Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2019 Oct 1.
Published in final edited form as: J Fam Psychol. 2018 Jun 25;32(7):936–946. doi: 10.1037/fam0000435

Marital Processes Linking Economic Hardship to Mental Health: The Role of Neurotic Vulnerability

Kandauda (KAS) Wickrama 1, Catherine Walker O’Neal 2,✉, Frederick O Lorenz 3
PMCID: PMC6205895  NIHMSID: NIHMS973782  PMID: 29939041

Abstract

Both enduring neurotic vulnerabilities and economic hardship have been shown to negatively influence marital behaviors, which have physical and mental health consequences. However, because most previous research is fragmented and has focused on the early years of marriage or relatively short periods of time, their long-term effects are unclear. Using data from the Iowa Midlife Transitions Project, with a sample of 370 married couples providing data from 1991 to 2001, we assessed enduring personal and couple vulnerabilities, trajectories of family economic hardship, and couples’ marital hostility using a comprehensive dyadic model to ascertain their influence on subsequent mental health. Couple marital hostility trajectories and neurotic vulnerabilities (both additively and interactively) were associated with changes in both spouses’ depressive symptoms. Results also indicated that couples’ marital hostility trajectories link trajectories of family economic hardship to subsequent changes in husbands’ and wives’ depressive symptoms. Last, associations between economic hardship trajectories, marital hostility trajectories, and depressive symptoms were moderated by couples’ neurotic vulnerability as captured by a product term of husbands’ and wives’ neurotic vulnerability. In general, these associations were amplified for couples with a high level of couple vulnerability and weakened (or altogether absent) for those with a low level of vulnerability.

Keywords: economic hardship, midlife, marital hostility, mental health, neuroticism

Marital relationships are affected by various personal characteristics established early in life, such as neuroticism, as well as life course experiences that occur before and during marriage. In particular, long-term economic difficulties are a life course experience with a continuing influence on marital relationships above and beyond enduring personal characteristics (Conger, Rueter & Elder, 1999; Karney, Bradbury, Fincham, & Sullivan, 1994). While changes in the marital relationship over time have been shown to influence later health outcomes (Miller, Hollist, Olsen, & Law, 2013; Wickrama, Lorenz, Conger, & Elder, 1997), little research has investigated changes in marital relationships as a central mediating construct in relation to stressful life experiences and later health outcomes from a life course stress process perspective (Pearlin, Schieman, Fazio, & Meersman, 2005), particularly in enduring marriages and during the middle years. The middle years include approximate ages between 40 and 60 years (Staudinger & Bluck, 2001) and are a time that is rich in life transition events (Lackman & James, 1997). More specifically, the current sample follows couples from their early middle years (average age of approximately 40) to their later middle years (average age of approximately 50). Consequently, this study focused on mental health consequences of stressful life experiences for individuals in enduring marriages in their middle years may identify unique patterns of findings compared to previous research on couples at different life stages.

From a life course framework, it is important to consider how the specific life experiences and developmental period of the current sample situate this study. In particular, the sample is comprised of Midwest couples from enduring marriages who experienced the “farm crisis” of the late 1980s as they were entering middle adulthood, a time when offspring were transitioning out of the home and aging parents were beginning to experience declines. The relevance of economic difficulties, as a particularly stressful life experience, for marital relationships has been repeatedly noted for couples who faced the “farm crisis” (Lorenz, Conger, Montague & Wickrama, 1993; Lorenz, Elder, Bao, Wickrama, & Conger, 2000). The lingering influence of these stressful life experiences may adversely impact their marital relationships over the middle years. While some of these marriages have dissolved in subsequent years, many have continued to experience life’s major transitions together (e.g., becoming “empty nesters”). As these husbands and wives approach their later middle years, the marital relationship may become even more salient to their well-being (Berg, Johnson, Meegan, & Strough, 2003). Research shows that as individuals age, they tend to keep and cultivate the relationships that are most emotionally important to them (Levenson, Carstensen, & Gottman, 1994). In particular, aging individuals generally place increased importance on the affiliation dimension of their marital relationship (Smith et al., 2009). Thus, it is reasonable to expect strong emotional responses to hostile behavioral in studies of older husbands and wives, and it is necessary to determine if findings from previous studies, mainly focused on couples in their early years of marriage and over short periods of time, generalize to the current study population.

Such studies can enhance our understanding of how changes in marital relationships during the middle years are related to mental health in later middle years. This knowledge has the potential to inform preventive intervention programs targeting middle-aged adults in enduring marriages. Consequently, we utilized prospective data collected from couples in their early middle years to examine couples’ hostile marital interactions as a pivotal marital process and assess the long-term antecedents of hostility (including experiencing economic hardship and neurotic vulnerabilities) as well as the subsequent mental health consequences.

The conceptual framework shown in Figure 1 is drawn from several theories of marriage including the vulnerability-stress adaptation model (VSAM; Karney & Bradbury, 1995), the self-regulatory depletion approach (Baumeister, 2002), and family system theory (Constantine, 1986; Larson, Wilson, & Beley, 1994). The VSAM (Karney & Bradbury, 1995) emphasizes the interplay between personal characteristics and stressful life circumstances affecting marital outcomes and provides a framework for understanding marital changes over the life course. Specifically, spouses’ enduring vulnerabilities (e.g., neuroticism) and life experiences (e.g., economic hardship) shape their adaptive processes (e.g., marital interactions), which may have negative consequences, such as declining global marital quality and poor health (Bookwala, 2005; Karney & Bradbury, 1995; Proulx, Helms, & Buehler, 2007). Moreover, family systems theory (Constantine, 1986; Larson et al., 1994) suggests that husbands’ and wives’ experiences and outcomes occur in a context of mutual influences and mutual interactions. This is supported by previous studies suggesting that the dyadic process of marital hostility and escalating hostility, in particular, can have mental health consequences for both husbands and wives.

Figure 1.

Figure 1

Open in a new tab

Conceptual framework for the current study.

Consistent with the VSAM, variations in behavioral trajectories of marital hostility are not expected to be random; instead, they are explained, in part, by specific factors, including personal characteristics and life course experiences (Karney & Bradbury, 1995; Larson et al., 1994; Wickrama, Bryant, & Wickrama, 2010; Wickrama, Surjadi, & Bryant, 2011). The VSAM specifically identifies family economic hardship as one stressful life experiences that is likely to impact marital behavior trajectories. Similarly, neuroticsm is a personal characteristic identified in the VSAM for its impact on marital behaviors. A phenotype of personal-trait neuroticism is neurotic vulnerability, which represents the psychological vulnerability component of neuroticism (Costa & McCrae, 1992) and is a focus on the current study. Consequently, we examine how both family economic hardship and the neurotic vulnerability of each spouse influence couple-level marital hostility. Utilizing a SEM framework offers the flexibility needed to incorporate a second-order (couple-level) growth curve to assess trajectories of husbands’ and wives’ hostile marital behaviors capturing both continuity (i.e., the level/severity) and change in couple-level hostility over time.

Moreover, we posit that couple members’ neurotic vulnerabilities can interactively influence marital hostility, resulting in acceleration (or deceleration) of the hypothesized life course stress process leading to poor mental health of both husbands and wives. That is, neurotic vulnerability of an individual is expected to be exacerbated by the neurotic vulnerability of their spouse (and vice versa) leading to more emotional and intense responses to life stressors (e.g., economic hardship and marital hostility) for both spouses. This interaction of couple members’ neurotic vulnerability, which we consider as a couple characteristic and hereafter refer to as couple’s neurotic vulnerability, was captured by a product term of husbands’ and wives’ neurotic vulnerability. In turn, as shown in Figure 1, we hypothesize trajectories of marital hostility contribute to subsequent poor mental health (i.e., depressive symptoms) of both partners.

This conceptual framework illustrated in Figure 1, drawing heavily from existing marriage theories represents a single, relatively comprehensive, life course stress process model taking a “long-view” over middle years (e.g., Elder, 2001; Pearlin et al., 2005). Our analysis of this framework addresses the limitations of much of the previous empirical research, which was unable to investigate these constructs together (i.e., stressful life experiences, enduring vulnerabilities, marital process, and mental health outcomes) in a single analytical framework considering dyadic effects over a long-period. Instead, past studies have been fragmented and investigated associations between selected constructs at a single time point. Consequently, spurious or limited-duration, short-term associations between study constructs could not be ruled out.

Furthermore, in addition to assessing the additive influences of individuals’ and couples’ neurotic vulnerabilities on marital hostility and mental health, we argue that the hypothesized associations among study constructs are moderated or amplified by couples' neurotic vulnerability. For couples with high scores of neurotic vulnerability, we expect that the hypothesized associations between family economic hardship, marital hostility, and husbands’ and wives’ depressive symptoms may be amplified (see Figure 1). Previous research has not investigated this potential amplifying effect of couple’s vulnerability on the associations involving economic hardship, marital hostility, and health outcomes. We discuss each of the study constructs and hypothesized paths in the paragraphs that follow.

Neurotic Vulnerability and Marital Hostility

As one of the “Big Five” personality traits, individuals high in trait neuroticism experience emotional instability and exhibit above average levels of anger, sadness, and worry (Costa & McCrae, 1992). Thus, in general, individuals with high levels of trait neuroticism are prone to psychological distress. However, trait neuroticism has five unique facets, or phenotypes, including angry hostility, depression, self-consciousness, impulsiveness, and vulnerability (Costa & McCrae, 1992). It is plausible that these facets are differentially related to various relational and health outcomes (Vassend, Orvik, Czajkowski, & Røysamb, 2016). In our examination of the impact of enduring personal vulnerability, we focus on the vulnerability facet of trait neuroticism, termed neurotic vulnerability (Costa & McCrae, 1992).

Research has shown that more vulnerable individuals (i.e., high scores for neurotic vulnerability) are more sensitive to stressful events and circumstances and exhibit amplified stress responses (e.g., they often panic easily when faced with a stressor and/or feel overwhelmed and unable to effectively manage stress) (Costa & McCrae, 1992). We expect that individuals with more neurotic vulnerability are less likely to manage stressful situations such as day-to-day economic difficulties in a calm manner and are more likely to become upset or agitated, which can surface as disruptive or hostile behavior within the marital relationship. On the other hand, individuals with less neurotic vulnerability are typically emotionally stable and even-tempered; as such, they may be more likely to handle stressful situations without expressing hostility and may manage stress in a manner that reduces the expression of marital hostility.

Previous research has shown that trait neuroticism is associated with the initial level of marital satisfaction (Donnellan, Conger & Bryant, 2004; Karney & Bradbury, 1997) as well as marital stability (Wickrama, Bryant, Conger, & Brennom, 2004) but not with changes in these marital attributes over time. Marital violence research has also documented that trait neuroticism is related to increased risk of perpetrating inter-personal violence (Hellmuth &McNulty, 2008). In general, these findings suggest that individuals with neurotic tendencies have marital process trajectories that are consistently unhealthy (i.e., related to the level of marital trajectories). Consequently, we hypothesize that neurotic vulnerability influences the level of hostile marital behavior trajectories.

The Interaction of Spouses’ Neurotic Vulnerabilities: A Couple Characteristic

Family systems theory (Constantine, 1986; Larson et al., 1994) suggests that an individual’s neurotic vulnerability can impact the neurotic vulnerability of his/her partner. That is, as previously discussed, the effect of being neurotically vulnerable may be exacerbated when their partner also has a high level of neurotic vulnerability. Furthermore, research supports the notion of personality convergence over time (Gonzaga et al., 2007) such that neurotic tendencies of a spouse may invoke or amplify similar tendencies in their partner and vice versa. We posit that this reciprocal within-couple process may increase the effect of individual vulnerabilities on numerous marital processes, including marital hostility, and may also directly contribute to the mental health of both spouses.

Neurotic Vulnerability and Poor Mental Health

Neurotic tendencies arise early in life and are one of the most established enduring vulnerability factors for subsequent depression (Kotov, Gamez, Schmidt, & Watson, 2010). Specifically, stress reactivity, which is captured by neurotic vulnerability, is an important risk factor for the development of depressive symptoms (Robinson & Alloy, 2003). In the present study, we expect that both husbands’ and wives’ neurotic vulnerability will be influential for their depressive symptoms a decade later, over and above the effects of marital hostility, even after controlling for the lagged effect of previous depressive symptoms. This may be attributed to the fact that neurotic vulnerability exerts a cumulative influence on depressive symptoms over the life course though other stress-related (mostly non-marital) mechanisms not captured in the present study.

Family Economic Hardship Trajectories and Marital Hostility

Previous experiences with economic problems, although potentially resolved, could have persistent effects on family well-being through several pathways. Chronic economic difficulties or stressful changes in economic well-being may lead to an accumulation of marital disagreements and conflicts with long-term marital consequences (Conger et al., 1999; Vinokur, Price, & Caplan, 1996). Also, consistent with spillover perspective, stressful experiences in other life domains can affect the marital relationship (Rogers & May, 2003). Theories of self-regulation suggest that the depletion of self-regulatory resources largely accounts for these spillover effects (Baumeister, 2002). That is, stressful family economic conditions erode spouses’ self-regulatory resources, thus compromising their energy and ability to manage relationship issues and conflicts and, ultimately, resulting in destructive marital interactions (Buck & Neff, 2012). Following this logic, we expect that, in the present study population who experienced the wake of the farm crisis in the late 1980’s as middle-aged mothers and fathers with adolescent children and aging parents, high levels of economic hardship will continue to negatively influence hostile marital behavior.

In contrast to the level of economic hardship (i.e., severity), economic hardship may be acute or transient. Recent research suggests that acute instances of economic hardship (e.g., sharp increases) are as detrimental for the well-being of family members as chronic economic hardship due to the stress of managing a decline in the family economy (Yuan, 2008; Wickrama et al., 2010). Thus, we expect that both the level and rate of change in family economic hardship trajectories will be associated with couples’ hostile behavior trajectories even after controlling for their neurotic vulnerability.

Mental Health Consequences of Marital Processes

The adverse influence of marital hostility and marital discord on husbands’ and wives’ mental health has been documented in previous research (Proulx, Buehler, & Helms, 2009; Whisman et al., 2015; Wickrama et al., 2010). However, as previously indicated, most of the research has focused on relatively short intervals in the life course. As a long-term process, marital hostility can be a source of chronic stress. In addition, marital hostility deprives couples of spousal warmth and support, which is important because the romantic relationship serves as the most intimate psychological, emotional, and physical source of support available to most adults (Bowlby, 1982). This centrality of the marital relationship may be particularly true for the “empty nester” couples comprising the current study. Thus, we expect that the association between marital hostility and husbands’ and wives’ mental health may by stronger than findings from previous studies of relatively younger couples over shorter periods of time. As such, we assess marital hostility trajectories with a second-order latent growth curve capturing both partners’ hostile behaviors across three time points. Furthermore, we consider their contribution to the generation of depressive symptoms after accounting for the effects of neurotic vulnerability and family economic difficulties.

Moderating Effect of Couples’ Neurotic Vulnerability

Furthermore, we expect that previously discussed mechanisms connecting family economic hardship, marital hostility, and depressive symptoms are moderated by the level of couples’ neurotic vulnerability (a product term computed from husbands’ and wives’ neurotic vulnerability). That is, in marriages with high couple neurotic vulnerability, stressful life experiences, such as economic hardship, are expected to elicit more intense behavioral and emotional responses resulting in a hostile marital context, which is expected to have a more detrimental effect on subsequent mental health. Thus, we hypothesize that associations between family economic hardship, marital hostility, and husbands’ and wives’ mental health outcomes in midlife will be amplified for those with high couple neurotic vulnerability. On the contrary, these associations are hypothesized to be weakened by low levels of couple neurotic vulnerability.

Method

Sample

The data used to evaluate these hypotheses are from the Iowa Midlife Transitions Project (MTP). This study is a prospective, longitudinal study of 451 white European American families from eight counties in rural Iowa (see Conger & Elder, 1994). The MTP originated from the Iowa Youth and Families Project (IYFP) in 1989, 1990, 1991, 1992, and 1994 before continuing as the MTP in 2001. The IYFP studied target adolescents and their families in north central Iowa. For both projects, trained field interviewers visited the families in their homes. During the visit, a trained interviewer asked each family member to complete a detailed questionnaire about family life, work, finances, friends, and mental and physical health status. Family members completed the questionnaires independently so that they could not see one another’s answers. All procedure were approved by the university’s institutional review board.

In 1991, the median age was 41 and 36 for husbands and wives, respectively. The median yearly income was $22,000 and $10,000 for husbands and wives, respectively. The median education for both husbands and wives was 13 years. On average, the couples had been married for 19 years and had three children from their marriage together. The median age of the youngest child was 12. Detailed information about the MTP and IYFP can be found in Conger and Conger (2002) and Conger and Elder (1994). Data for this study used 370 families who provided data at four time points (1991, 1992, 1994, and 2001).

We performed an attrition analysis to examine possible differences between the analyzed sample and those who were not included in the analyses (451–370). The education level of those who were not in the analysis was slightly lower than that of those who remained in the study. Those who were in the analysis and those who were not in the analysis were not significantly different in 1991 in their economic hardship, neurotic vulnerability, or depressive symptoms.

Measures

Neurotic Vulnerability

Utilizing the NEO-PI personality inventory, the psychological vulnerability facet of the neuroticism personality factor was assessed by the mean of 8 items (Costa & McCrae, 1992) in 1991. Items were rated on a 5-point scale ranging from 1 (strongly disagree) to 5 (strongly agree). Both husbands and wives independently responded to statement such as “I can handle myself pretty well in a crisis,” “It’s often hard for me to make up my mind,” and “When I’m under a great deal of stress, sometimes I feel like I’m going to pieces.” Responses were reverse scored when necessary so that higher scores reflect more neurotic vulnerability. The internal consistencies for this measure were .84 and .86 for husbands and wives, respectively. A couple-level measure of neurotic vulnerability was created by computing a product term (i.e., statistical interaction) from husbands’ and wives’ neurotic vulnerability scores after centering and transforming all values to a positive-scale beginning at zero (i.e., by adding the absolute number of the lowest negative value to all scores).

Family Economic Hardship

The list of economic problems was adapted from Dohrenwend, Askenasy, Krasnoff, and Dohrenwend (1978) to capture families’ economic circumstances. Separately for 1989, 1990, 1991, husbands’ and wives’ ”yes” responses to each of the 27 items were summed to indicate economic problems experienced by the family (1 = yes, 0 = no). The list of economic problems included items such as “received government assistance,” “borrowed money to help pay bills,” “sold possessions or cashed in life insurance,” “changed food shopping or eating habits to save money,” and “sold property to raise money.” Husbands’ and wives’ scores were then summed to create a family-level index of economic hardship.

Marital Hostility

In 1991, 1992, and 1994, each partner reported on their spouse’s expression of hostility. These assessments of perceived hostile interactions were developed for the IYFP (Matthews, Wickrama, & Conger, 1996). Sample hostility items include seven items which asked how often in the past year their spouse: “Got angry at you,” “shouted or yelled,” “insulted or swore at you,” and “criticized you or your ideas.” The items were rated using a 7-point Likert scale ranging from 1 = Never to 7 = Always with higher scores representing more hostile behavior. The internal consistencies for the resulting hostility measures ranged from .80 to .85 for husbands and wives across the years.

Depressive Symptoms

Nine items from the SCL90 (Derogatis & Melisaratos, 1983) were used to capture the self-reported ratings of depressive symptoms from the previous week for husbands and wives in 1991 and 2001. Sample items include, “Thoughts of ending your life,” “feelings of worthlessness,” and “feeling hopeless about the future.” These items were scored on a 5-point Likert scale from 1 = Not at all to 5 = Extremely with higher scores indicating more depressive symptoms. The internal consistencies for all depressive symptoms measures ranged from .80 to .91 for husbands and wives.

Analysis

Analyses were performed using structural equation modeling (SEM) with Mplus software (Muthen & Muthen, 1998–2010). A growth curve was estimated to assess the initial level and rate of change in family economic hardship with measurement occasions in 1989, 1990, and 1991. We also estimated latent growth curves for husbands’ and wives’ marital hostility (1991, 1992, and 1994). High correlations between the growth factors for husbands’ and wives’ marital hostility after taking contemporaneous correlations between husband and wife measurement errors into account suggested the existence of a higher-order growth construct capturing marital hostility at the couple-level even. Thus, growth curves of husbands’ and wives’ hostility were modeled as primary factors of a second-order growth curve calculating the initial level and rate of change in marital hostility as a dyadic construct. This approach is known as a factor-of-curves model (FCM; Duncan, Duncan, & Strycker, 2006; Wickrama, Lee, O’Neal, & Lorenz, 2016). These second-order growth parameters (i.e., initial level and slope) of marital hostility were examined as mediators between depressive symptoms outcomes and hypothesized antecedents, including family economic hardship growth parameters and neurotic vulnerabilities (i.e., husbands’ and wives’ vulnerability as well as couple-level neurotic vulnerabilities) after controlling for earlier levels of depressive symptoms. Controlling for earlier levels of depressive symptoms (1991) allows model paths to be interpreted as change in individuals’ depressive symptoms over time. Husbands’ and wives’ reports of neurotic vulnerability were allowed to correlate as were their depressive symptoms at both time points (i.e., 1991 and 2001).

In addition to assessing the interaction of spouses’ neurotic vulnerabilities as a product term capturing husbands’ and wives’ neurotic vulnerability, the moderating role of couple neurotic vulnerability on associations between family economic hardship, marital hostility, and depressive symptoms was assessed using a mean-split to create two groups of couples (high and low couple vulnerability). A model incorporating economic hardship trajectories, hostility trajectories, and depressive symptoms was then assessed separately for these two groups. We relied on a range of indices to evaluate the fit of our conceptual model, including the chi-square statistic, the comparative fit index (CFI), and the root mean square error of approximation (RMSEA). Bentler (1990) states that a CFI of more than .95 indicates a respectable model fit, and MacCallum, Browne, and Sugawara (1996) report that a RMSEA of .08 indicates reasonably good model fit as well.

Results

Descriptive statistics and bivariate correlations among study variables are shown in Table 1. The average economic hardship value averaged below the midpoint of the scale (m = 10.28 out of a possible 27 in 1989), yet these items reflect relatively notable instances of economic hardship (e.g., receiving government assistance), indicating a sizable amount of hardship among sample participants. Nevertheless, participants averaged below-midpoint levels of neurotic vulnerability, marital hostility, and depressive symptoms. Thus, the study sample couples were functioning relatively well.

Table 1.

Univariate and bivariate statistics for all study variables.

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16
1. H. Neur. Vulnerability --
2. W. Neur. Vulnerability .02 --
3. Couple Vulnerability .64*** .65*** --
4. Econ. Hard. (89) .29*** .13* .32*** --
5. Econ. Hard. (90) .18*** .14* .23*** .72*** --
6. Econ. Hard. (91) .18*** .14** .21*** .54*** .66*** --
7. H. Percept. W. Hostility (91) .14** .25*** .26*** .22*** .17** .12* --
8. H. Percept. W. Hostility (92) .08 .18*** .18*** .12* .14* .05 .76*** --
9. H. Percept. W. Hostility (94) .15** .21*** .22*** .17** .16** .15** .69*** .66*** --
10. W. Percept. H. Hostility (91) .13* .39*** .35*** .26*** .14** .20*** .58*** .46*** .43*** --
11. W. Percept. H. Hostility (92) .09 .30*** .26*** .17*** .12* .15** .52*** .52*** .42*** .79*** --
12. W. Percept. H. Hostility (94) .11 .31*** .25*** .13* .07 .16** .44*** .40*** .50*** .64*** .60*** --
13. H. Dep. Sym. (91) .50*** .08 .37*** .36*** .26*** .17*** .24*** .07 .15** .18*** .10 .12* --
14. H. Dep. Sym. (2001) .40*** .18*** .42*** .21*** .14** .16** .16** .14* .14* .07 .09 .07 .41*** --
15. W. Dep. Sym. (91) .00 .51*** .38*** .28*** .24*** .21*** .29*** .17** .23*** .49*** .37*** .29*** .21*** .11* --
16. W. Dep. Sym. (2001) .05 .34*** .24*** .14** .09 .25*** .05 .08 .07 .16** .18*** .18*** −.02 .17** .44*** --
Mean 2.20 2.36 .56 10.28 10.10 8.63 2.16 2.12 2.24 2.07 2.08 2.14 1.31 1.42 1.51 1.53
Std. Dev. .45 .50 .55 9.31 8.42 7.18 .75 .72 .71 .85 .82 .71 .45 .44 .57 .50
Open in a new tab

The univariate growth curve for family economic hardship from 1989 to 1991 fit the data well (χ2(df) = 1.69(1), CFI = .99, RMSEA = .04). The average initial level of economic hardship was 10.45 (p < .001), and across the sample, there was a general decrease in hardship over time (the mean rate of change of −0.60, p < .001), which suggests, in general, the families were recovering from the economic burden of the farm crisis. However, there was significant variability (p < .001) in the initial level and rate of change in economic hardship. Thus, some families may have experienced increases while others experienced decreases in economic hardship over time. That is, higher values for rate of change in economic hardship can be conceptualized as either a failure to encounter decline in economic hardship or an increase in economic hardship over time. A negative correlation coefficient between the initial level and slope represents a regression to the mean with families experiencing the most economic hardship reporting lower slopes than other families (r = −.58, p < .001).

We initially estimated a parallel process model comprised of two growth curves for husbands’ and wives’ marital hostility using data from 1991, 1992, and 1994. In this model, the initial levels and slopes were significantly correlated between husbands and wives. Because of the high correlations, we estimated a second-order growth curve (factor-of-curves model, FCM) of marital hostility using husbands’ and wives’ growth curves as indicators after taking contemporaneous correlations between husband and wife measurement errors into account. This second-order model showed a significant improvement over the parallel process model (reduction in chi-square for 3 df was 92.31) and was a good fit to the data (CFI = .98, RMSEA=.07, χ2(df) = 23.12 (9). Thus, we used the couple-level hostility growth curve to test our conceptual model. The average initial level of marital hostility was .94 (p < .001), and across the sample, there was a marginally significant general increase in marital hostility over time (mean slope of .09, p = .09). There was significant variability (p < .001) in the initial level and rate of change in marital hostility suggesting that some couples experienced an increase in hostility from 1991 to 1994 and others experienced a decrease in hostility.

Testing the Hypothesized Model

Next, we assessed the conceptual model presented in Figure 1 using a SEM with a latent growth curve capturing family economic hardship and a second-order FCM capturing marital hostility (see Figure 2). Model fit indices were: χ2(df) = 190.79(72), CFI = .94, RMSEA = .07, and the model accounted for 12.8% and 23.7% of the variation in husbands’ and wives’ depressive symptoms, respectively, in 2001. Both individual and couple neurotic vulnerability (the statistical interaction of spouses’ neurotic vulnerability) were associated with the initial level and rate of change in family economic hardship. More specifically, husbands’ and wives’ with more neurotic vulnerability reported higher average initial levels of economic hardship in 1989 (r = .33 and .26, respectively), but individuals with more neurotic vulnerability generally indicated less steep rates of change in economic hardship over time (r = −.38 and −.34 for husbands and wives, respectively), which indicates a potential regression to the mean over time. Couple neurotic vulnerability was not significantly associated with the initial level of economic hardship (r = −.13), but it was associated with the rate of change in economic hardship over time (β = .24).

Figure 2.

Figure 2

Open in a new tab

A SEM assessing individual and couple neuroticvulnerability, family economic hardship, marital hostility, and husbands’ and wives’ depressive symptoms.

Note: Only statistically significant paths are shown. Standardized coefficients are presented. The slope of economic hardship was significantly correlated with husband, wife, and couple vulnerability (r = −.38, −.34, and .24) (not shown in the figure).

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

Both neurotic vulnerability and economic hardship were related to the marital hostility. On average, wives’ neurotic vulnerability (measured in 1991) was related to the initial level of marital hostility (measured in 1991) (β = .41). That is, couples averaged a higher initial level of marital hostility when wives reported more vulnerability. Couple-level vulnerability was also related to a higher initial level of marital hostility (β = .29). Regarding economic hardship, couples with a higher initial level of economic hardship (in 1989), on average, experienced higher initial levels of marital hostility (in 1991) compared to couples with less economic hardship (β = .20). Similarly, couples with a faster rate of change in economic hardship from 1989–1991 tended to report steeper slopes for marital hostility (β = .47).

In turn, the rate of change in marital hostility was relevant for subsequent depressive symptoms of both spouses with a greater rate of change for marital hostility from 1991 to 1994 linked to more depressive symptoms in 2001 (husbands β = .14; wives β = .15). After accounting for the mediating role of hostility, neurotic vulnerability was also directly associated with depressive symptoms. Husbands’ and wives’ neurotic vulnerability as well as couple-level vulnerability in 1991 were all related to husbands’ reporting more depressive symptoms in 2001, on average (β = .26, .20, and .27, respectively). For wives, their neurotic vulnerability and the couple vulnerability was linked to their depressive symptoms in 2001 (β = .19 and .20, respectively). These findings pertaining to depressive symptoms in 2001 existed after accounting for husbands’ and wives’ depressive symptoms in 1991 (i.e., we predicted residualized changes in depressive symptoms in 2001). As expected, depressive symptoms in 1991 were predictive of symptoms in 2001 (β = .16 and .44 for husbands and wives, respectively).

Bootstrapping (available in Mplus) was utilized to provide evidence of the indirect effects of the rate of change in economic hardship (1989 to 1991) on both husbands’ and wives’ depressive symptoms in 2001 through the rate of change in marital hostility. The indirect effect was statistically significant for both husbands (α*β = .07, p < .05) and wives (α*β = .07, p < .05). Thus, while the rate of change in economic hardship was directly associated with both spouses’ subsequent depressive symptoms, it appears that the rate of change in economic hardship may also have an ancillary effect on depressive symptoms through associations with marital hostility.

A number of correlations were also taken into account within the model illustrated in Figure 2; for simplicity, some of these are not shown in Figure 2, but are summarized here. The level of economic hardship was also negatively correlated with the average rate of change in economic hardship (r = −.68, p < .001). Husbands’ and wives’ neurotic vulnerabilities were also significantly correlated (r = .50, p < .001) as were there depressive symptoms in 1991 (r = .40, p < .001). While their depressive symptoms were allowed to correlate in 2001, these values were not statically significant. The initial level and rate of change for marital hostility were negatively associated (r = −.52, p < .001).

We also tested a fully recursive model by expanding on the model shown in Figure 2 to incorporate missing paths from vulnerability constructs and economic hardship constructs to hostile behaviors and depressive symptoms. This fully recursive alternative model did not show an improvement in model fit, which suggests the more parsimonious model presented in Figure 2 is a better fit to the data.

Interpreting the Moderation of Couple Vulnerability

Last, we assessed whether couple neurotic vulnerability moderates the associations between economic hardship, marital hostility, and depressive symptoms. Because the interaction term is approximately normally distributed, the mean-split is an appropriate and meaningful cut-point of high and low neurotic vulnerability. In summary, a comparison of the model separately for couples with high and low couple neurotic vulnerability indicated an absence of detrimental effects of economic hardship and hostility on subsequent depressive symptoms for those with low couple neurotic vulnerability (i.e., couples with below average neurotic vulnerability) (see Figure 3). However, for couples with higher neurotic vulnerability scores, economic hardship and hostility were positively related to depressive symptoms. For instance, the initial level of economic hardship in 1989 was positively related to the initial level of marital hostility in 1991 for couples with high neurotic vulnerability (β = .49), whereas this effect was not present for those with low couple-level neurotic vulnerability (β = .01). Similarly, among couples with above average levels of neurotic vulnerability, a faster rate of change for economic hardship from 1991 to 1994 was associated with a faster rate of change in marital hostility from 1991 to 1994 (β = .95), but this association was not found for those with low couple-level neurotic vulnerability (β = .00). These group differences were statically significant.

Figure 3.

Figure 3

Open in a new tab

A SEM assessing the moderating role of couple-level vulnerability on the hypothesized associations.

Notes. *p<.05.*p< .01.*p<.001.Standardized coefficients are presented for the high vulnerability couples with coefficients for the low vulnerability couples in parentheses.

In the high couple vulnerability condition, the rate of change in marital hostility was associated with more depressive symptoms for both husbands and wives in 2001 (β = .23 and .27, respectively) but these associations were not statistically significant in the low vulnerability condition (β = .12 and .03 for husbands and wives, respectively). Low levels of couple neurotic vulnerability appear to buffer husbands from the detrimental effects of early depressive symptoms (measured in 1994) on subsequent depressive symptoms (measured in 2001) (β = .41 in the high condition and .05 in the low condition). However, this buffering effect was not evident for wives as depressive symptoms in 1994 were associated with their symptoms in 2001 regardless of couple-level neurotic vulnerability (β = .57 and .51 for the high and low conditions, respectively).

Discussion

Although previous research has shown a robust association between hostile marital processes and poor mental health (Proulx et al., 2009; Whisman et al., 2014; Wickrama et al., 2010, 2011), the association has not been adequately investigated within broader contexts or when considering the potential long-term effects in enduring relationships. Thus, largely drawing from contemporary theories of marriage, such as the vulnerability-stress adaptation model (VSAM; Karney & Bradbury, 1995), the present study examined a conceptual framework taking a long-view to examine how neurotic vulnerability, family economic hardship trajectories, and a hostile marital process combine to influence depressive symptoms over middle adulthood. Given the historical economic environment (i.e., a sample of couples experiencing relatively extreme economic hardship due to the farm crisis), we expected that hypothesized life course stress process model would explain later mental health inequalities.

Within the life course stress process model, four specific hypotheses were investigated. First, hostile marital behavior trajectories, as indicators of a hostile marital process, contribute to changes in both spouses’ depressive symptoms. Second, individuals’ neurotic vulnerability, as an enduring vulnerability, and couple-level neurotic vulnerability directly contribute to changes in husbands’ and wives’ depressive symptoms after accounting for economic hardship and hostile behavior trajectories. The third hypothesis is that marital hostility trajectories link neurotic vulnerabilities and trajectories of family economic hardship to subsequent changes in depressive symptoms, and the fourth hypothesis is that associations between family economic hardship trajectories, marital hostility trajectories, and subsequent depressive symptoms are moderated by couple-level neurotic vulnerability. In general, the results of the present study provided evidence for these expected associations. Furthermore, the results validate and strengthen conclusions drawn from existing cross-sectional and short-term longitudinal studies on stressful life events, marital processes, and mental health outcomes. Moreover, the findings serve to illustrate that these processes do, in fact, occur through long-term processes and are at play regardless of the enduring nature of marriages.

More specifically, for both husbands and wives, the rate of change in marital hostility from 1991 to 1994 was associated with their depressive symptoms after accounting for earlier levels of depressive symptoms, individual- and couple-level neurotic vulnerability, and economic hardship. This may be attributed to the fact that inter-personal hostile behavior and the manifestation of depressive symptoms are comorbid over time for both husbands and wives. That is, changes in the hostile couple context may represent a chronic stressor with emotional consequences for both spouses. Moreover, these hostile marital processes may be of consequence for depressive symptoms because they deprive couples of the psychological, emotional, and physical support provided by their spouse.

Results from the current study suggest that both individual- and couple-level neurotic vulnerabilities contribute to husbands’ and wives’ depressive symptoms after taking their marital hostility into account. It appears that neurotic vulnerability can be an important risk factor for the development of depressive symptoms. That is, individuals who possess negative tendencies may experience more depressive symptoms, possibly because their vulnerability elicits stronger emotional responses to negative life events and circumstances. These results also suggest that, in the middle years, husbands and wives may be exposed to numerous non-marital stressful events and circumstances with emotional consequences. More importantly, as evidence of the dynamic nature of the family system (Larson et al., 1994), the results indicate that neurotic vulnerability at the couple-level may contribute to husbands’ and wives’ depressive symptoms above and beyond the influence of each spouse’s neurotic vulnerability. This provide evidence for the mutual influences between husbands and wives in relation to neurotic vulnerabilities.

The results partially supported the fourth hypothesis because hostility, as a marital process, linked neurotic vulnerabilities to subsequent changes in mental health. We argued that the marital processes of these middle-aged couples include elements of stability but also change; the change comes from several sources, such as individual differences and life course experiences, including changes in family economic circumstances. Previous research suggests that neurotic vulnerabilities increase individuals’ risk for chronically high levels of hostility (representing a stability component) but does not affect the shape of these trajectories over time (change component) (Hellmuth & McNulty, 2008). These previous findings were supported by the current findings indicating statistically significant associations with the level of marital hostility, but not the rate of change in hostility over time. The results of the present study also showed that husbands’ neurotic vulnerability was not associated with marital hostility (after accounting for economic hardship and prior depressive symptoms), but wives’ neurotic vulnerability and couple-level neurotic vulnerability was linked to higher initial levels of hostility. Future research should extend this line of research to investigate how other couple-level vulnerabilities are related to changes in marriage and mental health over time.

Importantly, change in family economic hardship was related to the slope of marital hostility. That is, couples who experienced increases in family economic hardship generally reported increases in hostile behaviors as well. However, the level of economic hardship was not related to the rate of change in marital hostility. It seems that during their middle years husbands and wives are potentially more susceptible to changes in economic conditions than to the level (severity) of economic conditions. The present study revealed these dynamic associations using a growth curve analytical approach. This analytical approach provides additional insight into the processes at play in the self-regulation depletion hypothesis (Baumeister, 2002); in that, it is not simply the level of stressful family conditions that ultimately result in destructive marital interactions, but rather the change in stressful family conditions is particularly taxing for individuals’ self-regulatory resources that often serve to minimize the likelihood of negative marital exchanges.

Previous research has rarely investigated the moderating role of enduring vulnerabilities (e.g., neurotic vulnerability) in relation to marriage and health. The results of the present study provide evidence for the potential moderating role of couples’ neurotic vulnerability. These results suggest that the combination (i.e., a product term) of husbands’ and wives’ neurotic vulnerabilities (thereby reflecting a couple characteristic) form a vulnerability at the couple level that can be detrimental to the marriage. Thus, although the VSAM articulates enduring resources, such as individuals’ neurotic vulnerability, as a determinant of marital behavior (Karney & Bradbury, 1995), it appears that the combination of each partner’s neurotic vulnerability create a context that may not only influence interactions, but also intensify the emotional stress reactivity for both husbands and wives, conferring feelings of distress, including depressive symptoms. Particularly, during their middle years, a time that is dense in life transition events, couple interactive vulnerability may intensify emotional responses to stressful marriage because marital relationships are more salient for husbands and wives in later middle years. Furthermore, we posit that stress vulnerability may erode psychological resources, such as mastery and self-control, which have been shown to protect individuals from stress exposure and adversity (Pearlin et al., 2005). Future studies should extend on the current findings by considering other moderators of the processes identified, such as developmental markers capturing unique features of these enduring marriages. Furthermore, given that the transitions couples often experience in midlife is a central developmental marker of the midlife stage, future research should also explicitly assess the impact, either directly or as moderator, of particular midlife transition events (e.g., becoming empty nesters) on the processes identified.

There are several limitations to the current study that should be noted. First, the sample was comprised only of European-American individuals living in rural Iowa. Studies testing similar models with a more diverse population are needed. For instance, future samples should include multiple ethnicities, greater variation in length of marriage, and other geographic locations. Second, the historical and economic environment (i.e., the farm crisis) in which the respondents lived represents a rather unique context. Thus, future studies should investigate the study hypotheses involving family economic hardship and health with respondents who have experienced less extreme economic downturns. Third, some of study constructs (e.g., family economic hardship) were self-reported and, consequently, self-report biases are a possibility. Finally, because genetic research suggests that mental health influences of stressful experiences are moderated by individuals’ genotype, future studies should investigate potential genetic moderations of the observed associations.

Despite these limitations, the present study contributes to existing knowledge about the long-term combined influences of psychological and socioeconomic factors (i.e., enduring personal and couple vulnerabilities and economic hardship) on changes in marital processes and mental health outcomes for long-term married couples. The results suggest that personal and couple vulnerability (particularly neurotic vulnerability) and life experiences in the early years of middle adulthood can have implications for mental health in the later middle years; these implications may exist both directly and indirectly through couples’ marital processes, namely marital hostility in the current study. Thus, mental health services and marital therapy focusing on persistent influences of individual and couple vulnerabilities and life experiences may prove valuable, particularly when the relationship between these factors and marital changes are explicated. Furthermore, these results highlight the possible amplification of mental health effects due to the combination of both spouses’ neurotic vulnerabilities (which we label as couple-level vulnerability). These findings are particularly important for future family interventions and for counselors as they consider ways to protect individuals from the negative mental health consequences of adverse marital changes.

Acknowledgments

This research is currently supported by a grant from the National Institute on Aging (AG043599, Kandauda A. S. Wickrama, PI). The content is solely the responsibility of the authors and does not necessarily represent the official views of the funding agencies. Support for earlier years of the study also came from multiple sources, including the National Institute of Mental Health (MH00567, MH19734, MH43270, MH59355, MH62989, MH48165, MH051361), the National Institute on Drug Abuse (DA05347), the National Institute of Child Health and Human Development (HD027724, HD051746, HD047573, HD064687), the Bureau of Maternal and Child Health (MCJ-109572), and the MacArthur Foundation Research Network on Successful Adolescent Development Among Youth in High-Risk Settings.

Footnotes

An earlier version of this manuscript was presented at the National Council on Family Relations at the 2017 annual conference in Minneapolis, Minnesota.

Contributor Information

Kandauda (K.A.S.) Wickrama, Department of Human Development and Family Science, The University of Georgia, 107 Family Science Center I (House A), Athens, GA 30602, wickrama@uga.edu

Catherine Walker O’Neal, Department of Human Development and Family Science, The University of Georgia, 107 Family Science Center II (House D), Athens, GA 30602, cwalker1@uga.edu, 706-542-4922

Frederick O. Lorenz, Department of Statistics, Iowa State University, 1415 Snedecor Hall, Ames, IA 50011, folorenz@iastate.edu, 515-294-7531

References

  1. Baumeister RF. Ego depletion and self-control failure: An energy model of the self's executive function. Self and Identity. 2002;1(2):129–136. [Google Scholar]
  2. Bentler PM. Comparative fit indexes in structural models. Psychological Bulletin. 1990;107(2):238–246. doi: 10.1037/0033-2909.107.2.238. [DOI] [PubMed] [Google Scholar]
  3. Berg CA, Johnson MM, Meegan SP, Strough J. Collaborative problem-solving interactions in young and old married couples. Discourse Processes. 2003;35(1):33–58. [Google Scholar]
  4. Bookwala J. The role of marital quality in physical health during the mature years. Journal of Aging and Health. 2005;17(1):85–104. doi: 10.1177/0898264304272794. [DOI] [PubMed] [Google Scholar]
  5. Bowlby J. Attachment and Loss. 2. Vol. 1. New York: Basic Books; 1982. [Google Scholar]
  6. Buck AA, Neff LA. Stress spillover in early marriage: The role of self-regulatory depletion. Journal of Family Psychology. 2012;26(5):698–708. doi: 10.1037/a0029260. [DOI] [PubMed] [Google Scholar]
  7. Conger RD, Elder GH., Jr . Families in Troubled Times: Adapting to Change in Rural America. Social Institutions and Social Change. Aldine de Gruyter; Hawthorne, NY: 1994. [Google Scholar]
  8. Conger RD, Rueter MA, Elder GH., Jr Couple resilience to economic pressure. Journal of Personality and Social Psychology. 1999;76(1):54–71. doi: 10.1037//0022-3514.76.1.54. [DOI] [PubMed] [Google Scholar]
  9. Constantine LL. Family paradigms. New York: Guilford; 1986. [Google Scholar]
  10. Costa PT, McCrae RR. Normal personality assessment in clinical practice: The NEO Personality Inventory. Psychological Assessment. 1992;4(1):5–13. [Google Scholar]
  11. Derogatis LR, Melisaratos N. The brief symptom inventory: An introductory report. Psychological Medicine. 1983;13(3):595–605. [PubMed] [Google Scholar]
  12. Dohrenwend BS, Askenasy AR, Krasnoff L, Dohrenwend BP. Exemplification of a method for scaling life events: The PERI Life Events Scale. Journal of Health and Social Behavior. 1978;19(2):205–229. [PubMed] [Google Scholar]
  13. Donnellan MB, Conger RD, Bryant CM. The Big Five and enduring marriages. Journal of Research in Personality. 2004;38(5):481–504. [Google Scholar]
  14. Duncan TE, Duncan SC, Strycker LA. An introduction to latent variable growth curve modeling. Concepts, issues and applications. 2. Lawrence Earlbaum, Mahwah: 2006. [Google Scholar]
  15. Elder GH., Jr Families, social change, and individual lives. Marriage & Family Review. 2001;31(1–2):187–203. [Google Scholar]
  16. Gonzaga GC, Campos B, Bradbury T. Similarity, convergence, and relationship satisfaction in dating and married couples. Journal of Personality and Social Psychology. 2007;93(1):34–48. doi: 10.1037/0022-3514.93.1.34. [DOI] [PubMed] [Google Scholar]
  17. Hellmuth JC, McNulty JK. Neuroticism, marital violence, and the moderating role of stress and behavioral skills. Journal of Personality and Social Psychology. 2008;95(1):166–180. doi: 10.1037/0022-3514.95.1.166. [DOI] [PubMed] [Google Scholar]
  18. Karney BR, Bradbury TN. The longitudinal course of marital quality and stability: A review of theory, method, and research. Psychological Bulletin. 1995;118:3–34. doi: 10.1037/0033-2909.118.1.3. [DOI] [PubMed] [Google Scholar]
  19. Karney BR, Bradbury TN. Neuroticism, marital interaction, and the trajectory of marital satisfaction. Journal of Personality and Social Psychology. 1997;72(5):1075–1092. doi: 10.1037//0022-3514.72.5.1075. [DOI] [PubMed] [Google Scholar]
  20. Karney BR, Bradbury TN, Fincham FD, Sullivan KT. The role of negative affectivity in the association between attributions and marital satisfaction. Journal of Personality and Social Psychology. 1994;66(2):413–424. doi: 10.1037//0022-3514.66.2.413. [DOI] [PubMed] [Google Scholar]
  21. Kotov R, Gamez W, Schmidt F, Watson D. Linking “big” personality traits to anxiety, depressive, and substance use disorders: A meta-analysis. Psychological Bulletin. 2010;136(5):768–821. doi: 10.1037/a0020327. [DOI] [PubMed] [Google Scholar]
  22. Lackman ME, James B. Charting the Course of Midlife Development: An Overview. In: Lachmand ME, Boone-James J, editors. Multiple Paths of Midlife Development. Chicago: University of Chicago Press; 1997. pp. 1–20. [Google Scholar]
  23. Larson JH, Wilson SM, Beley R. The impact of job insecurity on marital and family relationships. Family Relations. 1994;43(2):138–143. [Google Scholar]
  24. Levenson RW, Carstensen LL, Gottman JM. Influence of age and gender on affect, physiology, and their interrelations: A study of long-term marriages. Journal of Personality and Social Psychology. 1994;67(1):56–68. doi: 10.1037//0022-3514.67.1.56. [DOI] [PubMed] [Google Scholar]
  25. Lorenz FO, Conger RD, Montague RB, Wickrama KAS. Economic conditions, spouse support, and psychological distress of rural husbands and wives. Rural Sociology. 1993;58(2):247–268. [Google Scholar]
  26. Lorenz FO, Elder GH, Bao WN, Wickrama KAS, Conger RD. After farming: Emotional health trajectories of farm, nonfarm, and displaced farm couples. Rural Sociology. 2000;65(1):50–71. [Google Scholar]
  27. MacCallum RC, Browne MW, Sugawara HM. Power analysis and determination of sample size for covariance structure modeling. Psychological Methods. 1996;1(2):130–149. [Google Scholar]
  28. Matthews LS, Wickrama KAS, Conger RD. Predicting marital instability from spouse and observer reports of marital interaction. Journal of Marriage and the Family. 1996;58(3):641–655. [Google Scholar]
  29. Miller RB, Hollist CS, Olsen J, Law D. Marital quality and health over 20 years: A growth curve analysis. Journal of Marriage and Family. 2013;75(3):667–680. [Google Scholar]
  30. Pearlin LI, Schieman S, Fazio EM, Meersman SC. Stress, health, and the life course: Some conceptual perspectives. Journal of Health and Social Behavior. 2005;46(2):205–219. doi: 10.1177/002214650504600206. [DOI] [PubMed] [Google Scholar]
  31. Proulx CM, Buehler C, Helms H. Moderators of the link between marital hostility and change in spouses’ depressive symptoms. Journal of Family Psychology. 2009;23(4):540–550. doi: 10.1037/a0015448. [DOI] [PMC free article] [PubMed] [Google Scholar]
  32. Proulx CM, Helms HM, Buehler C. Marital quality and personal well-being: A meta-analysis. Journal of Marriage and Family. 2007;69(3):576–593. [Google Scholar]
  33. Robinson MS, Alloy LB. Negative cognitive styles and stress-reactive rumination interact to predict depression: A prospective study. Cognitive Therapy and Research. 2003;27(3):275–291. [Google Scholar]
  34. Rogers SJ, May DC. Spillover between marital quality and job satisfaction: Long-term patterns and gender differences. Journal of Marriage and Family. 2003;65(2):482–495. [Google Scholar]
  35. Smith TW, Berg CA, Florsheim P, Uchino BN, Pearce G, Hawkins M, Olsen-Cerny C. Conflict and collaboration in middle-aged and older couples: I. Age differences in agency and communion during marital interaction. Psychology and Aging. 2009;24(2):259–273. doi: 10.1037/a0015609. [DOI] [PMC free article] [PubMed] [Google Scholar]
  36. Staudinger UM, Bluck S. A view on midlife development from life-span theory. In: Lachman ME, editor. Handbook of midlife development. New York: Wiley; 2001. pp. 3–39. [Google Scholar]
  37. Vassend O, Orvik BH, Czajkowski NO, Røysamb E. Facets of neuroticism and musculoskeletal symptoms. A study of middle-aged twins. Norsk Epidemiologi. 2016;26(1–2):47–56. [Google Scholar]
  38. Vinokur AD, Price RH, Caplan RD. Hard times and hurtful partners: How financial strain affects depression and relationship satisfaction of unemployedvas persons and their spouses. Journal of Personality and Social Psychology. 1996;71(1):166–179. doi: 10.1037//0022-3514.71.1.166. [DOI] [PubMed] [Google Scholar]
  39. Whisman MA, Robustelli BL, Beach SR, Snyder DK, Harper JM. Marital discord and depression in middle-aged and older couples: Is taxon status associated with depression? Journal of Social and Personal Relationships. 2015;32(7):967–973. [Google Scholar]
  40. Wickrama KAS, Bryant CM, Conger RD, Brennom JM. Change and continuity in marital relationships during the middle years. In: Conger RD, Lorenz FO, Wickrama KAS, editors. Continuity and Change in Family Relations: Theory, Methods, and Empirical Findings. Lawrence Erlbaum, Mahwah, New Jersey: 2004. pp. 123–143. [Google Scholar]
  41. Wickrama KAS, Bryant CM, Wickrama T. Perceived community disorder, hostile marital interactions, and self-reported health of African American couples: An interdyadic process. Journal of Personal Relationships. 2010;17(4):515–531. [Google Scholar]
  42. Wickrama KAS, Lee TK, O’Neal CW, Lorenz FO. Higher-Order Growth Curves and Mixture Modeling with Mplus: A Practical Guide. New York, NY: Taylor and Francis; 2016. [Google Scholar]
  43. Wickrama KAS, Surjadi FF, Bryant CM. Work, marriage, and health behaviors of African American working couples. Journal of Personal Relationships. 2011;18:519–531. [Google Scholar]
  44. Wickrama KAS, Lorenz FO, Conger RD, Elder GH., Jr Marital quality and physical illness: A latent growth curve analysis. Journal of Marriage and the Family. 1997;59(1):143–155. [Google Scholar]
  45. Yuan ASV. Exploring the changes in economic hardship and children's well-being over time: The “linked lives” of parents and children. Advances in Life Course Research. 2008;13:321–341. [Google Scholar]

RESOURCES