with positive outcomes of mental well-being (e.g., life satisfaction) compared with negative ones (e.g., depression; Antonucci et al., 2001; Lee & Szinovacz, 2016; Newsom et al., 2003). Newsom et al. (2003) suggest that people may receive an immediate boost to positive aspects of well-being from positive relational interactions, where marital strain may exact a longer-lasting impact on negative aspects of well-being, such as depression. Therefore, as a third contribution of our study, we examine several indicators of health and well-being (depression, happiness, and self-rated health), recognizing that the use of a single indicator may risk underestimating the effect of marital quality on well-being (e.g., Aneshensel, 2005) or likewise fail to capture important distinctions concerning the relative longitudinal effects of positive and negative aspects of marriage.

Literature Review

A Stress-Process Life Course Perspective for Conceptualizing the Relationship Between Marital Quality and Well-Being

The current study integrates insights from the stress process model into a life course framework to understand the relationship between marital quality and well-being. One of the paradigmatic principles of the life course is that of linked lives, which refers to interdependence in shared relationships (Elder et al., 2003). Prior work within the life course tradition has noted that the spousal relationship may take on increasing importance in later life as older adults downsize their social networks, due in part to transitions out of the paid labor force or adult children leaving the home (Hsieh & Hawkley, 2018; Lee & Szinovacz, 2016; Thomas et al., 2017).

The concept of linked lives is central to the idea that support and strain in relationships can affect health. Indeed, stressors (strain) and social support are core components of the stress process theory (Pearlin, 1989). Social support from spousal relationships can provide critical resources for health, while strain in such relationships (e.g., due to conflict) become sources of stress that can undermine health (Pearlin et al., 1981). On one hand, the receipt of support from a spouse is associated with a greater sense of meaning in life and a vast reservoir of resources, including better health behaviors (Reczek et al., 2014; Umberson, 1987) that may benefit both physical and mental health (Kawachi & Berkman, 2001). Because the spousal relationship becomes more salient in later life, these processes may take on added importance. Indeed, social support from a spouse/partner was more strongly related to lower depressive symptoms for older adults compared to their midlife counterparts (Thomas et al., 2017).

On the other hand, any stress caused by marital conflict can produce detrimental physiological consequences such as increased heart rate and impaired function of the immune system (Kiecolt-Glaser & Newton, 2001) and can undermine mental health by serving as a constant source of tension (Pearlin, 1989). Marital strain also exacts a larger toll on health for older adults compared to their younger counterparts (Umberson et al., 2006).

As we expand on in the next section, the current study examines the synergistic effect of support (positive quality) and strain (negative quality) among married older adults. We argue that this provides a more nuanced approach within the stress-process life course perspective, as it recognizes that a person may experience an overlap of positive and negative interaction within the same relationship.

Measuring Marital Quality: A Typology Approach

Marital quality is a multidimensional construct and several operationalizations have been proposed in the current literature (Robles et al., 2014). While not all encompassing, prior research has generally posited two basic models of measuring marital quality. The first approach takes the two relational dimensions of stress process theory, positive (support) and negative (strain) aspects of marital quality, and includes them as separate constructs (Liu & Waite, 2014; Umberson et al., 2006). The goal is to “isolate” the independent effects of each quality of marriage, net of the other, so they can be compared (Gilligan et al., 2015). Evidence from clinical studies and survey research both show that negative interactions within a marriage often leave a stronger imprint on people’s memory than positive ones (De Vogli et al., 2007; Lee & Szinovacz, 2016), and thus may outweigh the effects of positive interactions on health (Brooks & Dunkel Schetter, 2011). Yet, this first approach overlooks the fact that positive and negative aspect of marriages do not only have independent effects but may also have joint effects on health that each dimension alone cannot capture (Fingerman et al., 2006; Holt-Lunstad & Clark, 2014; Uchino et al., 2012).

Recognizing the limits to the first approach, a second approach has been to assess the joint effects of positive and negative features of marriage. Here, positive and negative spousal qualities are “contextualized such that the impact of positivity depends on the co-occurrence of negativity and vice versa” (Hsieh & Hawkley, 2018, p. 1321). This approach implies testing a statistical interaction to determine, for instance, how positive marital quality may condition (in this case, buffer), the harmful effects of negative ones. However, this approach is also not without limitations; as Hseih and Hawkley (2018) note, interaction terms only tell us the general (average) effect of positive (negative) aspects of marriage in the presence of negative (or positive) aspects. However, each member of a couple experiences a specific combination of positive and negative aspects of marriage, which statistical interaction terms cannot adequately convey.

A third approach to examining marital quality quickly gaining steam in social gerontology is to construct a typology of marital quality, which we adopt here. The typology approach involves using cut-off points to categorize individuals into groups based on scores on both the support and strain dimensions of marital quality, and typically sort couples into four different groups (Hsieh & Hawkley, 2018). Anchoring the two ends of the spectrum of marital quality are supportive marriages (high on positive aspects, and low on negative) and aversive marriages (low on positive, high on negative). However, Hseih and Hawkley (2018) also present two categories that are located somewhere in the middle of these opposing ends of marital quality. Ambivalent marriages are characterized by both high positive and high negative aspects (Fincham & Linfield, 1997; Uchino et al., 2012), while indifferent marriages capture relationships that have both low positive and low negative aspects (Fincham & Linfield, 1997; Uchino et al., 2012). As we outline below, this approach offers a more expansive view of the overlap of both dimensions of relationships within the stress process model, support and strain.

Categories of Marital Quality and Health

Based on previous work examining marital quality and health, it is well-documented that those in supportive marriages report better health than those in aversive marriages (Hsieh & Hawkley, 2018; Windsor & Butterworth, 2010). However, it remains unclear how the two middle categories fit into this pattern. Indifferent marriages are generally characterized by little engagement and communication in the marriage, where two people ignore and neglect each other, thus minimizing confrontation but lacking in support (Hseih & Hawkley, 2018). Though no prior study has considered the association of indifferent marriage with well-being, past studies have found that couples in indifferent marriages tend to report lower marital satisfaction and greater loneliness than individuals in supportive marriages, but higher satisfaction than those in aversive marriages (Fincham & Linfield, 1997; Hseih & Hawkley, 2018).

Ambivalent marriages present a more complicated case in potential linkages to health. Ambivalence is generally defined as “contradictions in relationships…that cannot be reconciled” (Lüscher & Pillemer, 1998, p. 416). People in ambivalent relationships feel torn about their feelings toward their spouse. Though not typically marked by abuse or cruelty, common sources of strain within this partnership could include partners getting on each other’s nerves or criticizing the behaviors of the other (Umberson, 1992). Compared to other social relationship types, both positive and negative exchanges with a spouse are the most common and intense (Chen & Feeley, 2014), and thus could hold tight linkages with health. Feelings of ambivalence tend to be relatively common in marital relationships and typically reflect past relationship difficulties or negative support exchanges (Birditt et al., 2009; Pillemer & Suitor, 2002).

Previous studies based on both experimental data and cross-sectional survey designs have tended to show that people in ambivalent relationships report worse cardiovascular health (Holt-Lunstad & Clark, 2014; Uchino et al., 2001), lower self-rated health (Fingerman et al., 2006), and poorer mental health (Newsom et al., 2003) relative to those in more positive quality marriages. This pattern occurs for two reasons. First, ambivalent relationships tend to be inherently stressful (Holt-Lunstad et al., 2007) and less predictable (Uchino et al., 2001). The mere presence of such a tie puts individuals on edge and stimulates greater cardiovascular activity (Holt-Lunstad et al., 2007). Indeed, people primed in an experiment to think about ambivalent relationships exhibit greater heart rate activity than those primed to think about purely problematic partners (Carlisle et al., 2012). Second, past studies have suggested that ambivalent relationships interfere with social support processes. These relationships tend to be characterized by less emotional support and more criticism (Reblin et al., 2010). One common source of criticism between spouses surrounds health behaviors, such as dietary and smoking practices (Umberson, 1992). Attempts to influence the health behaviors of one’s spouse, however, may be perceived as annoying or stressful to the other party (i.e., strain). In this situation, what may begin with good intentions to improve partner health could lead to higher levels of strain in the marriage (Birditt et al., 2009). This could reduce the likelihood of viewing that partner’s efforts at social support in other domains to be genuine or out of concern for one’s own health and cause further strain, which may in turn be associated with lower well-being. Based on the evidence reviewed above, we present our first two study hypotheses.

Hypothesis 1a: Older adults in aversive or indifferent marriages will report lower health and well-being compared to those in supportive marriage.

Hypothesis 1b: Older adults in ambivalent marriages will report lower health and well-being compared to individuals in supportive or aversive marriages.

Since little previous research has been conducted on the health of individuals in indifferent marriages, we thus leave it as an open question whether people in indifferent marriages will report greater or lower well-being than people in ambivalent marriages.

Gender Differences in the Relationship Between Marital Quality and Health?

Gender is a central focus of research on marriage and well-being and is a key determinant of experiences over the life course (Liu & Waite, 2014; Zhang & Hayward, 2006). The general consensus in the literature to date is that marital quality has stronger effects on the health of women compared to men, while marital status (being married) has stronger effects on men than women due to enhanced emotional support and health behavior regulation performed by one’s spouse (Umberson & Kroeger, 2016).

We draw from the gender-as-relational approach (see Thomeer et al., 2015; Umberson et al., 2018) to consider how the combination of marital strain and support may differentially impact the health and well-being of older married men and women. According to this perspective, men and women have idealized views of how each spouse should act (Peralta, 2007). This includes cultural norms which assign women the dual responsibilities of being more emotionally supportive and watching over the health of others, and position men as more independent risk-takers who are in need of health monitoring (Ganong & Larson, 2011).

These insights encapsulated in the gender-as-relational perspective are borne out in empirical work within the stress process model. Within a traditional marriage that typifies many later life partnerships (Davis & Greenstein, 2009), women are more likely to be providers of emotional support to their husbands (Umberson et al., 1996), which in turn promotes men’s well-being. Men receive emotional support primarily from their spouses, whereas women rely more heavily on their friends, relatives, and children (Gurung et al., 2003). Moreover, being married is more likely to be accompanied by social control of health behaviors for men compared to women, with wives often serving as the primary person that influences their husband’s health (Schafer, 2013; Umberson, 1992). This body of work suggests that:

Hypothesis 2a: Older males in ambivalent, indifferent, or aversive marriages will report lower health and well-being compared to those in supportive marriage.

On the other hand, however, some studies have suggested that marital strain may harm the health of women more than men (Kiecolt-Glaser & Newton, 2001). Older women, on the whole, tend to report a lower quality of marriage than their male counterparts, likely because they shoulder a greater burden of housekeeping duties and family care (Boerner et al., 2014; Umberson & Williams, 2005). What is more, women’s lower relative status in marriage (and society) compared to men, coupled with their stronger interpersonal disposition, could make women more vulnerable to stress in the marital partnership (Kiecolt-Glaser & Newton, 2001; Wanic & Kulik, 2011). Strain, as a form of relationship threat, may be more distressing for women who lack status and resources. Moreover, even within a well-functioning marriage, women tend to bear the brunt of emotional regulation within the relationship (Bloch et al., 2014), which can produce stress given that it is an arduous task filled with ambiguous cues (Loscocco & Walzer, 2013). Women might also be disappointed by a low level of support from their partner, even if strain was not present (i.e., in an indifferent marriage).

Hypothesis 2b: Older females in aversive, indifferent, or ambivalent marriages will report lower health and well-being compared to those in supportive marriage.

To help ensure that any observed association between marital quality type and well-being is robust to alternative explanations, our analyses include three types of covariates (all measured at baseline) that are related to both marital quality and physical and mental well-being: (1) socio-demographic covariates (age, race-ethnicity, education, number of children), (2) frequency of partnered sex and (3) baseline levels of each dimension of well-being.

Research Design

Sample

Data come from the first two waves of National Social Life, Health, and Aging Project (NSHAP). NSHAP is a nationally representative probability sample of community-dwelling older adults between the ages of ages 57 to 85. The first wave of NSHAP launched in 2005–2006 and included a sample of 3,005 adults ages 57 to 85 that were interviewed in their homes Five years later, in 2010–2011, 2,261 original respondents from Wave 1 were re-interviewed. In this study, we restrict our analytic sample to the 1,250 respondents who remained married to the same person at both Waves 1 and 2, and who were successfully interviewed at both waves.

Measures

Dependent variables

Depression

We used an 11-item short form of the CES-D scale of depressive symptoms scale at both Wave 1 and Wave 2. Sample items included: “I did not feel like eating—my appetite was poor” and “I felt that everything I did was an effort.” Response options for each item were: (1) “rarely or none of the time” to (4) “most of the time.” Scores were averaged, with a higher score indicating more depressive symptoms (α = 0.75 at Wave 1, and α = 0.77 at Wave 2).

Happiness

Respondents were asked to “rate their general happiness” at each wave. Response were coded on an ordinal scale from (1) “unhappy usually” to (5) “extremely happy.”

Self-Rated Health

Respondents were asked to “rate their general health” at each wave. Response were coded on an ordinal scale ranging from (1) “poor health” to (5) “excellent health.”

Primary independent variables

Marital Quality

We categorized marital quality based on responses to questions about spousal support and strain at Wave 1 of NSHAP (see Hsieh & Hawkley, 2018). Spousal support was assessed with two items: (a) “How often can you open up to your spouse if you need to talk about your worries?” and (b) “How often can you rely on your spouse for help if you have a problem?” (α = 0.59). Spousal strain was also composed of two items. (c) “How often does your spouse make too many demands on you?” and (d) “How often does your spouse criticize you?” (α = 0.62). For both strain and support items, response categories were: (1) “hardly ever” or “never”; (2) “some of the time”; (3) “often” In each case, higher scores represent higher spousal support or strain. There was a negative correlation between the spousal support and strain scales (polychoric ρ = −0.28). Online Appendix 1 shows the zero-order correlations among all items comprising the support and strain scales.

Taking the scores on these two dimensions of spousal support and strain, we created a four-category variable to denote marital quality. We defined high (vs. low) support as responding “often” or “hardly ever” to both support items, respectively, and high (vs. low) strain as responding “often” to both strain items. In our sample, most of the respondents (62% of men and 68% of women) reported that their spouse never or hardly ever makes too many demands on them or criticizes them (50% for men and 71% for women). Thus, respondents who perceived spousal demands or criticisms at least some of the time were defined as high (rather than low) in marital strain (see Hsieh & Hawkley, 2018, for a similar approach). We then generated a matrix of marital quality types using the dichotomous support and strain variables: supportive (high support and low strain), ambivalent (high support and high strain), indifferent (low support and low strain), and aversive (low support and high strain). In our analyses, we present coefficients for all possible pairwise comparisons in our four-category variable of marital quality.

Control Variables

We adjust for several demographic covariates that could confound the relationship between marital quality and health. Age was measured in years, and race was coded as a binary variable of white versus non-white respondents. The number of living children that respondent had was measured as the sum of living sons and living daughters (including stepchildren). Education was coded into a series of categories: (1) high school or less (reference group); (2) high school or equivalent; (3) vocational certification/some college/ associate degree; (4) bachelor’s degree or above.

Analyses also feature an adjustment for the sexual frequency of respondents at Wave 2, as it is a determinant of well-being in later life (DeLamater, 2012), and also a predictor of marital quality (DeLamater & Moorman, 2007). Respondents were asked, “how often they had sex with their spouse during the last 12 months.” Response categories ranged from (1) “once a month or less” to (5) “once a day or more.” In our sample, 29% of respondents reported that they did not have sex in the last 12 months, and they were coded to have a score of 0.

Supplemental analyses also adjusted for household assets. Main results were unchanged with its inclusion. We ultimately did not retain household assets in our final models due to high amounts of missing data (> 25%) and because it was highly redundant with our measure of education. Additional analyses also considered whether the respondent had been in a previous marriage (yes/no) and the duration of their current marriage (in years), measured at Wave 2 of the study (average = 31.03 years). Results remained unchanged with the addition of these latter two variables, so we ultimately excluded them from our final analyses.

We include a lagged version of each dependent variable, measured at Wave 1, to help ensure unbiased coefficients by adjusting for the autocorrelation between measures of health at both time points. The inclusion of baseline levels of well-being also safeguards against the possibility that any observed associations between marital quality and well-being are due to differential starting levels of well-being across categories of marital quality. Descriptive statistics for all study variables, stratified by gender, are found in Table 1. Tests for differences in proportions (χ²) or means (t statistic) between men and women are presented in Table 1.

Table 1.

Sample Descriptive Statistics, National Social Life, Health, and Aging Project (NSHAP), Stratified by Gender.

Variables	Total Sample (N = 1,250)			Male (N = 769)		Female (N = 481)		χ²/t	df	Diff Test (p)
	Range	Mean/Prop	SD	Mean/Prop	SD	Mean/Prop	SD	χ²/t	df	Diff Test (p)
Depression, Wave 2	1, 4	1.89	0.32	1.88	0.32	1.90	0.33	−1.56	1,248	n.s.
Happiness, Wave 2	1, 5	3.71	0.84	3.74	0.85	3.67	0.83	1.27	1,248	n.s.
Self-Rated Health, Wave 2	1, 5	3.29	1.06	3.27	1.05	3.22	1.08	−0.59	1,248	n.s.
Marital Quality								17.65	3	**
Supportive	0, 1	0.56		0.53		0.60
Ambivalent	0, 1	0.18		0.22		0.12
Indifferent	0, 1	0.15		0.14		0.16
Aversive	0, 1	0.11		0.11		0.11
Age	57, 85	66.63	6.91	66.96	7.17	66.09	6.44	2.05	1,248	**
Race								0.02	1	n.s.
White		0.76		0.76		0.77
Non-white		0.24		0.24		0.23
Education								15.98	3	***
< High school		0.16		0.17		0.13
High school or equivalent		0.23		0.21		0.27
Some college		0.32		0.30		0.37
Bachelor’s degree or higher		0.29		0.32		0.23
Number of children	0, 22	3.15	1.92	3.23	1.94	3.03	1.89	1.66	1,248	n.s.
Frequency of sex at Wave 2	0, 5	1.14	1.20	1.20	1.19	1.05	1.22	1.99	1,248
Depression, Wave 1	1, 4	1.89	0.32	1.87	0.30	1.92	0.33	−2.27	1,248	n.s.
Happiness, Wave 1	1, 5	3.78	0.81	3.83	0.81	3.71	0.79	2.45	1,248	*
Self-rated Health, Wave 1	1, 5	3.44	1.04	3.46	1.04	3.42	1.04	0.78	1,248	n.s.

Note. n.s. = not significant.

*p < .05. **p < .01. ***p < .001.

Analytic Plan

Following previous longitudinal research on marital quality and health with the NSHAP data, we conducted a gender stratified analysis (Liu & Waite, 2014), with a total of 769 males and 481 females. We employed Wald tests to formally test for the equality of gender coefficients for categories of marital quality and well-being (Agresti & Finlay, 2009). This test uses seemingly unrelated regression techniques to fit separate models for men and women and calculates a χ² statistic to test whether the coefficients differ across gender. Finally, since less than 5% of our sample had missing data on our study variables, listwise deletion was used to handle missing data.

Results

Table 1 displays descriptive statistics for our sample. Across our three indicators of well-being, depression was stable between waves, as respondents had an average of 1.89 on depression at both waves. Happiness declined slightly from 3.78 at Wave 1 to 3.71 at Wave 2. Self-rated health also declined from 3.44 at Wave 1 to 3.29 at Wave 2, likely attributable to the physical decline that accompanies the aging processes. It is also notable that at Wave 2, there were no significant differences in depression, happiness, or self-rated health between married men and women.

Looking at our measure of marital typology, 60% of women fit into the supportive marriage category, while only 53% of men find themselves in a supportive marriage. Men were almost twice as likely to be in an ambivalent marriage (22%) compared to women (12%). Online Appendix 2 shows how other demographic characteristics are associated with the assignment of marital categories. Briefly, we note that there were no differences in marital category by age. Relative to Whites, non-Whites were more likely to be in ambivalent marriages, and less likely to be in supportive marriages. Finally, those with a bachelor’s degree or higher were less likely to be in an indifferent or ambivalent marriage, and more likely to be in a supportive marriage.

Does Marital Category Influence Health and Well-Being? Results From Regression Analysis

Table 2 displays results from a series of regression analyses for each of our measures of well-being for the male sample, and Table 3 for an identical set of models pertaining to the female sample. Figure 1 displays the focal association between our typology of marital quality and our three measures of well-being for all of the six models that we tested (left panel for males, right panel for females). These visualizations allow for accessible comparisons of the association between all four marital types to each other, and we reference relevant comparisons below. Average marginal effects are shown for depression, and the predicted probabilities of being in the highest category of happiness (“extremely happy “) and self-rated health (“excellent”) are displayed, with 95% confidence intervals shown in every instance.

Table 2.

Coefficients From OLS Regressions and Ordinal Logistic Regression of Health Outcomes on Marital Quality, Male Sample.

Variables	Model 1	Model 2	Model 3
	Depression	Happiness	Health
Marital Quality
Ambivalent vs. Supportive	.01	−0.30	−0.13
	(.03)	(.20)	(.19)
Indifferent vs. Supportive	−.02	−0.52*	−0.55**
	(.03)	(.23)	(.22)
Aversive vs. Supportive	.06	−0.90**	−0.77**
	(.04)	(.27)	(.26)
Indifferent vs. Ambivalent	.01	−0.22	−0.68**
	(.04)	(.26)	(.25)
Aversive vs. Ambivalent	.06	−0.60*	−0.89**
	(.04)	(.30)	(.29)
Aversive vs. Indifferent	.05	−0.39	−0.21
	(.05)	(.31)	(.30)
Control Variables
Age	.00	−0.00	−0.02
	(.00)	(.01)	(.01)
Non-white ^a	−.00	−0.29	−0.20
	(.03)	(.19)	(.18)
High school or equivalent ^b	−.06	0.09	0.51*
	(.04)	(.25)	(.24)
Some College ^b	−.14***	−0.19	0.67**
	(.03)	(.24)	(.23)
Bachelor’s Degree or higher ^b	−.15***	0.17	0.93***
	(.03)	(.24)	(.24)
Frequency of Sex, Wave 2
Once a month or less ^c	−.02	0.15	0.47*
	(.03)	(.21)	(.20)
2–3 Times a month ^c	−.07	0.32	0.53*
	(.03)	(.22)	(.22)
Once or twice a week ^c	−.10*	0.74**	0.69**
	(.04)	(.25)	(.25)
3–6 Times a week ^c	−.02	0.95	0.39
	(.08)	(.53)	(.51)
Once a day or more ^c	−.12	0.87	2.60**
	(.13)	(.85)	(.89)
Number of Living Children	.00	0.03	0.13**
	(.04)	(.04)	(.04)
Depression, Wave 1	.38***
	(.04)
Happiness, Wave 1		1.29***
		(.10)
Self-Rated Health, Wave 1			1.38***
			(.09)

Note. N = 769.

^a Compared to White.

^b Compared to less than high school.

^c Compared to not at all.

* p < .05. ** p < .01. *** p < .001.

Table 3.

Coefficients From OLS Regressions and Ordinal Logistic Regression of Health Outcomes on Marital Quality, Female Sample.

Variables	Model 1	Model 2	Model 3
	Depression	Happiness	Health
Marital Quality
Ambivalent vs. Supportive	−.04	−0.41	−0.27
	(.05)	(.32)	(0.30)
Indifferent vs. Supportive	−.03	−0.47	0.05
	(.04)	(.27)	(0.27)
Aversive vs. Supportive	.06	−1.11**	0.27
	(.05)	(.33)	(0.30)
Indifferent vs. Ambivalent	.01	−0.05	0.32
	(.06)	(.38)	(0.36)
Aversive vs. Ambivalent	.10	−0.69	0.54
	(.06)	(.42)	(0.38)
Aversive vs. Indifferent	.09	−0.64	0.22
	(.06)	(.38)	(0.36)
Control Variables
Age	.00	−0.01	−0.02
	(.00)	(.02)	(0.02)
Non-white ^a	−.03	−0.21	−0.40
	(.04)	(.24)	(0.24)
High school or equivalent ^b	.11*	−0.09	−0.39
	(.05)	(.35)	(0.34)
Some College ^b	.09	0.23	0.09
	(.05)	(.32)	(0.32)
Bachelor’s Degree or higher ^b	−.02	0.64	0.34
	(.03)	(.35)	(0.36)
Frequency of Sex, Wave 2
Once a month or less ^c	−.05	−0.07	0.19
	(.04)	(.26)	(0.24)
2 to 3 Times a month ^c	−.09*	0.21	0.85**
	(.04)	(.28)	(0.28)
Once or twice a week ^c	−.06*	0.50	0.56
	(.05)	(.33)	(0.31)
3–6 Times a week ^c	−.15	0.49	0.37
	(.10)	(.61)	(0.58)
Once a day or more ^c	.12	−1.00	−0.16
	(.15)	(.93)	(0.86)
Number of Living Children	.00	0.08	−0.04
	(.04)	(.05)	(0.05)
Depression, Wave 1	.40***
	(.04)
Happiness, Wave 1		1.09***
		(.14)
Self-Rated Health, Wave 1			1.34***
			(1.12)

Note. N = 481.

^a Compared to White.

^b Compared to less than high school.

^c Compared to not at all.

*p < .05. **p < .01. ***p < .001.

Figure 1.

Marginal effects of marital quality on depression, happiness, and self-rated health, male (N = 769) and female sample (N = 461). Note. Estimates are derived by Models 1–3 of Table 2 (men) and Table (3) women for each indicator of health and well-being. All other covariates are held at their respective means.

Beginning first with depression (Model 1), there were no significant differences between any two sets of marital types, for either men or women. Moving to the results for happiness in Model 2, men in indifferent as well as aversive marriages report significantly lower happiness and self-rated health (b = −0.52, p < .05 and b = −.0.90, p < .01, respectively) compared with men in supportive marriages. Men in aversive marriages also reported lower happiness relative to men in ambivalent marriages (b = −0.60, p < .05). For women, only those in aversive marriages reported lower happiness relative to those in supportive marriages (b = −1.11, p < .01). Wald tests of equality of coefficients revealed no statistically significant differences between the coefficients for older men and women for happiness for any of these pairwise comparisons. Even where a significant coefficient for both men and women occurred (aversive versus supportive), the effect was similar for both groups. This is shown clearly in Figure 1, in the second row: both men and women in aversive marriages were almost half as likely to fall into the top category of happiness (predicted probability = 0.10 for men, and 0.07 for women) compared to those in supportive marriages (prob. = 0.20 for men, and 0.18 for women).

Finally, Model 3 presents analyses for self-rated physical health. Men in indifferent (b = −0.55, p < .01) and aversive (b = −0.77, p < .01) marriages reported lower self-rated health compared to men in supportive marriages. What is more, men in indifferent marriages had lower self-rated health relative to those in ambivalent marriages (b = −0.68, p < .01), as did men in aversive marriages (b = −0.89, p < .01). The results for self-rated health paint a much different picture for women. Indeed, no pairwise comparison revealed statistically significant differences among groups of marital quality for women.

Wald tests for equality of coefficients between gendered samples showed that there were gender differences in the indifferent versus supportive (χ² = 3.97, p = 0.04) and aversive versus supportive (χ² = 6.91, p = 0.009) marriage comparisons. As displayed in the final row of graphs in Figure 1 (self-rated health), men in indifferent marriages were 5% less likely to be in excellent health than those in supportive marriages (prob. = 0.09 versus 0.14, respectively). Men in aversive marriages were almost half as likely to be in this top category of health compared to those in supportive marriages (prob. = 0.08 versus 0.14).

In sum, we find some support for Hypothesis 1a and Hypothesis 2b: older men and women in aversive marriages, and older men in indifferent marriages, generally report lower well-being than those in supportive marriages. We do not find support for Hypothesis 1b, however. Older men and women in ambivalent marriages did not report lower well-being than those in supportive marriages, nor higher well-being than those in indifferent or aversive marriages. With respect to gender differences, we found evidence consistent with Hypothesis 2a: men in aversive marriages, and in some cases, indifferent marriages, had lower health and well-being than their female counterparts. Women experienced only lower happiness in aversive compared with supportive marriages.

Supplemental analyses

As robustness checks on our main findings, we conducted two additional analyses. First, we tested two-way interactions between support and strain rather than using our categorical approach of marital quality. Second, we tested two-way interactions between gender and marital category in the full sample instead of gender stratifying the sample. Results from these analyses are presented in Online Appendix 3 and 4, respectively. In each case, the results are consistent with the interpretations we offered above.

Discussion

This study adopted a stress-process life course perspective to understand how specific types of marital qualities are related with older men’s and women’s health and well-being longitudinally. In alignment with the linked lives principle of the life course, which suggests that spousal relationship takes on increasing importance with age, our first key finding was that older adults in aversive marriages tended to have worse health over time compared to those in supportive marriages. This result was documented for both happiness and self-rated health for men, and happiness only for women. This is consistent with results from numerous other studies showing that low quality marriages undermine the health and well-being of older adults, while positive quality is health enhancing (Hsieh & Hawkley, 2018; Liu & Waite, 2014; Umberson et al., 2006). However, our marital typology approach clarifies conventional approaches to studying the effects of relationship quality within the stress process model (Pearlin, 1989; Pearlin et al., 1981) because it suggests that negative quality (in the absence of positive quality) and positive quality (in the absence of negative quality) bear associations with well-being. This is a key theme that reverberates through our results: positive and negative marital qualities synergistically influence the health and well-being of married individuals in later life.

A second and more novel finding of the current study was that older men and women in ambivalent marriages did not report lower well-being relative to those in supportive marriages. At first glance, this result appears to be at odds with past research showing health deficits for those in ambivalent marriages compared to supportive marriages (Holt-Lunstad & Clark, 2014; Hold-Lunstad et al., 2007; Uchino et al., 2001). Past research has tended to rely on non-representative cross-sectional survey (Uchino et al., 2001, 2012) or experimental data (Holt-Lunstad & Clark, 2014), primarily of younger individuals. However, research specific to ambivalence in later life relationships that document negative consequences for well-being studies have relied on cross-sectional data and examined relational dyads between older parents and their children (Fingerman et al., 2006; Gilligan et al., 2015) or kin and non-kin relationships more generally (Rook et al., 2011). Our study utilized nationally representative, longitudinal data specific to the marriage at later stages of the life course, which has the advantage of adjusting for baseline levels of health and moving us closer to (but not fully establishing) a causal understanding of the relationship between marital quality and health.

Why might ambivalent relationships have consistently null longitudinal associations with health across several indicators of well-being, for both older men and women? Older adults may be motivated to adapt to any negative exchanges within a long-term marriage, at least 5 years or more, in our study, with an average duration of 31.21 years (standard deviation = 19.43). Such familiarity with one’s partner may motivate individuals to engage in relationship maintenance strategies to momentarily ease tension or strain, such as comprising, accepting partial blame, or engaging in conciliatory behaviors such as forgiveness (e.g., Rook et al., 2011). Ambivalent relationships to partners may thus have a positive component that boosts positive affect, at least temporarily (e.g., Ingersoll-Dayton et al., 1997). In relationships that have stood the test of time, relationship strain may involve criticism or nagging on the part of one spouse (e.g., regarding poor health behaviors), which could be perceived as nothing more than minor inconveniences. Moreover, it stands to reason that even in marriages characterized by greater strain, small or fleeting windows of positivity could offset some of the more negative ones. Finally, compared to previous work that has considered both positive and negative dimensions of marital quality, the NSHAP measures of spousal support and strain were each only two items and had relatively low reliability coefficients. This could potentially introduce measurement error compared to past studies which use a 6-item scale with extensive response options (Fincham & Linfield, 1997).

Lastly, our results also suggested important gender differences for the impact of marital quality on physical health. Men in indifferent and aversive marriages reported lower self-rated health relative to those in supportive marriages. For the latter comparison, marriage increases the likelihood that one will have access to social support much more for men than women (Umberson et al., 1996), so a wholly unsupportive spouse may be more detrimental for the health of older men. Men in supportive marriages, on the other hand, may develop a tacit understanding of their partner’s perspective and limit the need for potentially antagonistic conversations with their spouse. These men might also tend to value communal goals in their relationships that are pursued via mutual cooperation with their wives. Finally, according to the gender-as-relational perspective, structural systems of gender dictate norms that women should serve as agents of social control over the health behavior of men (Rook et al., 2011; Umberson, 1992). Men who lack a supportive partner even in the context of low marital conflict may report worse physical health as a result of not having someone actively watching over their health behavior.

Older adults in indifferent marriages perceive few positive aspects of the relationship (Hseih & Hawkley, 2018), and this appears detrimental to the physical health of men. While some research has suggested that marital quality tends to matter more for the health of women (Thomeer et al., 2015), there is also some evidence that women report lower marital satisfaction than men (Umberson & Williams, 2005). One speculative interpretation is that older women are more conditioned to an indifferent spouse who shows little positivity or negativity toward them. Similar to what we argued above, men tend to be more heavily reliant on their female partners for taking care of their health needs, including seeking preventative medical care (e.g., Schafer, 2013); thus, their health may disproportionately suffer from having a disengaged partner.

Taken together, three main conclusions can be drawn with respect to gender differences in the effect of marital quality on well-being. First, that differences between men and women only emerged for self-rated physical health suggests the importance of considering multiple indicators of well-being (Aneshensel, 2005; Newsom et al., 2003; Robles et al., 2014), and is consistent with previous research which documents more robust associations between marital quality and physical health compared to mental well-being (Thomas et al., 2017). Second, the general consensus to date in the literature is that marital quality has stronger effects on the health of women, but marital status (just being married) has stronger effects for men (Robles et al., 2014; Umberson & Kroeger, 2016). Our longitudinal study adds additional complexity to this conclusion by suggesting that gender differences in the effects of marital quality and health depend on the timing in the life course that they are studied (later life), as well as the outcomes under consideration. Finally, other than for happiness, our results suggested little evidence that marital quality mattered for older women. Some previous evidence within the gender-as-relational perspective suggests there could be health costs of providing for men’s health for women (Umberson et al., 2015), but our study does not support this. While marital strain may exact a short-term toll on the health of women (Proulx et al., 2007), it is possible that older women receive health protection from their larger social networks and more diverse sources of social support.

Limitations and Applications

We acknowledge several limitations of the current study. First, our results may have been influenced by selection bias, as only older adults that remained in the same marriage over the years between Waves 1 and 2 of the NSHAP study were included in our analytic sample. Supplemental analyses, however, reveal that those in aversive marriages were not more likely to drop out of the analytic sample between waves. Relative to those in supportive marriages, older individuals in ambivalent marriages were less likely to drop out of, while those in indifferent marriages were more likely to fall out of the sample between waves. Given this latter pattern, the estimates we reported here may be conservative ones, especially since older adults in indifferent marriages may have had worse health at baseline and follow-up as a result of lower marital quality. Moreover, couples who remained together over the study period despite a high prevalence of negative qualities in their marriage may have established tried-and-true coping mechanisms to overcome these rough patches.

Second, the scales we derived our typology of marital strain and support from had low α reliability coefficients. This is likely because only an abbreviated number of items from the original spousal support/strain scale developed by Schuster and colleagues (1990) were included in the NSHAP survey. However, previous research with the NSHAP data has tended to favor categorical over continuous approaches to measuring marital quality to circumvent measurement error (see Choi & Ha, 2011). Similarly, we did not assess change in marital quality over the study period. A promising direction for future research is to track change in marital type over time (Warner & Adams, 2016); it is possible, for instance, that ambivalent may, with time, work through the aspects in their marriage that are causing strain, thus coming to resemble more closely the marital characteristics of those in supportive marriages. Moreover, it is likely that the effects of supportive, or aversive marriages on health may accumulate over time, further widening the gap between these groups.

Finally, future research could integrate measures of partner health and mental health by using dyadic data to test the influence of marital categories on subsequent health. Extant research suggests that poor physical, cognitive, and mental health on the part of one or both spouses is associated with higher marital conflict and lower levels of engagement with one’s partner, likely due to higher stress caused by the debilitating nature of long-term illness (Coyne et al., 1987; Wong & Hseih, 2019). In ancillary analyses, we conducted actor-partner interdependence models using dyadic data only from Wave 2 of NSHAP (N = 913 married couples), as it was not available at Wave 1. Adjusting for three facets of partner’s health, self-rated physical health, depressive symptoms, and cognitive health (see Shega et al., 2014), the main pattern of results we reported was not altered. An additional exploratory analysis showed that lower physical health on the part of husbands was associated with a higher likelihood of wives being in an aversive (but not indifferent) marriage; health did not affect marital category assignment for men. At least in these cross-sectional analyses, the health of one’s partner may influence the probability of falling into each category of marital type but does not ultimately change the effect of marital type on health and well-being. Longitudinal research designs in a dyadic framework are needed to more fully explore the effect of partner’s health on selection into marital category and the health and well-being of their spouse, and thus help to resolve these issues of causal ordering.

Conclusion

Our results show that simply assessing one dimension of marital quality without the other may obscure the natural heterogeneity that exist in the marital lives of older couples. While we hope this approach guides future scholarly work on marital quality and health, we also see value in the findings of our study for family therapists and medical professionals who counsel older adults experiencing marital problems. The results of the current study suggest that some benefit may be gained by helping each member of the couple gain an appreciation of how the infusion of positivity into a relationship can prevent the health of both partners from spiraling downward, even amid negative relational aspects and challenges. On this front, it is clear that more research is needed to identify how later life marital types are linked with different aspects of interpersonal coping strategies among male and female partners, and how they may ultimately affect health and well-being.

Supplemental Material

SUPPLEMENTARY_MATERIAL - Marital Quality and Well-Being Among Older Adults: A Typology of Supportive, Aversive, Indifferent, and Ambivalent Marriages

SUPPLEMENTARY_MATERIAL for Marital Quality and Well-Being Among Older Adults: A Typology of Supportive, Aversive, Indifferent, and Ambivalent Marriages by Yingling Liu and Laura Upenieks in Research on Aging

Footnotes

Declaration of Conflicting Interests

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

Funding

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

ORCID iD

Yingling Liu

Supplemental Material

Supplemental material for this article is available online.

Author Biographies

Yingling Liu is a PhD candidate of sociology and the research analyst of Center for Community Research and Development at Baylor University. Her research focuses on health ineuqality, aging and well-being, and applied sociology.

Laura Upenieks, PhD, is an assisstant professor of sociology at Baylor University. Her current research focuses on intergrating life course models of health and social network analysis to better understand the relationship between religion and well-being, and on how life course inequalities shape health outcomes during the COVID-19 pandemic.

References

Agresti

Finlay

(2009). Statistical methods for the social sciences: With SPSS from A to Z: A brief step-by-step manual. Allyn & Bacon.

Aneshensel

C. S.

(2005). Research in mental health: Social etiology versus social consequences. Journal of Health and Social Behavior, 46(3), 221–228.

Antonucci

T. C.

Lansford

J. E.

Akiyama

(2001). Impact of positive and negative aspects of marital relationships and friendships on well-being of older adults. Applied Developmental Science, 5(2), 68–75.

Birditt

K. S.

Miller

L. M.

Fingerman

K. L.

Lefkowitz

E. S.

(2009). Tensions in the parent and adult child relationship: Links to solidarity and ambivalence. Psychology and Aging, 24(2), 287–295.

Birmingham

W. C.

Uchino

B. N.

Smith

T. W.

Light

K. C.

Butner

(2015). It’s complicated: Marital ambivalence on ambulatory blood pressure and daily interpersonal functioning. Annals of Behavioral Medicine, 49(5), 743–753.

Bloch

Haase

C. M.

Levenson

R. W.

(2014). Emotion regulation predicts marital satisfaction: More than a wives’ tale. Emotion, 14(1), 130–144.

Boerner

Jopp

D. S.

Carr

Sosinsky

Kim

S.-K.

(2014). “His” and “Her” marriage? The role of positive and negative marital characteristics in global marital satisfaction among older adults. The Journals of Gerontology Series B: Psychological Sciences and Social Sciences, 69(4), 579–589.

Brooks

K. P.

Dunkel Schetter

(2011). Social negativity and health: Conceptual and measurement issues. Social & Personality Psychology Compass, 5, 904–918.

Carlisle

Uchino

B. N.

Sanbonmatsu

D. M.

Smith

T. W.

Cribbet

M. R.

Birmingham

Vaughn

A. A.

(2012). Subliminal activation of social ties moderates cardiovascular reactivity during acute stress. Health Psychology, 31(2), 217–225.

10.

Carr

Freedman

V. A.

Cornman

J. C.

Schwarz

(2014). Happy marriage, happy life? Marital quality and subjective well-being in later life. Journal of Marriage and Family, 76(5), 930–948.

11.

Chen

Feeley

T. H.

(2014). Social support, social strain, loneliness, and well-being among older adults: An analysis of the health and retirement study. Journal of Social and Personal Relationships, 31(2), 141–161.

12.

Choi

N. G.

J. H.

(2011). Relationship between spouse/partner support and depressive symptoms in older adults: Gender difference. Aging & Mental Health, 15(3), 307–317.

13.

Coyne

J. C.

Kessler

R. C.

Tal

Turnbull

Wortman

C. B.

Greden

J. F.

(1987). Living with a depressed person. Journal of Consulting and Clinical Psychology, 55(3), 347–352.

14.

Davis

S. N.

Greenstein

T. N.

(2009). Gender ideology: Components, predictors, and consequences. Annual Review of Sociology, 35, 87–105.

15.

DeLamater

(2012). Sexual expression in later life: A review and synthesis. Journal of Sex Research, 49(2–3), 125–141.

16.

DeLamater

Moorman

S. M.

(2007). Sexual behavior in later life. Journal of Aging and Health, 19(6), 921–945.

17.

De Vogli

Chandola

Marmot

M. G.

(2007). Negative aspects of close relationships and heart disease. Archives of Internal Medicine, 167(18), 1951–1957.

18.

Elder

G. H.

Jr Johnson

M. K.

Crosnoe

(2003). The emergence and development of the life course. In Mortimer

J. T.

Shanahan

M. J.

(Eds.), Handbook of the life course (pp. 3–19). Plenum Press.

19.

Fincham

F. D.

Linfield

K. J.

(1997). A new look at marital quality: Can spouses feel positive and negative about their marriage. Journal of Family Psychology, 11(4), 489–502.

20.

Fingerman

K. L.

Chen

P.-C.

Hay

Cichy

K. E.

Lefkowitz

E. S.

(2006). Ambivalent reactions in the parent and offspring relationship. The Journals of Gerontology: Series B, 61(3), P152–P160.

21.

Galinsky

A. M.

Waite

L. J.

(2014). Sexual activity and psychological health as mediators of the relationship between physical health and marital quality. The Journals of Gerontology Series B: Psychological Sciences and Social Sciences, 69(3), 482–492.

22.

Ganong

Larson

(2011). Intimacy and belonging: The association between sexual activity and depression among older adults. Society and Mental Health, 1(3), 153–172.

23.

Gilligan

Suitor

J. J.

Feld

Pillemer

(2015). Do positive feelings hurt? Disaggregating positive and negative components of intergenerational ambivalence. Journal of Marriage and Family, 77(1), 261–276.

24.

Gurung

R. A.

Taylor

S. E.

Seeman

T. E.

(2003). Accounting for changes in social support among married older adults: Insights from the MacArthur studies of successful aging. Psychology and Aging, 18(3), 487–496.

25.

Holt-Lunstad

Birmingham

Jones

B. Q.

(2008). Is there something unique about marriage? The relative impact of marital status, relationship quality, and network social support on ambulatory blood pressure and mental health. Annals of Behavioral Medicine, 35(2), 239–244.

26.

Holt-Lunstad

Clark

B. D.

(2014). Social stressors and cardiovascular response: Influence of ambivalent relationships and behavioral ambivalence. International Journal of Psychophysiology, 93(3), 381–389.

27.

Holt-Lunstad

Uchino

B. N.

Smith

T. W.

Hicks

(2007). On the importance of relationship quality: The impact of ambivalence in friendships on cardiovascular functioning. Annals of Behavioral Medicine, 33(3), 278–290.

28.

Hsieh

Hawkley

(2018). Loneliness in the older adult marriage: Associations with dyadic aversion, indifference, and ambivalence. Journal of Social and Personal Relationships, 35(10), 1319–1339.

29.

Ingersoll-Dayton

Morgan

Antonucci

(1997). The effects of positive and negative social exchanges on aging adults. The Journals of Gerontology Series B: Psychological Sciences and Social Sciences, 52(4), S190–S199.

30.

Kawachi

Berkman

L. F.

(2001). Social ties and mental health. Journal of Urban Health, 78(3), 458–467.

31.

Kiecolt-Glaser

J. K.

Newton

T. L.

(2001). Marriage and health: His and hers. Psychological Bulletin, 127(4), 472–503.

32.

Lee

H. J.

Szinovacz

M. E.

(2016). Positive, negative, and ambivalent interactions with family and friends: Associations with well-being. Journal of Marriage and Family, 78(3), 660–679.

33.

Liu

Waite

(2014). Bad marriage, broken heart? Age and gender differences in the link between marital quality and cardiovascular risks among older adults. Journal of Health and Social Behavior, 55(4), 403–423.

34.

Loscocco

Walzer

(2013). Gender and the culture of heterosexual marriage in the United States. Journal of Family Theory & Review, 5(1), 1–14.

35.

Lüscher

Pillemer

(1998). Intergenerational ambivalence: A new approach to the study of parent-child relations in later life. Journal of Marriage and the Family, 413–425.

36.

Newsom

J. T.

Nishishiba

Morgan

D. L.

Rook

K. S.

(2003). The relative importance of three domains of positive and negative social exchanges: A longitudinal model with comparable measures. Psychology and Aging, 18(4), 746–754.

37.

Pearlin

L. I.

(1989). The sociological study of stress. Journal of Health and Social Behavior, 30, 241–256.

38.

Pearlin

L. I.

Menaghan

E. G.

Lieberman

M. A.

Mullan

J. T.

(1981). The stress process. Journal of Health and Social behavior, 22(4), 337–356.

39.

Peralta

R. L.

(2007). College alcohol use and the embodiment of hegemonic masculinity among European American men. Sex Roles, 56(11–12), 741–756.

40.

Pillemer

Suitor

J. J.

(2002). Explaining mothers’ ambivalence toward their adult children. Journal of Marriage and Family, 64(3), 602–613.

41.

Proulx

C. M.

Helms

H. M.

Buehler

(2007). Marital quality and personal well-being: A meta-analysis. Journal of Marriage and Family, 69(3), 576–593.

42.

Reblin

Uchino

B. N.

Smith

T. W.

(2010). Provider and recipient factors that may moderate the effectiveness of received support: Examining the effects of relationship quality and expectations for support on behavioral and cardiovascular reactions. Journal of Behavioral Medicine, 33(6), 423–431.

43.

Reczek

Liu

Spiker

(2014). A population-based study of alcohol use in same-sex and different-sex unions. Journal of Marriage and Family, 76(3), 557–572.

44.

Robles

T. F.

Slatcher

R. B.

Trombello

J. M.

McGinn

M. M.

(2014). Marital quality and health: A meta-analytic review. Psychological Bulletin, 140(1), 140.

45.

Rook

August

Sorkin

(2011). Social network functions and health. In Contrada

Baum

(Eds.), The handbook of stress science: Biology, psychology, and health (pp. 123–135). Springer.

46.

Schafer

M. H.

(2013). Discussion networks, physician visits, and non-conventional medicine: Probing the relational correlates of health care utilization. Social Science & Medicine, 87, 176–184.

47.

Schuster

T. L.

Kessler

R. C.

Aseltine

R. H.

(1990). Supportive interactions, negative interactions, and depressed mood. American Journal of Community Psychology, 18, 423–438.

48.

Shega

J. W.

Sunkara

P. D.

Kotwal

Kern

D. W.

Henning

S. L.

McClintock

M. K.

Schumm

Waite

L. J.

Dale

(2014). Measuring cognition: The Chicago cognitive function measure in the National Social Life, Health and Aging Project, Wave 2. Journals of Gerontology Series B: Psychological Sciences and Social Sciences, 69(Suppl_2), S166–S176.

49.

Thomas

P. A.

Liu

Umberson

(2017). Family relationships and well-being. Innovation in Aging, 1(3), igx025.

50.

Thomeer

M. B.

Reczek

Umberson

(2015). Gendered emotion work around physical health problems in mid-and later-life marriages. Journal of Aging Studies, 32, 12–22.

51.

Uchino

B. N.

Cawthon

R. M.

Smith

T. W.

Light

K. C.

McKenzie

Carlisle

Gunn

Birmingham

Bowen

(2012). Social relationships and health: Is feeling positive, negative, or both (ambivalent) about your social ties related to telomeres? Health Psychology: Official Journal of the Division of Health Psychology, American Psychological Association, 31(6), 789–796.

52.

Uchino

B. N.

Holt-Lunstad

Uno

Flinders

J. B.

(2001). Heterogeneity in the social networks of young and older adults: Prediction of mental health and cardiovascular reactivity during acute stress. Journal of Behavioral Medicine; New York, 24(4), 361–382.

53.

Umberson

(1987). Family status and health behaviors: Social control as a dimension of social integration. Journal of Health and Social Behavior, 28(3), 306–319.

54.

Umberson

(1992). Gender, marital status and the social control of health behavior. Social Science & Medicine, 34(8), 907–917.

55.

Umberson

Chen

M. D.

House

J. S.

Hopkins

Slaten

(1996). The effect of social relationships on psychological well-being: Are men and women really so different? American Sociological Review, 61(5), 837–857.

56.

Umberson

Donnelly

Pollitt

A. M.

(2018). Marriage, social control, and health behavior: A dyadic analysis of same-sex and different-sex couples. Journal of Health and Social Behavior, 59(3), 429–446.

57.

Umberson

Kroeger

R. A.

(2016). Gender, marriage, and health for same-sex and different-sex couples: The future keeps arriving. In McHale

S. M.

King

Van Hook

Booth

(Eds.), Gender and couple relationships (pp. 189–213). Springer International Publishing.

58.

Umberson

Thomeer

M. B.

Lodge

A. C.

(2015). Intimacy and emotion work in lesbian, gay, and heterosexual relationships. Journal of Marriage and Family, 77(2), 542–556.

59.

Umberson

Williams

(2005). Marital quality, health, and aging: Gender equity? The Journals of Gerontology: Series B, 60(Special_Issue_2), S109–S113.

60.

Umberson

Williams

Powers

D. A.

Liu

Needham

(2006). You make me sick: Marital quality and health over the life course*. Journal of Health and Social Behavior, 47(1), 1–16.

61.

Waite

L. J.

Gallagher

(2000). The case for marriage. Doubleday.

62.

Wanic

Kulik

(2011). Toward an understanding of gender differences in the impact of marital conflict on health. Sex Roles, 65(5–6), 297–312.

63.

Warner

D. F.

Adams

S. A.

(2016). Physical disability and increased loneliness among married older adults: The role of changing social relations. Society and Mental Health, 6(2), 106–128.

64.

Windsor

T. D.

Butterworth

(2010). Supportive, aversive, ambivalent, and indifferent partner evaluations in midlife and young-old adulthood. Journals of Gerontology Series B: Psychological Sciences and Social Sciences, 65(3), 287–295.

65.

Wong

J. S.

Hsieh

(2019). Functional status, cognition, and social relationships in dyadic perspective. The Journals of Gerontology: Series B, 74(4), 703–714.

66.

Zhang

Hayward

M. D.

(2006). Gender, the marital life course, and cardiovascular disease in late midlife. Journal of Marriage and Family, 68(3), 639–657.

Supplementary Material

Please find the following supplemental material available below.

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

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

0.00 MB

0.10 MB