Health continuity over mid-later years in enduring marriages: Economic pressure as couple- and individual-level mediator

Abstract

Health problems in midlife have been shown to continue into later adulthood. This continuity may be attributed to social selection and social causation, with longitudinal sequential associations between health problems and economic pressure (EP) over the life course creating mediational effects (health → EP → health). Moreover, in enduring marriages, this potential mediation may operate as a dyadic process over time. Yet this mediational process involving health problems and economic pressure has not been adequately investigated in couple contexts. Using a sample of 257 husbands and wives in enduring marriages, we investigated multilevel mediation processes between health problems and EP from midlife to later adulthood. The results indicate unique couple- and individual-level (only for husbands) mediation processes of health problems in the couple context, emphasizing the value of considering both couple- and individual-level health processes when developing health interventions.

Keywords

Couple contexts economic pressure health problems

Past studies suggest that the incidence of diseases and physical impairment increases notably over the mid-later years (Lorenz, Wickrama, Conger, & Elder, 2006), which contributes to physical health problems in later adulthood (Ohrnberger, Fichera, & Sutton, 2017; hereafter referred as health continuity). In addition, consistent with the social selection perspective (Warren, 2009), life course studies suggest long-term associations between early health problems and later economic hardship (Clarke, O’Malley, Schulenberg, & Johnson, 2010). Consistent with the social causation perspective (Conger & Donnellan, 2007), however, previous studies have demonstrated the influence of economic problems on subsequent physical health (e.g., Wickrama, Kwag, Lorenz, Conger, & Surjadi, 2010). Taken together, these findings suggest economic pressure (EP) may act as a mediator of health continuity from midlife to later adulthood (hereafter referred as health mediational processes; Lee & Jackson, 2017). That is, earlier health contributes to subsequent EP, which, in turn, impacts health later in the life course.

This mediational process aligns with the stress proliferation hypothesis within the life course stress process perspective (Pearlin, Schieman, Fazio, & Meersman, 2005; Thoits, 2010). Stress proliferation refers to a process in which an initial stressor gives rise to a secondary stressor/s in the same stress domain or in additional domains (Pearlin et al., 2005). Stress proliferation may occur in short or long periods of time over the life course (Thoits, 2010). Although a past study investigated the health mediational role of EP at a single time point (Warren, 2009), longitudinal changes (trajectories) in EP have not been considered, which limits our understanding of longitudinal mediational processes that explain health continuity from midlife to later adulthood. Moreover, recent relationship research suggests that individual- and couple-level processes may be uniquely involved (e.g., hybrid models; Galovan, Holmes, & Proulx, 2017). Thus, we posit that health mediational processes may be multilevel with distinct individual- (within-couple contexts) and couple-level (between-couple contexts) processes.

Individual-level mediational processes of health problem

At the individual level, health mediational processes are aligned with the actor–partner interdependence model perspective (APIM; Kenny, Kashy, & Cook, 2006) including two distinct effects: (a) actor effects and (b) partner, or crossover, effects. Consistent with social selection perspective (Warren, 2009), one’s own health problems not only influence his/her EP trajectories (actor effects) but also may influence the partner’s trajectories of EP (partner effects). For example, an individual’s poor health can lead to his/her own EP due to limited workforce participation (actor effects; Leonard, Hughes, & Pruitt, 2017), yet a partner’s poor health may also lead to increasing EP because the individual is likely to reduce their work hours to support their partner (partner effects; Bubonya, Cobb-Clark, & Wooden, 2017).

Similarly, one’s own EP trajectories not only influence his/her own health problems (actor effects) but also may influence the partner’s health problems (partner effects). In terms of actor effects, the social causation perspective (Conger & Donnellan, 2007) highlights how EP, a particularly potent stressor, negatively affects individuals’ health through various psycho-behavioral mechanisms, including the reduced ability to manage health behaviors (e.g., sedentary behavior or unhealthy eating; risky health life style; Lee, Wickrama, & O’Neal, 2019). These behaviors influence the development of numerous cardiometabolic risk factors (e.g., diabetes, hypertension, thickening of arteries, and heart attacks) in mid to late life (actor effects; Wickrama, O’Neal, & Neppl, in press). In addition, individuals’ EP increase their own psychological distress (e.g., anxiety and depressive symptoms; Falconier, 2010), which may also contribute to physical health problems (actor effects; Lee, Wickrama, & O’Neal, 2019).

Moreover, social causation may operate through partner effects. Previous studies suggest that spouses’ daily life activities are intertwined, and one spouse’s individual attributes, such as mood, behavior, and stresses, can affect himself/herself as well as the partner (emotional contagion hypotheses; Kiecolt-Glaser & Wilson, 2017). According to the relational perspective (Berscheid & Ammazzalorso, 2001), exposure to a partner’s EP can function as a potent stressor, influencing spouse’s physical health problems because spouses’ psychological distress (and related health behaviors) stemming from EP are interdependent and can be transmitted between spouses. That is, a spouse’s economic stress can spillover to his/her partner, which may have physical health consequences through the same psycho-behavioral mechanisms discussed previously. Taken together, at the individual level, there may be a health mediating role of EP involving actor and partner effects. However, past studies have not adequately explored these social selection and social causation perspectives involving actor and partner effects.

Couple-level mediational processes of health problems

In addition, due to the couple context representing shared experiences and feelings between couple members, there may exist dyadic health and EP constructs (LeBlanc, Frost, & Wight, 2015). With regard to health, recent research suggests that couples’ health may converge over time because of increasing similarity of partners’ attitudes and health-related behaviors, such as diet, physical activity, smoking, and alcohol consumption (a phenomenon termed “love sick”: Kiecolt-Glaser & Wilson, 2017). This research supports the existence of a dyadic construct of physical health representing the commonalities in health shared by spouses. This dyadic construct may contribute to health continuity of husbands and wives; couple-level health continuity (e.g., Lee et al., 2019).

Like these shared health experiences, in enduring relationships, experiences on financial hardship are often similar for spouses (e.g., Barton & Bryant, 2016) because both spouses are exposed to the same or similar life events. These shared stress feelings of economic problems may form unique couple-level EP constructs. This is consistent with previous research noting that EP exists at the couple-level across midlife as a shared negative stressful experience (e.g., Barton & Bryant, 2016). Taken together, based on previous research, there is reason to believe that couple-level common constructs of health continuity and EP exist and may play a role in the longitudinal couple-level health mediational processes from midlife to later adulthood. However, previous research has not explored potential multilevel health mediation processes (couple and individual levels), which limits our understanding of how couple- and individual-level processes uniquely contribute to health mediational processes from midlife to later adulthood. Using prospective data collected from husbands and wives over a period of 25 years (1991–2015), this study examines the unique individual- and couple-level mediating influence of EP trajectories on health continuity over the mid-later years (40–65 years). Figure 1 presents the hypothesized model of the present study.

Figure 1.
Hypothesized multilevel mediational processes of health problems in midlife.

Method

Participants and procedures

The data used to evaluate these hypotheses are from the Iowa Youth and Family Project (IYFP, 1989–1994; mean age = 40 and 39 years in 1989 for husbands and wives, respectively), which was later continued as two panel studies: the Midlife Transitions Project (2001) and the Later Adulthood Study (2015). Together, these projects provide data over 27 years on rural families from a cluster of eight counties in North Central Iowa that closely mirror the economic diversity of the rural Midwest. The IYFP began as a study of rural couples with at least one-seventh grade child in 1989 (Conger & Elder, 1994). The protocol and all study procedures were approved by the Institutional Review Board at the principal investigator’s university (Iowa State University 1989–2001; University of Georgia 2015).

The current study utilized data from 257 couples who participated in 1990, 1991, 1992, 1994, 2001, and 2015 data collections and were consistently married throughout the study period. Of the couples excluded from the current analyses (43% of the original sample), the majority were excluded because they divorced or separated by 2015. An attrition analysis compared demographic characteristics (i.e., age, education level [≥2 years of college = 54.8% and 54.7% for husbands and wives, respectively], and annual household income [<$40,000 = 79.5%]), psychopathology, divorce proneness, and physical health problems at 1989 between the current analytic sample of consistently married couples and couples who were excluded from the current analyses due to divorce/separation or study attrition. The only significant difference noted was for divorce proneness, with higher scores reported for couples who were excluded from the current analysis due to study attrition or divorce (ts = 12.32 and 14.64, p < .001 for husbands and wives). On average, the couples had been married for 19 years and had three children in 1989. Because there are very few minorities in the rural area studied, all participating families were White.

Measures

Economic pressure

EP was measured using three items separately from husbands and wives in 1990, 1991, 1992, 1994, and 2001 (Whitbeck et al., 1991). Participants were asked the extent to which they were having difficulty paying their bills (1 = no difficulty at all, 5 = a great deal of difficulty), whether they had money left over at the end of the month (1 = more than enough money left over, 5 = not enough to make ends meet), and whether their income never seems to catch up with their expenses (1 = strongly disagree, 5 = strongly agree). The items were averaged with higher scores indicating more financial strain. Average reliability coefficients across all study time points were above .85 for wives and husbands.

Physical health problems

In 1990 and 2015, husbands’ and wives’ physical health problems were measured using two subscales: (a) poor global health and (b) physical diseases. Poor global health—Participants indicated on a scale from 1 = excellent to 5 = poor: “How would you rate your overall physical health?” Higher values indicate worse health. Physical diseases—Respondents indicated whether they suffered from any symptoms or diseases (diagnosed by a physician) from a list of 48 physical health ailments in the previous 2 years. Sample items include “diabetes,” “cancer,” “arthritis,” “stroke,” “high cholesterol,” “high blood pressure,” and “peptic ulcer.” Items were coded as 1 = yes and 0 = no. A sum score was computed representing the number of illnesses or chronic health conditions the individual experienced in the previous 2 years (Lorenz et al., 2006).

Statistical analyses

First, intraclass correlations (ICCs; Kenny et al., 2006) were calculated to examine the proportions of couple- and individual-level variances in health problems and EP. These values provide context regarding the relative magnitude of couple- and individual-level variability of study variables (i.e., poor global health, physical diseases, and EP). Second, latent constructs of health problems were estimated for husbands and wives in midlife (1990) and later adulthood (2015) using measures of poor global health and physical diseases as two repeated indicators. Regression paths were then specified not only between health problems in 1990 and 2015 but also between husbands’ and wives’ constructs (representing an APIM specification; Kenny et al., 2006). Third, a dyadic latent growth curve model investigated EP trajectories during midlife (1990–2001) for husbands and wives. Time scores of the growth model were centered at the second measurement point (1991) to adjust for trajectory effects of early EP (1990). Fourth, the dyadic latent growth curve model was specified within the APIM estimated in Step 2 to investigate the longitudinal individual-level mediation processes of health. In the model, actor and partner effects were estimated by specifying regression paths between both spouses’ health problems and EP (see Figure 1).

Fifth, to estimate longitudinal couple-level health mediational processes, higher-order latent variables (Wickrama, Lee, O’Neal, & Lorenz, 2016; see Figure 1) were specified for couple-level physical health and EP trajectories in the individual-level latent factors. That is, a latent construct of couple health was specified from husbands’ and wives’ latent constructs of physical health (i.e., primary latent factors). The same approach was used for higher-order latent growth variables of couples’ EP. To estimate these couple-level common (or shared) latent variables, we fixed all factor loadings between higher-order latent factors and primary latent factors to 1 (fixed-effect model; Schofield & Abraham, 2017). In a dyadic model, this approach is also known as a common fate model (Ledermann & Kenny, 2012). Longitudinal mediation processes of health problems were then specified using higher-order latent factors. Given that higher-order latent factors (i.e., couple-level health problems and EP) take into account couple-level common (shared) variables, the higher-order mediation effects (couple health problems → couple EP trajectories → couple health problems) reflect longitudinal couple-level health mediational processes.

Last, in the couple-level health mediational processes described in Step 5, the individual-level health mediational paths described in Step 4 were reestimated using the residual variances of the primary latent factors for health problems and EP trajectories. The residual variances of primary latent factors after taking higher-order factors into account represent unique (unshared) variances of the primary factors (Schofield & Abraham, 2017). That is, in our hypothesized couple-level mediational model, the residual variances of individual-level latent factors reflect unique variances of husbands’ and wives’ health problems and EP trajectories after taking the couple-level common (shared) health problems and EP trajectories into account. Thus, in these analyses, coefficients for individual-level health mediational processes reflect unique individual-level health mediational process after adjusting for couple-level health mediational processes. All mediation effects were tested using a bootstrapping approach (MacKinnon, 2011).

For model evaluation, the Comparative Fit Index (CFI; acceptable fit > .90; Little, 2013) and Root Mean Square Error of Approximation (RMSEA; acceptable fit < .08; Little 2013) were utilized. Maximum likelihood estimation with robust standard errors which handles missing values by implementing full information maximum likelihood was used in Mplus (version 8.00; Muthén & Muthén, 2017).

Results

Individual- and couple-level variability in health problems and EP

To examine individual- and couple-level variability in health problems and EP, ICCs were calculated. The ICCs of health problem indicators (poor global health and physical diseases in 1990 and 2015) were, on average, .42 (ranged from .27 [in physical diseases in 1990] to .53 [poor global health in 2015]). These results suggest that approximately 42% of the total variance in health problems was explained by couple-level variance (with the remaining 58% of the total variance in health problems existing at the individual level). ICCs of EP were, on average, .64 (ranged from .63 in 2015 to .65 in 1991). Thus, on average, approximately 64% of the total variance in EP was explained by couple-level variance (with individual-level variance explaining 36% of the total variance in health problems). Taken together, the results indicate couple- and individual-level variations in physical health and EP in the couple context.

Individual-level mediation processes of health problems

For husbands and wives, the mean of EP decreased over time (see the Online Supplemental Table 1). Conversely, both husbands’ and wives’ poor global health increased between 1990 and 2015. In addition, husbands’ physical diseases typically increased more than wives’ diseases from 1990 to 2015 (t = 5.89, p < .001; see the Online Supplemental Table 1). The APIM results indicated strong actor effects for husbands and wives (βs = .86 and .89, p < .001, respectively). However, partner effects were not significant (see corresponding coefficients in the Online Supplemental Figure 1). Next, results from the dyadic growth curve model of EP (CFI = .97; RMSEA = .07) showed that wives averaged higher initial levels of EP compared to husbands’ EP (means = 2.66 and 2.74, p < .001 for husbands and wives, respectively; Wald test = 3.98, p < .05). However, both husbands’ and wives’ slope decreased over time (means = −.26 and −.27, p < .001, respectively) with no statistical difference between husbands and wives (Wald test = .06, p = .81). All standardized coefficients for the actor–partner mediation processes are shown in Figure 2 (CFI = .96, RMSEA = .06). For actor mediation processes, for both husbands and wives, two significant mediation effects were found: through the initial level (β = .08 and .13 for husbands and wives, respectively, p < .05) and slope (β = .11 and .15 for husbands and wives, respectively, p < .05) of their own EP.

Figure 2.
Actor–partner mediation process of health problems. Note. Standardized coefficients are shown. HP = health problem; I = initial level; S = linear slope; EP = economic pressure. The (H) and (W) superscripts represent husband and wife, respectively. Husbands’ and wives’ EP in 1990 were accounted for when estimating latent growth factors of EP. Only statistically significant coefficients are shown. CFI = .96. RMSEA = .06. p < .05; p < .01; p < .001. CFI = Comparative Fit Index; RMSEA = Root Mean Square Error of Approximation.

In addition, two partner mediation processes were noted. Wives’ health problems in 1990 were related to increased initial levels of husbands’ EP, which, in turn, appeared to increase husbands’ health problems in 2015 (i.e., wives’ partner mediation process of health problems). This mediation effect was statistically significant (β = .13, p < .05). The second partner mediation process indicated that husbands’ health problems in 1990 were related to an increased downward slope for their own EP, which, in turn, was linked to more health problems for wives in 2015 (i.e., husbands’ partner mediation process of health problems; β = .17, p < .05). Total indirect actor effects between husbands and wives were similar (R ² = .23 and .20, respectively). Total indirect effects for the partner effects between husbands and wives were also similar (R ² = .21 and .27, respectively).

Unique couple- and individual-level mediation process of health problems

All estimated coefficients for couple-level health mediation processes are shown in Figure 3 (CFI = .96; RMSEA = .05). The results indicated couple-level continuity in health problems from 1990 to 2015 (β = .30, p < .001). In addition, two significant mediation effects were detected at the couple level: (a) through the initial level of couples’ EP (β = .17, p < .05) and (b) through the slope of couples’ EP (β = .13, p < .01). After accounting for these couple-level mediation process, at the individual-level, continuity in health problems from 1990 to 2015 remained strong for both husbands and wives (βs =.55 and .48, p < .001, respectively). In addition, actor mediation effects for husbands’ health problems were still significant through their initial level (β = .13, p < .05) and slope (β = .13, p < .05) of EP. However, the actor mediation effect of health problems was not significant for wives’ initial level (β = .02, p = .38) or slope (β = .04, p = .28) of EP. In addition, the two partner mediation processes noted in the previous individual-level models became nonsignificant after accounting for the couple-level processes (not shown in the figure; β = .05, p = .28 and β = .03, p = .56 for wives’ partner mediation process and husbands’ partner mediation process, respectively). After taking into account the couple-level mediation processes, the total indirect effects for husbands’ actor effects were relatively higher than the total indirect effects of wives (R ² = .22 and .08, respectively). The total indirect partner effects between husbands and wives were small and similar in magnitude (R ² = .03 and .02, respectively).

Figure 3.
Couple- and individual-level mediation process of health problems. Note. Standardized regression coefficients are shown. Standardized factor loadings are shown in parenthesis. HP = health problem; I = initial level; S = linear slope; EP = economic pressure. The (H), (W), and (C) superscripts represent husband, wife, and couple, respectively. CFI = .96. RMSEA = .05. Nonsignificant partner effect coefficients are not shown. Grayed paths = factor loadings; Solid paths = significant paths; Dotted paths = nonsignificant paths. p < .05; p < .01; p < .001. CFI = Comparative Fit Index; RMSEA = Root Mean Square Error of Approximation.

Discussion

The present study investigated associations between health problems and EP examining EP over time as a mediational process contributing to the continuity in health problems across midlife and later adulthood. A series of longitudinal dyadic model approaches allowed for an examination of these associations at the individual (actor and partner mediation) and couple levels. Previous couple research has mainly emphasized either individual-level associations (e.g., APIM) or couple-level associations (e.g., common fate model). However, findings from the ICCs supported the existence of both individual- and couple-level variabilities in EP and health problems suggesting multilevel (i.e., individual and couple levels) structures in these attributes. Thus, findings from the current study extend previous research by identifying the unique (unshared) individual- and couple-level common (shared) health mediational processes from midlife to later adulthood.

Individual-level health mediational processes (actor and partner effects)

The findings indicated significant health continuity for husbands and wives from midlife to later adulthood (health problems in 1990 → health problems in 2015; see Figure 2) even after accounting for mediation processes. This health continuity suggests that individual health problems during midlife have persistent effects on later health problems.

Mediational processes were also identified with EP mediating husbands’ and wives’ health continuity from midlife to later adulthood (i.e., actor effects). This is consistent with the social selection perspective (Warren, 2009), which proposes that individuals’ health and well-being have long-term influence on their social circumstances and behaviors. It is also consistent with the social causation perspective (Conger & Donnellan, 2007), which posits that social conditions lead to variation in health and well-being. Because both perspectives were supported, findings suggest that these two perspectives are not isolated. Instead, they are sequentially connected from midlife to later adulthood (longitudinal health sequential processes; health problems → EP → health problems).

Moreover, these findings extend previous research by identifying social selection and social causation effects may occur for partner effects (i.e., the partner mediation of health problems) in enduring couple relationships. Previous studies mainly emphasized partner effects during midlife using social causation perspectives. For example, Wickrama, Kwag, Lorenz, Conger, and Surjadi (2010) reported that couple economic hardship influenced each partner’s health problems during midlife. This is consistent with the current study, which also identified that longitudinal changes in husbands’ EP in midlife (i.e., the slope of EP) were generally associated with an increase in wives’ health problems in later adulthood. Interestingly, the results indicated that longitudinal changes in wives’ EP did not explain husbands’ health problems in later adulthood. This is consistent with previous research noting that spouse’s health consequences from EP (e.g., job loss) were more severe when it was the husband, rather than the wife, who experienced EP (Bubonya et al., 2017). The findings of the current study support the evidence for this partner effect. Importantly, the current study extends previous findings by identifying that husband’s EP in midlife predicted wife’s health problems in later adulthood.

In addition, the current study identified partner effects in social selection processes (health → EP). Wives’ health problems in midlife had a proximal effect on husbands’ EP (i.e., initial levels of husbands’ EP). This may be due in part to wives reducing their provision of support while experiencing health problems, as low levels of partner social support are linked to spouses experiencing increased family-to-work conflict (Selvarajan, Cloninger, & Singh, 2013), which can intensify EP (Schieman & Young, 2011). The results suggest this crossover is specific to wives’ midlife health problems and husbands’ EP, which extends on previous findings by identifying potential gender differences in this partner effect.

The results did not demonstrate partner effects for husbands (i.e., effects of husbands’ health problems on wives’ EP). However, husbands’ health problems were positively associated with wives’ health problems, which, in turn, influenced wives’ health problems. Together, these associations imply wives’ health problems may serve as a mediator linking husbands’ health problems and wives’ EP. However, given that husbands’ and wives’ health problems were measured at the same time point (1990), we were not able to conduct test this potential mediation with the necessary temporal precedence (i.e., the cause precedes the effect; Shadish, Cook, & Campbell, 2002). Future studies exploring this complex mediational process should explore this possibility in greater detail. Nonetheless, the findings of this study suggest dynamic actor–partner social selection and causation processes in couple contexts from midlife to later adulthood.

Couple-level health mediational processes

The findings demonstrate longitudinal health concordance across mid-later years by identifying couple-level continuity in health problems. Previous studies suggest that couples transmit health behaviors (sleep, exercise, and diet quality) to one another, and these can converge in a common path to promote shared risk for health problems (Meyler, Stimpson, & Peek, 2007). Our findings support this view by identifying the longitudinal mutual influence (i.e., interdependence; Galovan et al., 2017) of physical health problems between husbands and wives from midlife to later adulthood.

In addition, we examined couple-level EP trajectories. These couple-level EP trajectories mediated couple-level health continuity from midlife to later adulthood, and these couple-level mediation processes remained significant even after accounting for the individual-level health mediation processes, indicating unique couple-level health mediation processes (see Figure 3). Previous studies mainly focused on examining individual health sequential processes (e.g., Warren, 2009; health problems → EP → health problems). However, our results emphasize the importance of considering couple-level contexts as similar patterns of health sequential processes were found for husbands and wives in enduring couple relationships.

Unique individual-level health mediational processes

In the higher-order model (see Figure 3), the results indicated significant individual-level health continuity (husbands’ and wives’ health problems in 1990 → health problems in 2015) even after adjusting for couple-level health mediational processes. These findings of individual- and couple-level processes provide clear support for the existence of multilevel structures of health continuity for enduring couples.

Importantly, some of the individual-level actor and partner effects disappeared when the couple-level processes were incorporated in the model. For example, wives’ actor mediation processes and partner mediation processes for both husbands and wives were not detected when considering couple-level health processes. One plausible explanation is that there may be strong associations (i.e., similarities) between husbands and wives not only in the same domain (e.g., associations between husbands’ and wives’ health problems) but also across different domains (e.g., husbands’ health problems ↔ wives’ EP) over time. This would explain why most individual-level actor–partner health mediation processes may be explained by couple-level mediational processes.

However, even after taking into account couple-level health mediation processes, the actor mediation effects for husbands’ health problems remained significant. Psycho-behavioral mechanisms may drive this individual-level association with husbands’ poor health increasing the probability of job loss. The persistence of this effect for men, but not women, is consistent with previous research reporting that husbands’ income is a stronger determinant of household income compared to wives’ income (Nadim, 2016; Ulbrich, 1988), which may contribute to men’s financial pressure exhibiting a stronger impact on their depressive symptoms (e.g., Ross & Huber, 1985). In addition, previous studies indicate that husbands with elevated health problems are less likely to seek spousal support compared to wives (Vogel, Wester, Hammer, & Downing-Matibag, 2014), which may increase husbands’ social isolation (e.g., loneliness; psychological mechanism). Together, these multiple components of psychological distress (depressive symptoms and social isolation) may increase husbands’ reliance on maladaptive behavioral coping strategies (behaviors), such as binge drinking and risk-taking, which often result in health problems (Polenick, Renn, & Birditt, 2018). Future studies examining the persistent individual-level processes for men should account for these psycho-behavioral mechanisms to enhance understanding of these gender differences.

Limitations, future directions, and implications

Several limitations must be acknowledged. First, the sample was comprised only of White couples living in rural Iowa. Studies testing similar models with a more diverse population that include multiple ethnicities, greater variation in length of marriage, and occupations are necessary. Second, self-reported information regarding health problems may be a source of bias. Future research should extend this line of research using clinical measures of health outcomes. Third, we limited our analyses to couples in enduring marriages due to our interest in health concordance. While results may vary somewhat for people who marry later in life as they have less shared history with their partner, the general relationship processes are expected to be largely similar but perhaps smaller in magnitude. Similarly, we did not account for how divorce or other marital disruptions (i.e., extended separations) may impact these findings, although these marital transitions likely have economic and health consequences.

In sum, the present study contributes to existing knowledge about unique individual- and couple-level health mediational processes over an extended period of time (1990–2015) in a sample of enduring marriages. Importantly, the findings indicate that couples’ health problems in midlife continue over the life course through couple-level EP, creating couple-level health mediational processes. These findings emphasize the importance of couple-level intervention efforts to improve couple health in middle and later years. In addition, the results suggest that unique individual-level health mediational processes (particularly for husband) operate simultaneously. Consequently, clinical and prevention efforts to improve physical health for married adults in midlife and later adulthood should carefully consider multilevel health mediation processes at the individual and couple levels. Integrated individual and couple healthcare prevention is one strategy that may prove useful for reducing older adults’ health problems (Heitler, 2001). Accordingly, health planners and activists should focus on strengthening both individual- and couple-level behavioral coping mechanisms, such as pleasurable joint activities and engagement in joint and individual health promotion behaviors, which can safeguard individuals from physical health problems in their later life.

Supplemental material

Supplemental Material, Online_Supplemental_Material_XL - Health continuity over mid-later years in enduring marriages: Economic pressure as couple- and individual-level mediator

Supplemental Material, Online_Supplemental_Material_XL for Health continuity over mid-later years in enduring marriages: Economic pressure as couple- and individual-level mediator by Tae Kyoung Lee, Kandauda A. S. Wickrama and Catherine Walker O’Neal in Journal of Social and Personal Relationships

Footnotes

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work 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.

ORCID iD

Tae Kyoung Lee

Catherine Walker O’Neal

Open research statement

This research was not pre-registered. The data and materials used in the research are available upon request by emailing txl371@med.miami.edu.

Supplemental material

Supplemental material for this article is available online.

References

Barton

A. W.

Bryant

C. M.

(2016). Financial strain, trajectories of marital processes, and African American newlyweds’ marital instability. Journal of Family Psychology, 30, 657–664. doi:10.1037/fam0000190

Berscheid

Ammazzalorso

(2001). Emotional experience in close relationships. In Fletcher

G. J.

Clark

M. S.

(Eds.), Interpersonal process. Blackwell handbook of social psychology (pp. 308–330). Malden, MA: Blackwell.

Bubonya

Cobb-Clark

D. A.

Wooden

. (2017). Job loss and the mental health of spouses and adolescent children. Journal of Labor Economics, 6, 6. doi:10.1186/s40172-017-0056-1

Clarke

P. J.

O’Malley

P. M.

Schulenberg

J. E.

Johnston

L. D.

(2010). Midlife health and socioeconomic consequences of persistent overweight across early adulthood: Findings from a national survey of American adults (1986−2008). American Journal of Epidemiology, 5, 540–548. doi:10.1093/aje/kwq156

Conger

R. D.

Donnellan

M. B.

(2007). An interactionist perspective on the socioeconomic context of human development. Annual Review of Psychology, 58, 175–199. doi:10.1146/annurev.psych.58.110405.085551

Conger

R. D.

Elder

G. H.

Jr . (1994). Families in troubled times: Adapting to change in rural America. New York, NY: Aldine de Gruyter.

Falconier

M. K.

(2010). Female anxiety and male depression: Links between economic strain and psychological aggression in Argentinean couples. Family Relations, 59, 424–438. doi:10.1111/j.1741-3729.2010.00613.x

Galovan

A. M.

Holmes

E. K.

Proulx

C. M.

(2017). Theoretical and methodological issues in relationship research: Considering the common fate model. Journal of Social and Personal Relationships, 34, 44–68. doi:10.1177/0265407515621179

Heitler

(2001). Combined individual/marital therapy: A conflict resolution framework and ethical considerations. Journal of Psychotherapy Integration, 11, 349–383. doi:1016606317558

10.

Kenny

D. A.

Kashy

D. A.

Cook

W. L.

(2006). Dyadic data analysis. New York, NY: Guilford Press.

11.

Kiecolt-Glaser

J. K.

Wilson

S. J.

(2017). Lovesick: How couples’ relationships influence health. Annual Review of Clinical Psychology, 13, 421–443. doi:10.1146/annurev-clinpsy-032816-045111

12.

LeBlanc

A. J.

Frost

D. M.

Wight

(2015). Minority stress and stress proliferation among same-sex and other marginalized couples. Journal of Marriage and Family, 77, 40–59. doi:10.1111/jomf.12160.

13.

Ledermann

Kenny

D. A.

(2012). The common fate model for dyadic data: Variations of a theoretically important but underutilized model. Journal of Family Psychology, 26, 140–148. doi:10.1037/a0026624

14.

Lee

Jackson

(2017). The simultaneous effects of socioeconomic disadvantage and child health on children’s cognitive development. Demography, 54, 1845–1871. doi:10.1007/s13524-017-0605-z

15.

Lee

T. K.

Wickrama

K. A. S.

O’Neal

C. W

. (2019). Midlife general psychopathology trajectories and later-life physical health in husbands and wives. Health Psychology, 38(6),553–562.

16.

Lee

T. K.

Wickrama

K. A. S.

O’Neal

C. W.

(2019). Early socioeconomic adversity and cardiometabolic risk in young adults: Mediating roles of risky health lifestyle and depressive symptoms. Journal of Behavioral Medicine, 42, 150–161. doi:10.1007/s10865-018-9952-5

17.

Leonard

Hughes

A. E.

Pruitt

S. L.

(2017). Understanding how low-socioeconomic status households cope with health shocks: An analysis of multisector linked data. The Annuals of the American Academy of Political and Social Sciences, 669, 125–145. doi:10.1177/0002716216680989

18.

Little

T. D.

(2013). Longitudinal structural equation modeling. New York, NY: Guilford.

19.

Lorenz

F. O.

Wickrama

K. A. S.

Conger

R. D.

Elder

G. H.

Jr . (2006). The short-term and decade-long effects of divorce on women’s midlife health. Journal of Health and Social Behavior, 47, 111–125. doi:10.1177/002214650604700202

20.

MacKinnon

D. P.

(2011). Integrating mediators and moderators in research design. Research on Social Work Practice, 21, 675–681. doi:10.1177/1049731511414148

21.

Meyler

Stimpson

J. P.

Peek

M. K.

(2007). Health concordance within couples: A systematic review. Social Science and Medicine, 64, 2297–2310. doi:10.1016/j.socscimed.2007.02.007

22.

Muthén

L. K.

Muthén

B. O.

(2017). Mplus user’s guide (8th ed.). Los Angeles, CA: Muthén & Muthén.

23.

Nadim

(2016). Undermining the male breadwinner ideal? Understanding of women’s paid work among second-generation immigrants in Norway. Sociology, 50, 109–124. doi:10.1177/0038038514560259

24.

Ohrnberger

Fichera

Sutton

(2017). The relationship between physical and mental health: A mediation analysis. Social Science and Medicine, 195, 42–49. doi:10.1016/j.socscimed.2017.11.008

25.

Pearlin

L. I.

Schieman

Fazio

E. M.

Meersman

S. C.

(2005). Stress, health, and the life course: Some conceptual perspectives. Journal of Health and Social Behavior, 46, 205–219. doi:10.1177/002214650504600206

26.

Polenick

C. A.

Renn

B. N.

Birditt

K. S.

(2018). Dyadic effects of depressive symptoms on medical morbidity in middle-aged and older couples. Health Psychology, 37, 28–36. doi:10.1037/hea0000573

27.

Ross

C. E.

Huber

(1985). Hardship and depression. Journal of Health and Social Behavior, 26, 312–328. doi:10.2307/2136655

28.

Schieman

Young

(2011). Economic hardship and family-to-work conflict: The importance of gender and work conditions. Journal of Family and Economic Issues, 32, 46–61. doi:10.1007/s10834-010-9206-3

29.

Schofield

T. J.

Abraham

W. T.

(2017). Intergenerational continuity in attitudes: A latent variable family fixed-effects approach. Journal of Family Psychology, 31, 1005–1016. doi:10.1037/fam0000375

30.

Selvarajan

T. T.

Cloninger

P. A.

Singh

(2013). Social support and work-family conflict: A test of an indirect effects model. Journal of Vocational Behavior, 83, 486–499. doi:10.1016/j.jvb.2013.07.004

31.

Shadish

W. R.

Cook

T. D.

Campbell

D. T.

(2002). Experimental and quasi-experimental designs for generalized causal inference. Boston, MA: Houghton Mifflin.

32.

Thoits

P. A.

(2010). Stress and health major findings and policy implications. Journal of Health and Social Behavior, 51, S41–S53. doi:10.1177/0022146510383499

33.

Ulbrich

P. M.

(1988). The determinants of depression in two-income marriages. Journal of Marriage and Family, 50, 121–131. doi:10.2307/352433

34.

Vogel

D. L.

Wester

S. R.

Hammer

J. H.

Downing-Matibag

T. M.

(2014). Referring men to seek help: The influence of gender role conflict and stigma. Psychology of Men and Masculinity, 15, 60–67. doi:10.1037/a0031761

35.

Warren

J. R.

(2009). Socioeconomic status and health across the life course: A test of the social causation and health selection hypotheses. Social Forces, 87, 2125–2153. doi:10.1353/sof.0.0219

36.

Whitbeck

L. B.

Simons

R. L.

Conger

R. D.

Lorenz

F. O.

Shirley

Elder

G. H.

Jr . (1991). Family economic hardship, parental support, and adolescent self-esteem. Social Psychology Quarterly, 54, 353–363. doi:10.2307/2786847

37.

Wickrama

K. A. S.

Kwag

K. H.

Lorenz

F. O.

Conger

R. D.

Surjadi

F. F.

(2010). Dynamics of family economic hardship and the progression of health problems of husbands and wives during the middle years: A perspective from rural Mid-West. Journal of Aging and Health, 22, 1132–1157. doi:10.1177/0898264310377353

38.

Wickrama

K. A. S.

Lee

T. K.

O’Neal

C. W.

Lorenz

F. O.

(2016). Higher order growth curves and mixture modeling with Mplus: A practical guide. New York, NY: Routledge.

39.

Wickrama

K. A. S.

O’Neal

C. W.

Neppl

T. K

. (in press). Midlife family economic hardship and later life cardiometabolic health: The protective role of marital integration. The Gerontologist. doi:10.1093/geront/gny047

Supplementary Material

Please find the following supplemental material available below.

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

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

0.00 MB

0.47 MB