Skip to main content
NCBI home page
The Journals of Gerontology Series B: Psychological Sciences and Social Sciences logoLink to The Journals of Gerontology Series B: Psychological Sciences and Social Sciences
. 2015 Mar 12;71(1):165–176. doi: 10.1093/geronb/gbv014

Marital Quality and Cognitive Limitations in Late Life

Minle Xu 1,✉, Patricia A Thomas 2, Debra Umberson 1
PMCID: PMC4841010  PMID: 25765315

Abstract

Objectives.

Identifying factors associated with cognitive limitations among older adults has become a major public health objective. Given the importance of marital relationships for older adults’ health, this study examines the association between marital quality and change in cognitive limitations in late life, directionality of the relationship between marital quality and cognitive limitations, and potential gender differences in these associations.

Method.

Latent growth curve models were used to estimate the association of marital quality with change in cognitive limitations among older adults and the direction of the association between marital quality and cognitive limitations using 4 waves of the Americans’ Changing Lives survey (N = 841).

Results.

Results indicate that more frequent negative (but not positive) marital experiences are associated with a slower increase in cognitive limitations over time, and the direction of this association does not operate in the reverse (i.e., cognitive limitations did not lead to change in marital quality over time). The association between negative marital experiences and cognitive limitations is similar for men and women.

Discussion.

The discussion highlights possible explanations for the apparent protective effect of negative marital experiences for older adults’ cognitive health over time, regardless of gender.

Key Words: Cognitive limitations, Gender, Marital quality, Older adults.

Cognitive limitations, which refer to impairment in one or more domains of cognitive function (e.g., attention, memory, and executive function), range from mild age-associated cognitive decline to severe cognitive impairment (Stott, 2006). Age is a strong predictor of cognitive limitations, with decline in cognition becoming common after age 60 (Atti et al., 2010). Individuals diagnosed with severe cognitive limitations experience increased risk for disability, mortality, and elevated health care costs (Plassman et al., 2008). Families with cognitively impaired members are often faced with significant caregiving burden (Fisher et al., 2011). These wide-ranging adverse consequences of cognitive limitations, along with population aging, urgently call for research to identify factors that may contribute to or provide protection from cognitive decline among older adults.

The link between social relationships and cognitive limitations in later life has received substantial attention. Larger social networks seem to be protective against increases in cognitive limitations (Zunzunegui, Alvarado, Del Ser, & Otero, 2003). In contrast, social isolation and loneliness are associated with increased cognitive limitations irrespective of education, physical and mental health status, and health behavior (Shankar, Hamer, McMunn, & Steptoe, 2013). However, past studies have predominantly focused on the size of social networks and/or the frequency of social engagement and do not consider how the quality of any particular social relationship might influence cognitive limitations. Yet a significant literature establishes that the quality of relationships, particularly marital relationships, is especially important in shaping health outcomes (Robles, Slatcher, Trombello, & McGinn, 2014). For older people who are married, their spouse is typically their primary social connection; the quality of this relationship may then be particularly salient to health in aging populations. Marital quality has been linked to changes in physical health status over time, and this association is stronger at older ages (Umberson, Williams, Powers, Liu, & Needham, 2006). Marital quality may also be associated with cognitive limitations, in part, by influencing psychological distress levels, exposure to stress, and health habits that are known to contribute to cognitive decline over time (Lee et al., 2010; Wilson et al., 2007).

The reverse may also be true, however, with cognitive limitations leading to change in marital quality. Prior studies suggest that cognitive limitations are inversely associated with marital quality (Laganá, Spellman, Wakefield, & Oliver, 2011). However, most of this evidence is based on data from clinical populations, and it is unclear whether cognitive limitations significantly influence change in marital quality over time in general population-based samples. Past research also points to the possibility of gender differences in the association between marital quality and cognitive limitations. Marital quality seems to be more important to the physical functioning of women than men (Kiecolt-Glaser & Newton, 2001). For example, prior studies suggest that marital strain may be more detrimental to the immune functioning (Kiecolt-Glaser et al., 1993) and inflammatory responses (Whisman & Sbarra, 2012) of women than men.

The present study extends work on social relationships and health in late life and, more specifically, work on marital quality and cognitive status by addressing the following questions: (a) Is marital quality (both positive and negative aspects of marital quality) associated with change in cognitive limitations among those aged 60 and older? (b) Is the direction of the association between marital quality and cognitive limitations driven primarily from marital quality to change in cognitive limitations or vice versa? (c) Does the association between marital quality and cognitive limitations differ for men and women? We examine these questions with data from four waves of the nationally representative Americans’ Changing Lives (ACL) panel survey. Although various instruments for measuring cognitive limitations have been developed for clinical settings, the ACL includes a measure that is validated for use in community-based samples (McDowell, 2006).

Conceptual Framework

Marital quality consists of both positive (e.g., love, support, and satisfaction) and negative (e.g., conflict, demands, and strains) dimensions. Kiecolt-Glaser and Newton’s conceptual framework (2001) suggests that both positive and negative dimensions of marital relationships have an impact on physical health primarily through psychological, physiological, and behavioral pathways. Many of these pathways have also been linked to cognitive limitations in older populations (Lee et al., 2010; Lupien, Maheu, Tu, Fiocco, & Schramek, 2007; Wilson et al., 2007).

Psychological well-being serves as one pathway through which marital quality might affect cognitive limitations. Positive marital quality is characterized by emotional support exchanges between marital partners. Emotional support, in turn, contributes to psychological well-being and lower risk for psychological distress, and psychological distress has been associated with risk for cognitive decline (Wilson et al., 2007). In addition, Seeman, Lusignolo, Albert, and Berkman (2001) have linked emotional support (from network members) to better cognitive functioning among older adults. Positive marital quality may also buffer the deleterious effects of stress on cognitive health through the provision of instrumental and emotional support (Cohen, 2004). Yet negative marital quality may lead to psychological distress that, in turn, undermines cognitive health.

Negative marital quality, as a source of stress, affects cognitive limitations through physiological pathways as well. Past research clearly shows that stressful relationships impair health (Robles & Kiecolt-Glaser, 2003). Marital strain has detrimental physiological consequences (e.g., increased heart rate, impaired immune function) that can undermine health (Kiecolt-Glaser & Newton, 2001). Further, research on stress and cognition suggests that the activation of hypothalamic-pituitary-adrenal axis in response to stress leads to subsequent production of the stress hormone cortisol, and high levels of circulating cortisol have damaging effects on cognitive functioning (De Kloet, Oitzl, & Joëls, 1999; Lupien et al., 2007). In a large population-based study with middle- to older-aged adults, Leng and colleagues (2013) found that those who reported the highest levels of stress in life had poorer cognitive functioning a decade later. However, moderate levels of stress and stress-induced circulating cortisol facilitate cognitive functioning (De Kloet et al., 1999). Comijs and colleagues (2011) found that experiences of mild chronic strain such as conflict with others over the previous three years was associated with better cognitive functioning among older adults. Thus, whether negative marital quality enhances or compromises cognitive functioning may depend on the intensity of stress that negative marital quality induces.

Health behaviors constitute another pathway through which marital quality might influence cognitive limitations. There is substantial evidence that health-promoting behaviors (e.g., good sleep habits, physical activity, and a healthy diet) protect against cognitive limitations whereas health risk behaviors (e.g., smoking, inactivity, and heavy drinking) compromise cognitive health (Lee et al., 2010). Since psychological distress and stress lead to more health risk behaviors, positive marital experiences may slow the increase in cognitive limitations among older adults by reducing stress levels and the risky behaviors that are used to cope with stress. Positive marital experiences may also cultivate a sense of responsibility to stay healthy in order to care for one’s spouse, and this goal of staying healthy may impede health-compromising behaviors and facilitate health-promoting behaviors (Umberson, 1987). In contrast, negative marital experiences/marital strain may generate stress that contributes to more health risk behaviors that, in turn, undermine cognitive health (Kassel, Stroud, & Paronis, 2003).

Theoretical and empirical work on social control suggests an additional possible explanation for a link between marital strain and health behaviors (Umberson, 1992). Spouses often attempt to influence each other’s health habits in ways that may promote better health habits and, potentially, better cognitive health. But, at the same time, these attempts to influence a spouse can be annoying and stressful. In this scenario, efforts to influence a spouse’s health habits (known as social control; Umberson, 1987) may actually contribute to higher levels of stress and marital strain (Birditt & Antonucci, 2008) even while having positive effects on health habits and cognitive health.

In sum, several bodies of prior research suggest possible underlying pathways linking both positive and negative marital quality to cognitive limitations among older adults. Although it is beyond the scope of the present study to thoroughly test each of these specific mechanisms (e.g., health behaviors, stress reactions), past empirical and theoretical work on these mechanisms provides sound reasons to expect linkages between marital quality and cognitive limitations.

Previous Research on Social Relationships and Cognitive Limitations

Recent studies have shown that being married (compared to being unmarried) protects against increased cognitive limitations among older adults (Hakansson et al., 2009; Karlamangla et al., 2009). While these studies focus on the relationship between marital status and cognitive limitations, they neglect to consider the quality of marital relationships. Another line of research focuses on social engagement and cognitive limitations and indicates that overall emotional support from one’s spouse, other family, and friends is protective against deterioration in cognitive functioning over time even after controlling for a variety of demographic, health status and health behavior characteristics, including educational attainment, depression, and chronic conditions (Seeman et al., 2001, 2011). In their cross-sectional analysis, Seeman and colleagues (2001) found that greater frequency of conflict or demands from network members is associated with better cognitive functioning among men and women aged 70–79. An additional study by Comijs and colleagues (2011) also shows that having serious conflicts with others in the previous 3 years is associated with improved learning of new information among older adults.

Using longitudinal data from the national survey of Midlife Development in the United States, Seeman and colleagues (2011) found that average social strain or conflict with a range of network members at Wave 1 was negatively related to executive functioning (cognitive processes that coordinate thought and action to obtain goals) but not episodic memory (a type of long-term memory of past experiences and events in one’s life) among older adults at Wave 2. Results from Roozendaal, Okuda, de Quervain, and McGaugh (2006) suggest that stress-induced hormones have dual effects of both reinforcing and compromising episodic memory, potentially explaining Seeman and colleagues’ nonsignificant effect of relationship strain on episodic memory. The social engagement literature often assesses average/overall emotional support from all network members, but the quality of any particular social relationship’s contribution to cognitive limitations is not typically addressed. Particularly, prior cross-sectional and longitudinal studies have not distinguished marital relationships from other relationships. Taken together, past research suggests that both positive and negative dimensions of marital quality may be important but neglected factors associated with cognitive limitations among older adults.

Directionality of the Association Between Marital Quality and Cognitive Limitations

The conceptual models described above imply that marital quality impacts cognitive limitations. Yet cognitive status may influence levels of marital quality as well. Cognitive limitations may compromise marital quality due to diminished sexuality, companionship, reciprocity, and ambiguity about the future of the relationship (Evans & Lee, 2014). Most empirical studies on the impact of cognitive limitations on marital quality have focused on the marital quality of people who are caregivers to a cognitively-impaired spouse.

Recent studies, however, have begun to examine marital quality as reported by cognitively impaired individuals. In their community-based study, Laganá and colleagues (2011) found that cognitive limitations are negatively related to marital adjustment and satisfaction among older women. Clare and colleagues (2012) reported that ratings of positive marital experiences from people with early-stage dementia, albeit lower, were not significantly different from those in a control group of persons who did not have dementia, and perceived marital quality from people with dementia did not decline significantly over an 18-month period. An important unaddressed question is whether cognitive limitations are associated with change in positive or negative dimensions of marital quality over a long period of time. Thus, examining the direction of the association between marital quality and cognitive limitations will provide a more thorough understanding of the link between marital quality and cognitive limitations.

Gender Differences in the Association Between Marital Quality and Cognitive Limitations

Gender as a structural location shapes experiences of marriage for men and women (Martin, 2004), and gendered experiences in marriage may contribute to gendered patterns in the association between marital quality and cognitive limitations. Women may experience higher levels of depression relative to men when confronted with marital strain as women are more likely than men to internalize rather than externalize their problems (Rosenfield, Lennon, & White, 2005). Clinical and laboratory-based studies on the association between marital interactions and physiological reactions also indicate a significant gender difference, with women generally more reactive than men to marital conflict (Robles & Kiecolt-Glaser, 2003). For example, marital strain seems to impair immune functioning in women more than men (Kiecolt-Glaser et al., 1993). Marriage also reinforces the gendered performance of social control in relation to health behaviors. As monitoring and caring for a spouse’s health are more likely to be viewed as feminine tasks, women may be more likely than men to engage in efforts to influence their spouse’s health habits (Umberson, 1992; Reczek & Umberson, 2012). Indeed, Umberson (1992) analyzes national data and shows that women are significantly more likely than men to tell or remind their spouse to do things to protect their health.

Clinical research further suggests that the impact of cognitive limitations on marital quality may differ by gender. For instance, Clare and colleagues (2012) found that ratings of marital quality by women with early-stage dementia declined whereas ratings by men with early-stage dementia increased during an 18-month follow-up. Although this gender difference was not statistically significant, studies tracking changes over a longer period of time may find significant gender differences in the impact of cognitive limitations on marital quality. Building on this prior work, we consider whether there are gender differences in the association between marital quality and cognitive limitations among older adults.

In sum, the aim of the current study is to examine whether positive and negative aspects of marital quality are associated with cognitive limitations over time, the direction of the relationship between cognitive limitations and marital quality, and potential gender differences in these associations.

Methods

Data

The ACL survey involved face-to-face interviews in the contiguous United States in 1986, 1989, 1994, and 2001/2002 (House, 2007). This nationally representative study includes individuals aged 25 and older in the Wave 1 sample with an oversampling of African Americans and adults aged 60 and older (N = 3617). The current study focuses on 841 white and African American adults (very few respondents reported other racial/ethnic groups) who were married and aged 60 or older at Wave 1, with 486 women and 355 men. Ideally, we would restrict our sample to respondents who were continuously married across all four waves. However, this would result in a very small sample with only 156 individuals aged 60 and older. To account for marital transitions over time, therefore, we created two flag variables to indicate whether respondents were widowed or divorced/separated after Wave 1 and included these flags as control variables in all analytical models.

Full Information Maximum Likelihood (FIML) estimation was employed to deal with missing data including panel attrition. FIML estimation is superior over other ways (e.g., listwise deletion, pairwise deletion, and similar response pattern imputation) of handling missing data in terms of parameter estimate bias, parameter estimate efficiency, and model fit (Enders & Bandalos, 2001). FIML procedures incorporate all available information in the sample for analyses, regardless of whether respondents participated in every wave of the survey (Wickrama, Mancini, Kwag, & Kwon, 2013).

As marital quality is inversely associated with mortality risk (Robles et al., 2014), those who survived over the 15-year period would be a healthier subpopulation of older adults. To account for mortality selection, we used a Heckman-type correction approach. A discrete-time hazard model for mortality on all respondents at Wave 1 over the 15-year interval was estimated as a function of Wave 1 variables associated with mortality risk, including self-rated health, number of children, religion, age, race, gender, education, income, and employment status (Rogers, Hummer, & Nam, 2000). The predicted mortality hazard from this model was included as a control variable in all subsequent analytical models. Following the Heckman-type correction, unobserved factors related to cognitive limitations and mortality are taken into account when estimating cognitive limitations (Liu, 2012). To further account for panel attrition, we included the number of waves in which respondents participated as a control variable (Warner & Brown, 2011).

The panel feature of the ACL study makes it particularly suitable for our analysis. The multiple waves of data enable us to track trajectories of change in individuals’ cognitive limitations, as well as assess the association between marital quality and change in cognitive limitations over time. The ACL study measures marital quality in each of the four waves as well, allowing us to address the directionality of the association between marital quality and cognitive limitations. We exclude people in cohabiting relationships from our analyses because marriage and cohabitation differ in their respective effects on health in the United States, and the gap between marriage and cohabitation on health can be partly explained by levels of relationship quality (Marcussen, 2005).

Measures

Cognitive limitations.

The ACL survey adopted the Short Portable Mental Status Questionnaire (SPMSQ) to test respondents’ cognitive limitations. The SPMSQ assesses respondents’ memory, knowledge of current events, and ability to perform mathematical tasks and was designed to identify cognitive deficits among people dwelling in both community and institutionalized settings (McDowell, 2006; Pfeiffer, 1975). Cognitive limitation is a combined score measured with four items. Respondents were asked: (a) “What is the date today—month, day, and year?” (b) “What day of the week is it?” (c) “What is the name of the president of the United States?” (d) “Subtract 3 from 20 and tell me the number you get. Then, keep subtracting 3 from this number and each new number you get, telling me the results as you go (Stop when the answer is 2 or less).” We eliminated the item concerning identification of the previous president because the survey year at Wave 2 occurred shortly after the 1988 presidential election and very few respondents answered the question incorrectly at this wave. Each cognitive limitation item was coded as 0 “correct” or 1 “incorrect”. Higher scores indicate poorer cognitive functioning. A natural logarithmic transformation was applied to normalize cognitive limitations at each wave.

Positive marital experience is a summary score composed of three items: (a) “Taking all things together, how satisfied are you with your marriage?” (b) “How much does your (husband/wife) make you feel loved and cared for?” and (c) “How much is (he/she) willing to listen when you need to talk about your worries or problems?” Responses for each item ranged from 0 “not at all” to 4 “a great deal”. Higher scores imply more satisfactory marital relationships.

Negative marital experience is a composite score consisting of two items: (a) “Taking everything into consideration, how often do you feel bothered or upset by your marriage?” and (b) “How often would you say the two of you typically have unpleasant disagreements or conflicts?” Each item included response categories ranging from 0 “never” to 4 “daily or almost daily”. Higher scores reflect more frequent experience of marital strain.

Control variables.

Previous studies suggest socioeconomic status is associated with the chance of marrying, staying married, and marital quality, as well as cognitive limitations (Cagney & Lauderdale, 2002; Smock, 2004). Particularly, high levels of education were associated with decreased odds of cognitive limitations among older adults even after controlling for current income (Cagney & Lauderdale, 2002). To control for the confounding effects of socioeconomic factors, we included education, income, and employment as control variables in our models. Education was coded as three dummy variables: less than high school, some college, college graduate and above (with high school graduate as the reference category). Total family income from all sources for the last 12 months consists of ten categories ranging from less than $5,000 to $80,000 or more. Employment status at Wave 1 was coded as 0 “not employed” and 1 “employed”. Other control variables that capture respondents’ demographic status are age, race, gender, and remarriage status. Age at Wave 1 ranges from 60 to 92 (in years) and was centered at its mean. Race is included as a dummy variable with 0 “white” and 1 “black”. Gender was coded as a dummy variable with 0 “male” and 1 “female”. We differentiated those in their first marriages from those in remarriages with a dummy variable (0 “first marriage” and 1 “remarried”).

Analytical Strategy

We used standard latent growth curve models (LGCM) to estimate the association of baseline marital quality with initial levels of cognitive limitations as well as the rate of change in cognitive limitations from Wave 1 to Wave 4. Thus change over time in this study means change over the 15-year period. The LGCM allows each respondent to have individual trajectories of cognitive limitations over time. The intercept (initial level) and the slope (growth rate over time) are two latent factors that vary randomly across respondents. The equations for the linear LGCM are:

yit=π0i+π1ixt+ eit (1)
π0i=b00+b01wi+r0i (2)
π1i= b10+b11wi+ r1i (3)

Equation 1 examines within-individual change over time. y it is the outcome variable (cognitive limitations) for individual i (i = 1, 2, 3,…, 841) at time point t (t = 1, 2, 3, or 4). π 0i is the latent intercept and π1i is the linear slope. x t is the time score that indicates number of years since Wave 1. Equations 2 and 3 examine between-individual change over time. b 00 is the average cognitive limitations at Wave 1 and b 10 is the average rate of change in cognitive limitations from Wave 1 to Wave 4 after controlling for all covariates. w i represents the time-invariant covariates. e it represents within-individual residual; r 0i and r 1i are between-individual residuals.

We used parallel LGCM to investigate the directionality of the relationship between marital quality and cognitive limitations. The parallel LGCM estimates individuals’ trajectories of marital quality and cognitive limitations simultaneously and tests whether the intercept of marital quality relates to the slope of cognitive limitations and whether the intercept of cognitive limitations is associated with marital quality over time.

Results

Table 1 shows the descriptive statistics for all variables of interest. Both positive and negative marital quality increased slightly over the four waves. Compared to men, women scored lower on positive marital experiences and higher on negative marital experiences. Even though women scored higher on cognitive limitations than men across waves, the differences between men and women are not statistically significant. The proportion of women who completed high school is higher compared to men. The proportion employed is higher for men than women, and men had higher household income relative to women. Women were more likely to become widowed over time and participated in more waves of the survey compared with men. The mortality hazard for women is lower than that for men.

Table 1.

Descriptive Statistics of All Variables (N = 841)

Total Women Men
Mean SD Mean SD Mean SD
Age (centered at 68) 0.14 6.39 0.02 5.87 0.32 7.03
Gender (1 = female) 0.58 ── ── ── ── ──
Race (1 = African American) 0.21 ── 0.20 ── 0.23 ──
Education
 Less than high school 0.45 ── 0.43 ── 0.48 ──
 High school 0.29 ── 0.33** ── 0.24 ──
 Some college 0.16 ── 0.16 ── 0.16 ──
 College and above 0.10 ── 0.08 ── 0.12 ──
Remarriage status (1 = remarried) 0.28 ── 0.26 ── 0.31 ──
Employment (1 = employed) 0.27 ── 0.19*** ── 0.38 ──
Household income 4.55 2.44 4.35** 2.38 4.83 2.50
Widowed after Wave 1 0.29 ── 0.42*** ── 0.12 ──
Divorced/separated after Wave 1 0.02 ── 0.02 ── 0.02 ──
Number of waves 2.75 1.07 2.90*** 1.05 2.55 1.06
Positive marital experiences at Wave 1 9.88 2.23 9.51*** 2.37 10.38 1.91
Positive marital experiences at Wave 2 9.63 2.42 9.15*** 2.58 10.27 2.02
Positive marital experiences at Wave 3 9.63 2.42 9.15*** 2.58 10.27 2.02
Positive marital experiences at Wave 4 10.17 2.09 9.75* 2.20 10.59 1.89
Negative marital experiences at Wave 1 2.13 1.51 2.22* 1.57 2.00 1.43
Negative marital experiences at Wave 2 2.31 1.60 2.49*** 1.63 2.06 1.53
Negative marital experiences at Wave 3 2.35 1.52 2.52* 1.56 2.15 1.44
Negative marital experiences at Wave 4 2.38 1.55 2.49 1.55 2.27 1.56
Cognitive limitations at Wave 1 (log transformed) 0.35 0.45 0.36 0.45 0.34 0.45
Cognitive limitations at Wave 2 (log transformed) 0.30 0.44 0.32 0.44 0.29 0.43
Cognitive limitations at Wave 3 (log transformed) 0.39 0.44 0.40 0.43 0.37 0.45
Cognitive limitations at Wave 4 (log transformed) 0.40 0.45 0.42 0.47 0.36 0.43
Mortality hazard 0.03 0.03 0.02*** 0.02 0.05 0.04
Open in a new tab

Notes. At Wave 2: N = 666 (N = 399 women, N = 267 men); at Wave 3: N = 501 (N = 316 women, N = 185 men); at Wave 4: N = 288 (N = 197 women, N = 91 men).

*p < .05, **p < .01, ***p < .001 for significant differences between men and women.

Consistent with previous research, the results from the unconditional LGCM (without covariates) indicate that levels of cognitive limitations increase over the 15-year period among older adults (b 10 = .009; p < .001). The average level of cognitive limitations at Wave 1 across all older adults is .326 (p < .001). We also find significant variation in levels of cognitive limitations among individuals at Wave 1 and marginally significant variation for the slope of cognitive limitations (results not shown).

Table 2 presents results from the LGCM estimating the associations of marital quality with the initial levels of cognitive limitations and the growth rate in cognitive limitations over time. Age is inversely associated with initial levels of cognitive limitations at Wave 1. African Americans and those with less than a high school education had higher initial levels of cognitive limitations. Participation in more waves of the study is associated with lower initial levels of cognitive limitations. The mortality hazard over the 15-year period is positively related to initial levels of cognitive limitations. Gender, however, is not significantly associated with initial levels of cognitive limitations.

Table 2.

Relationships Between Baseline Marital Quality and Cognitive Limitations From Linear Growth Curve Model

Intercept Slope
Estimate SE Estimate SE
Positive marital experiences .010 .007 .000 .001
Negative marital experiences .007 .010 −.003* .001
Age (centered at 68) −.008* .004 .002*** .001
Gender (1 = female) .069 .036 −.002 .005
Race (1 = African American) .127*** .035 .006 .005
Educationa
 Less than high school .188*** .033 .003 .004
 Some college −.019 .040 .002 .005
 College and above −.019 .047 −.004 .005
Remarriage status (1 = remarried) −.024 .029 .005 .004
Employment (1 = employed) −.007 .032 .000 .004
Household income −.012 .006 .001 .001
Widowed after Wave 1 .042 .031 −.002 .004
Divorced/separated after Wave 1 −.118 .092 .026* .010
Number of waves −.036* .015 −.003 .003
Mortality hazard 1.721* .835 −.223 .181
Means of growth parameters .159 .114 .039* .019
Variances in growth parameters .046*** .008 .000 .000
Pseudo-R square .368 .869
Model fit index CFI=.961 RMSEA=.025
Open in a new tab

Notes. CFI = comparative fit index; RMSEA = root mean square error of approximation.

aThe reference group is respondents with high school education.

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

Results from Table 2 also indicate that negative marital experiences are significantly associated with the slope of cognitive limitations over time. To illustrate this relationship, we graphed the association between negative marital experiences and change in cognitive limitations (Supplementary Figure 1). Figure 1 shows the predicted trajectories of cognitive limitations for those with high (1 SD above the mean) and low (1 SD below the mean) frequency of negative marital experiences. Older adults with more frequent negative marital experiences at wave 1 experienced a slower rate of increase in cognitive limitations over time; whereas older adults with less frequent negative marital experiences at Wave 1 experienced a faster increase in cognitive limitations over time. Higher frequency of negative marital experiences is inversely associated with cognitive limitations over time, suggesting beneficial effects of negative marital experiences on cognitive functioning. Still, gender is not a significant predictor of change in cognitive limitations over time.

Results further indicate that the older a respondent was at Wave 1, the faster the increase in cognitive limitations over time. Respondents who were divorced between Wave 1 and 4 experienced a more rapid increase in cognitive limitations over the 15-year period. Contrary to our expectations, the associations between positive/negative marital experiences and cognitive limitations are similar for men and women (results not shown).

Table 3 presents results regarding the directionality of the relationship between marital quality and cognitive limitations. Results indicate that the intercept of negative marital experiences is significantly associated with the growth rate of cognitive limitations over time among older adults. The higher the frequency of initial negative marital experiences, the slower the rate of increase in cognitive limitations over time. The intercept of cognitive limitations is not significantly associated with changes in marital quality over time, however, suggesting that the flow is from marital quality to cognitive limitations rather than vice versa. Consistent with the descriptive results, the parallel models indicate that, compared to men, women reported fewer positive marital experiences as well as higher levels of negative marital experiences at baseline. Gender, however, was not significantly related to changes in marital quality/cognitive limitations over time. Moreover, the association between marital quality and cognitive limitations in the parallel models is similar for men and women (results not shown).

Table 3.

Direction of the Relationships Between Marital Quality and Cognitive Limitations

Intercept of Marital Quality Slope of Marital Quality Intercept of Cognitive Limitations Slope of Cognitive Limitations
Positive marital experiences
 Intercept of positive marital experiences ── ── ── .001
 Intercept of cognitive limitations ── .017 ── ──
 Means 10.635*** .034 .277*** .016
 Variance 2.823*** .000a .047*** .000
 Pseudo-R square .120 1.000 .364 .760
 Model fit index CFI = .975 RMSEA = .021
Negative marital experiences
 Intercept of negative marital experiences ── ── ── −.004*
 Intercept of cognitive limitations ── −.043 ── ──
 Means 1.867*** .007 .278*** .037*
 Variance 1.277*** .002 .047*** .000
 Pseudo-R square .038 .483 .361 .889
 Model fit index CFI = .962 RMSEA = .024
Open in a new tab

Notes. CFI = comparative fit index; RMSEA = root mean square error of approximation. ── suggests the parameter was not included in model. All the control variables were included in both models.

aThe variance of the slope for positive marital quality is constrained to be 0 due to lack of variation over time.

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

Discussion

The consequential impact of cognitive limitations on older adults’ quality of life, family caregivers, and the American health care system has stimulated numerous studies seeking to detect factors that may contribute to or prevent cognitive decline among older adults (Fisher et al., 2011). Given the salience of marital relationships for older adults, the current study investigates whether marital quality is associated with change in cognitive limitations over time, the direction of the association between marital quality and cognitive limitations, and the moderating effects of gender on the association between marital quality and cognitive limitations. Our results from LGCM analyses indicate that cognitive limitations increase over time among older adults but negative marital experiences are associated with a slower increase in cognitive limitations over time. Our results also suggest that the association between marital quality and cognitive limitations flows primarily from marital quality to cognitive limitations rather than vice versa. We do not find evidence of an association between positive marital experiences and change in cognitive limitations or a moderating effect of gender on the association between marital quality and cognitive limitations.

The finding that more frequent negative marital experiences at wave 1 are associated with a slower increase in cognitive limitations over time is not consistent with the inverse association of negative marital experiences and physical health that has been well documented in past research (Umberson et al., 2006). This may occur because the majority of older adults in the current sample had relatively infrequent negative marital experiences, and the occasional experience of marital strain over an extended period of time may create moderate stress that serves to sharpen cognitive functioning. This explanation is supported by prior studies showing that moderate stress leads to mildly elevated cortisol levels; in turn, moderately elevated cortisol is associated with enhanced cognitive functioning (De Kloet et al., 1999). A recent longitudinal study also shows that older adults who experienced moderate chronic strain (e.g., serious conflicts with others or illness of a relative) over a three year period were better at learning new information than those who did not experience these chronic strains (Comijs et al., 2011). More population-based studies are needed to examine the relationship between varying levels of chronic stress and cognitive functioning over time and, ideally, these studies would follow individuals over time to assess their levels of cognitive functioning prior to and after exposure to stress.

The social control model suggests that some of the regulatory actions that spouses enact in an effort to promote healthier behaviors may actually impose strain on their marital relationship. If this occurs, more positive health behaviors could also contribute to a slower increase in cognitive limitations over time. However, we conducted an ancillary analysis that suggests controlling for smoking, alcohol consumption, and physical activity at Wave 1 does not alter the significant association between negative marital experiences and change in cognitive limitations over time. Future studies should examine possible behavioral pathways linking negative marital quality and cognitive limitations in greater detail.

Negative marital experiences may also signify invisible support from a spouse that is not recognized as support by the recipient of that support. Invisible support is more conducive to recipients’ health than is perceived support, which is more likely to involve some emotional costs (e.g., a cost to self-esteem; Bolger, Zuckerman, & Kessler, 2000). For instance, older adults may feel bothered or upset when their spouse constantly urges or reminds them to adopt healthier lifestyles (e.g., eat healthier food, take medicines on time) that are salutary to cognitive health. It is also possible that older adults who frequently engage in challenging or demanding situations (e.g., marital conflict) are cognitively healthier (a “use it or lose it” process) because conflicted interactions entail intensive cognitive processing including attention, reasoning, speed of processing, language, and executive functioning (Seeman et al., 2011). The association between negative marital experiences and change in cognitive limitations also suggests a possible mechanism linking social integration to cognitive functioning. The marital relationship, as a key source of social integration, may slow the increase in cognitive limitations over time as a result of cognitive demands and social control efforts that are associated with strained interactions. This explanation also suggests why social isolation or loneliness, due to a lack of social interaction and social control, is associated with higher levels of cognitive limitations (Shankar et al., 2013).

Thoits (1995) has proposed that individuals actively search for solutions to marital problems or conflict (e.g., reconstructing the meaning of negative marital experiences) to allay anxiety and stress as well as promote health. Therefore, married older adults may be more resilient to the effects of negative marital experiences on their cognitive status because they have developed effective coping strategies that turn out to be protective against a steeper growth in cognitive limitations. For instance, older adults experiencing high levels of marital strain may be more likely to participate in social activities or to seek social support from their children, friends, and neighbors in an effort to cope with strain in their marriage. In turn, these social resources may protect individuals from decline in cognitive limitations. Even though supplementary analyses indicate that the significant association between negative marital experiences and cognitive limitations remains unchanged after controlling for social integration and social support at Wave 1, future research should investigate whether other psychosocial factors (e.g., positive emotions and positive reappraisal) explain the association between negative marital experiences and slower growth in cognitive limitations.

We also raised questions about the direction of the association between marital quality and cognitive limitations. Results from parallel LGCM analyses show that negative marital experiences are significantly and inversely associated with cognitive limitations over time. Yet no significant association between initial levels of cognitive limitations and change in marital quality was found. Marital quality among people in early-stage dementia does not typically decline precipitously (Clare et al., 2012); rather, as suggested by previous qualitative work, closeness and intimacy may actually increase among people in early-stage dementia (Harris, 2009). Those with higher levels of cognitive limitations may also be less aware and care less about the subtleties of their marital interactions. If this is the case, those with no/fewer cognitive limitations might be more likely to experience and report marital strain. Supplementary analyses, however, indicate that cognitive limitations are not associated with initial levels or change in marital quality over time even among those with no/low levels of initial cognitive limitations. In future studies, longitudinal dyadic data would be well suited to understanding how marital quality and cognitive limitations are linked and unfold over time, from the point of view of both the impaired and unimpaired partner.

Our study suggests no significant gender differences in the estimated effects of marital quality on change in cognitive limitations. Even though past theoretical work on gender and lab-based studies suggest that women are more vulnerable to marital strain compared to men, population-based longitudinal studies suggest that women are not more susceptible to the effects of marital strain on health (Umberson & Williams, 2005). Our results are also consistent with other longitudinal studies on marital quality and physical health (Umberson et al., 2006). Using the first three waves of the ACL survey, Umberson and colleagues (2006) found no significant gender difference in the relationship between marital quality and change in self-rated health after controlling for sociodemographic characteristics. Results from our parallel models also suggest that the association between initial levels of cognitive limitations and change in marital quality over time does not vary significantly by gender, lending additional support to the clinical findings by Clare and colleagues (2012). Our findings then contribute to growing population-based evidence that marital quality affects various aspects of men’s and women’s health in similar ways.

We note several limitations of the present study. One limitation lies in the lack of a more comprehensive assessment of cognitive limitations, which is a common issue associated with measures of cognitive status in large scale surveys, primarily due to time and logistical constraints in data collection (Lachman & Spiro, 2002). Even though the SPMSQ is a well-accepted measure of cognitive limitations in survey data, this approach cannot capture/compare all of the important dimensions of cognition. The SPMSQ is primarily useful for identifying cognitive deficits as well as cognitive limitations related to dementia but lacks the sensitivity to diagnose neuropsychological aspects of cognitive aging (McDowell, 2006). Moreover, significant cognitive limitations observed at older ages result from years of cumulative decline in fluid cognition that varies from year to year even among younger respondents. Measures of fluid cognition are needed to better address possible causal relationships between marital quality and cognitive limitations. Future research should investigate whether and how marital quality is associated with different aspects of cognitive limitations by employing tests that provide in-depth appraisals of specific cognitive functions such as orientation, executive functioning, attention and memory as well as more sensitive cognitive measures to identify decline in fluid cognition at younger ages. Another limitation associated with this study is selection bias. Older adults who remain married at age 60 and older are likely to be healthier than their unmarried counterparts. Therefore, the findings of this study only apply to the selected population of people who are married in older adulthood as those experiencing extremely strained marital relationships may have divorced. Net of these limitations, this study advances our understanding of the association between marital quality and changes in cognitive limitations over time.

Our findings add to emerging evidence that marital relationships benefit cognitive functioning (Hakansson et al., 2009; Karlamangla et al., 2009), albeit in a somewhat counterintuitive fashion. It seems that supportive aspects of relationships are less important than marital strain in relation to change in cognitive limitations over time among older adults. Moreover, marital strain is associated with less, not more, acceleration in cognitive limitations over time, for both men and women. Future research should aim to identify and better understand the underlying mechanisms connecting marital strain to a slower increase in cognitive limitations over time.

Supplementary Material

Supplementary material can be found at: http://psychsocgerontology.oxfordjournals.org/

Supplementary Data
supp_71_1_165__index.html (764B, html)

Acknowledgments

This research has received support from grant, 5 T32 HD007081, Training Program in Population Studies, and the center grant 5 R24 HD042849, awarded to the Population Research Center at The University of Texas at Austin by the Eunice Kennedy Shriver National Institute of Child Health and Human Development. M. Xu developed the conceptual framework for this study, conducted the data analysis and wrote the paper. P. A. Thomas supervised the data analysis, drafted and revised the article, and assisted developing the conceptual framework. D. Umberson planned the study, drafted and revised the article, and assisted with the conceptual framework.

References

  1. Atti A. R., Forlani C., De Ronchi D., Palmer K., Casadio P., Dalmonte E., Fratiglioni L. (2010). Cognitive impairment after age 60: Clinical and social correlates in the Faenza Project. Journal of Alzheimer’s Disease, 21, 1325–1334. doi:10.3233/JAD2010091618 [DOI] [PubMed] [Google Scholar]
  2. Birditt K., Antonucci T. C. (2008). Life sustaining irritations? Relationship quality and mortality in the context of chronic illness. Social Science & Medicine (1982), 67, 1291–1299. doi:10.1016/j.socscimed.2008.06.029 [DOI] [PMC free article] [PubMed] [Google Scholar]
  3. Bolger N., Zuckerman A., Kessler R. C. (2000). Invisible support and adjustment to stress. Journal of Personality and Social Psychology, 79, 953–961. doi:10.1037/00 22-3514.79.6.953 [DOI] [PubMed] [Google Scholar]
  4. Cagney K. A., Lauderdale D. S. (2002). Education, wealth, and cognitive function in later life. The Journals of Gerontology, Series B: Psychological Sciences and Social Sciences, 57, P163–P172. doi:10.1093/geronb/57.2.P163 [DOI] [PubMed] [Google Scholar]
  5. Clare L., Nelis S. M., Whitaker C. J., Martyr A., Markova I. S., Roth I., … Morris R. G. (2012). Marital relationship quality in early-stage dementia. Alzheimer Disease & Associated Disorders, 26, 148–158. doi:10.1097/wad.0b013e318 221ba23 [DOI] [PubMed] [Google Scholar]
  6. Cohen S. (2004). Social relationships and health. The American Psychologist, 59, 676–684. doi:10.1037/0003-066X.59.8.676 [DOI] [PubMed] [Google Scholar]
  7. Comijs H. C., van den Kommer T. N., Minnaar R. W. M., Penninx B. W. J. H., Deeg D. J. H. (2011). Accumulated and differential effects of life events on cognitive decline in older persons: Depending on depression, baseline cognition, or ApoE ε4 status? The Journals of Gerontology, Series B: Psychological Sciences and Social Sciences, 66 (suppl 1), i111–i120. doi:10.1093/geronb/gbr019 [DOI] [PubMed] [Google Scholar]
  8. De Kloet E. R., Oitzl M. S., Joëls M. (1999). Stress and cognition: Are corticosteroids good or bad guys? Trends in Neurosciences, 22, 422–426. doi:10.1016/s0166-2236 (99)01438-1 [DOI] [PubMed] [Google Scholar]
  9. Enders C. K., Bandalos D. L. (2001). The relative performance of full information maximum likelihood estimation of missing data in structure equation models. Structural Equation Modeling, 8, 430–457. doi:10.1207/S15328007SEM080_5 [Google Scholar]
  10. Evans D., Lee E. (2014). Impact of dementia on marriage: A qualitative systematic review. Dementia, 13, 330–349. doi:10.1177/1471301212473882 [DOI] [PubMed] [Google Scholar]
  11. Fisher G. G., Franks M. M., Plassman B. L., Brown S. L., Potter G. G., Llewellyn D. … Langa K. M. (2011). Caring for individuals with dementia and cognitive impairment, not dementia: Findings from the aging, demographics, and memory study. Journal of the American Geriatrics Society, 59, 488–494. doi:10.1111/j.1532-5415.2010.03304.x [DOI] [PMC free article] [PubMed] [Google Scholar]
  12. Hakansson K., Rovio S., Helkala E., Helkala E., Vilska A., Winblad B., … Kivipelto M. (2009). Association between mid-life marital status and cognitive function in later life: Population based cohort study. British Medical Journal, 339, 1–8. doi:10.1136/bmj.b2 462 [DOI] [PMC free article] [PubMed] [Google Scholar]
  13. Harris P. (2009). Intimacy, sexuality and early-stage dementia: The changing marital relationship. Alzheimer’s Care Today, 10, 63–77. doi:10.1097/ACQ.0b013e3181a40d80 [Google Scholar]
  14. House J. S. (2007). Americans’ changing lives: Waves I, II, III, and IV, 1986, 1989, 1994, and 2002 [Computer file]. ICPSR04690-v3. Ann Arbor, MI: University of Michigan, Institute for Social Research, Survey Research Center. [Google Scholar]
  15. Karlamangla A. S., Miller-Martinez D., Aneshensel C. S., Seeman T. E., Wight R. G., Chodosh J. (2009). Trajectories of cognitive function in late life in the United States: Demographic and socioeconomic predictors. American Journal of Epidemiology, 170, 331–342. doi:10.1093/aje/kwp154 [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. Kassel J. D., Stroud L. R., Paronis C. A. (2003). Smoking, stress, and negative affect: Correlation, causation, and context across stages of smoking. Psychological Bulletin, 129, 270–304. doi:10.1037/0033-2909.129.2.270 [DOI] [PubMed] [Google Scholar]
  17. Kiecolt-Glaser J. K., Malarkey W. B., Chee M., Newton T., Cacioppo J. T., Mao H. Y., Glaser R. (1993). Negative behavior during marital conflict is associated with immunological down-regulation. Psychosomatic Medicine, 55, 395–409. doi:10.1097/ 00006842-199309000-00001 [DOI] [PubMed] [Google Scholar]
  18. Kiecolt-Glaser J. K., Newton T. L. (2001). Marriage and health: His and hers. Psychological Bulletin, 127, 472–503. doi:10.1037/0033-2909.127.4.472 [DOI] [PubMed] [Google Scholar]
  19. Lachman M., Spiro A. (2002). Critique of cognitive measures in the health retirement study (HRS) and the asset and health dynamics among the oldest old (AHEAD) study. Retrieved from http://hrsonline.isr.umich.edu/sitedocs/dmc/Lachman_hrscognitive.pdf [Google Scholar]
  20. Laganá L., Spellman T., Wakefield J., Oliver T. (2011). Ethnic minority status, depression, and cognitive failures in relation to marital adjustment in ethnically diverse older women. Clinical Gerontologist, 34, 173–189. doi:10.1080/07317115.2011.554627 [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. Lee Y., Back J. H., Kim J., Kim S. H., Na D. L., Cheong H. K. … Kim Y. G. (2010). Systematic review of health behavioral risks and cognitive health in older adults. International Psychogeriatrics, 22, 174–187. doi:10.1017/S1041610209991189 [DOI] [PubMed] [Google Scholar]
  22. Leng Y., Wainwright N. W., Hayat S., Stephan B. C., Matthews F. E., Luben R. … Brayne C. (2013). The association between social stress and global cognitive function in a population-based study: The European Prospective Investigation into Cancer (EPIC)-Norfolk study. Psychological Medicine, 43, 655–666. doi:10.1017/S0033291712001316 [DOI] [PubMed] [Google Scholar]
  23. Liu H. (2012). Marital dissolution and self-rated health: Age trajectories and birth cohort variations. Social Science & Medicine (1982), 74, 1107–1116. doi:10.1016/j.socscimed.2011.11.037 [DOI] [PubMed] [Google Scholar]
  24. Lupien S. J., Maheu F., Tu M., Fiocco A., Schramek T. E. (2007). The effects of stress and stress hormones on human cognition: Implications for the field of brain and cognition. Brain and Cognition, 65, 209–237. doi:10.1016/j.bandc.2007.02.007 [DOI] [PubMed] [Google Scholar]
  25. Marcussen K. (2005). Explaining differences in mental health between married and cohabi-ting individuals. Social Psychology Quarterly, 68, 239–257. doi: 10.2307/4148772 [Google Scholar]
  26. Martin P. Y. (2004). Gender as social institution. Social Forces, 82, 1249–1273. doi:10.1353 /sof.2004.0081 [Google Scholar]
  27. McDowell I. (2006). Measuring health: A guide to rating scales and questionnaires. New York, NY: Oxford University Press. [Google Scholar]
  28. Pfeiffer E. (1975). A short portable mental status questionnaire for the assessment of organic brain deficit in elderly patients. Journal of the American Geriatrics Society, 23, 433–441. [DOI] [PubMed] [Google Scholar]
  29. Plassman B. L., Langa K. M., Fisher G. G., Heeringa S. G., Weir D. R., Ofstedal M. B., … Wallace R. B. (2008). Prevalence of cognitive impairment without dementia in the United States. Annals of Internal Medicine, 148, 427–434. doi: 10.7326/0003-4819-148-6-200803180-00005 [DOI] [PMC free article] [PubMed] [Google Scholar]
  30. Reczek C., Umberson D. (2012). Gender, health behavior, and intimate relationships: Lesbian, gay, and straight contexts. Social Science & Medicine (1982), 74, 1783–1790. doi:10.1016/j.socscimed.2011.11.011 [DOI] [PMC free article] [PubMed] [Google Scholar]
  31. Robles T. F., Kiecolt-Glaser J. K. (2003). The physiology of marriage: Pathways to health. Physiology & Behavior, 79, 409–416. doi:10.1016/S0031-9384(03)00160-4 [DOI] [PubMed] [Google Scholar]
  32. Robles T. F., Slatcher R. B., Trombello J. M., McGinn M. M. (2014). Marital quality and health: A meta-analytic review. Psychological Bulletin, 140, 140–187. doi:10.1037/a0031859 [DOI] [PMC free article] [PubMed] [Google Scholar]
  33. Rogers R., Hummer R., Nam C. (2000). Living and dying in the U.S.A.: Behavioral, health, and social differentials in adult mortality. San Diego, CA: Academic Press. [Google Scholar]
  34. Roozendaal B., Okuda S., de Quervain D. J., McGaugh J. L. (2006). Glucocorticoids interact with emotion-induced noradrenergic activation in influencing different memory functions. Neuroscience, 138, 901–910. doi:10.1016/j.neuroscience.2005.07.049 [DOI] [PubMed] [Google Scholar]
  35. Rosenfield S., Lennon M. C., White H. R. (2005). The self and mental health: Self-salience and the emergence of internalizing and externalizing problems. Journal of Health and Social Behavior, 46, 323–340. doi:10.1177/002214650504600402 [DOI] [PubMed] [Google Scholar]
  36. Seeman T. E., Lusignolo T. M., Albert M., Berkman L. (2001). Social relationships, social support, and patterns of cognitive aging in healthy, high-functioning older adults: MacArthur Studies of Successful Aging. Health Psychology, 20, 243–255. doi:10.1037/0278-6133.20.4.243 [DOI] [PubMed] [Google Scholar]
  37. Seeman T. E., Miller-Martinez D. M., Stein Merkin S., Lachman M. E., Tun P. A., Karlamangla A. S. (2011). Histories of social engagement and adult cognition: Midlife in the U.S. study. The Journals of Gerontology, Series B: Psychological Sciences and Social Sciences, 66(Suppl. 1), i141–i152. doi:10.1093/geronb/gbq091 [DOI] [PMC free article] [PubMed] [Google Scholar]
  38. Shankar A., Hamer M., McMunn A., Steptoe A. (2013). Social isolation and loneliness: Relationships with cognitive function during 4 years of follow-up in the English Longitudinal Study of Ageing. Psychosomatic Medicine, 75, 161–170. doi:10.1097/PSY.0b013e31827f09cd [DOI] [PubMed] [Google Scholar]
  39. Smock P. J. (2004). The wax and wane of marriage: Prospects for marriage in the 21st century. Journal of Marriage and Family, 66, 966–973. doi:10.1111/j.00222445.2004. 00066.x [Google Scholar]
  40. Stott D. J. (2006). Cognitive decline in ageing. Asian Journal of Gerontology & Geriatrics, 1, 21–25. [Google Scholar]
  41. Thoits P. A. (1995). Stress, coping, and social support processes: Where are we? What next? Journal of Health and Social Behavior, 35, 53–79. doi:10.2307/2626957 [PubMed] [Google Scholar]
  42. Umberson D. (1992). Gender, marital status and the social control of health behavior. Social Science & Medicine, 34, 907–917. doi:10.1016/0277-9536(92)90259-s [DOI] [PubMed] [Google Scholar]
  43. Umberson D. (1987). Family status and health behaviors: Social control as a dimension of social integration. Journal of Health and Social Behavior, 28, 306–319. doi:10.2307/ 2136848 [PubMed] [Google Scholar]
  44. Umberson D., Williams K. (2005). Marital quality, health, and aging: Gender equity? The Journals of Gerontology, Series B: Psychological Sciences and Social Sciences, 60, S109–S113. doi:10.1093/geronb/60.Special_Issue_2.S109 [DOI] [PMC free article] [PubMed] [Google Scholar]
  45. Umberson D., Williams K., Powers D. A., Liu H., Needham B. (2006). You make me sick: Marital quality and health over the life course. Journal of Health and Social Behavior, 47, 1–16. doi:10.1177/002214650604700101 [DOI] [PMC free article] [PubMed] [Google Scholar]
  46. Warner D. F., Brown T. H. (2011). Understanding how race/ethnicity and gender define age-trajectories of disability: An intersectionality approach. Social Science & Medicine (1982), 72, 1236–1248. doi:10.1016/j.socscimed.2011.02.034 [DOI] [PMC free article] [PubMed] [Google Scholar]
  47. Whisman M. A., Sbarra D. A. (2012). Marital adjustment and interleukin-6 (IL-6). Journal of Family Psychology, 26, 290–295. doi:10.1037/a0026902 [DOI] [PMC free article] [PubMed] [Google Scholar]
  48. Wickrama K. K., Mancini J. A., Kwag K., Kwon J. (2013). Heterogeneity in multidimensional health trajectories of late old years and socioeconomic stratification: A latent trajectory class analysis. The Journals of Gerontology, Series B: Psychological Sciences and Social Sciences, 68, 290–297. doi:10.1093/geronb/gbs111 [DOI] [PubMed] [Google Scholar]
  49. Wilson R. S., Schneider J. A., Boyle P. A., Arnold S. E., Tang Y., Bennett D. A. (2007). Chronic distress and incidence of mild cognitive impairment. Neurology, 68, 2085–2092. doi:10.1212/01.wnl.0000264930.97061.82 [DOI] [PubMed] [Google Scholar]
  50. Zunzunegui M. V., Alvarado B. E., Del Ser T., Otero A. (2003). Social networks, social integration, and social engagement determine cognitive decline in community-dwelling Spanish older adults. The Journals of Gerontology, Series B: Psychological Sciences and Social Sciences, 58, S93–S100. doi:10.1093/geronb/58.2.S93 [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Supplementary Data
supp_71_1_165__index.html (764B, html)
supp_gbv014_Supplementary_Figure_1.jpg (12.2KB, jpg)

Articles from The Journals of Gerontology Series B: Psychological Sciences and Social Sciences are provided here courtesy of Oxford University Press

RESOURCES