Abstract
Romantic relationships are ubiquitous among adolescents and adults the world over. More than 90% of adults in the United States will marry at some point, and cohabitation is increasingly common among unmarried adults. Intimate relationships are arguably the main way that we fulfill our fundamental need for connection. In the United States and many other countries, for individuals in a committed monogamous romantic relationship, the relationship itself becomes one of the key contexts for mental health. The association between relationship distress and various forms of psychopathology is as strong as many other well-known predictors of mental illness. In this article, I discuss how relationships that become unsatisfying, distressed, or conflicted are a precursor to the experience of mental illness. I also discuss how the romantic relationship may trigger a diathesis for psychopathology. That diathesis may be biological (e.g., genetic) or psychological (e.g., cognitive, emotional).
Mental health is a staggering public health problem in the United States, with approximately 50% of the population meeting criteria for a disorder at some point in their lives (Kessler et al., 2005). In the search for etiological factors contributing to psychopathology, one frequently studied but still underemphasized risk factor is the quality of a current, committed romantic relationship. In this article, I review what is known about the strength and direction of the association between relationship distress and various common forms of psychopathology. I then discuss potential mechanisms for the effect of distress on psychopathology. I build the argument that a distressed and unsatisfying romantic relationship may be one of the most important social-environmental triggers for the expression of psychopathology.
Relationship Distress and Psychopathology
In the United States, approximately 90% of the population will marry at some point in their lives. Among younger generations, for whom marriage is increasingly postponed, cohabitation is the norm; for instance, from 2013 to 2017, the percentage of U.S. adults aged 18 to 44 who had ever cohabitated (59%) was larger than the percentage who had ever married (50%; Horowitz et al., 2019). For those who are partnered, through cohabitation or marriage, a satisfying, well-functioning romantic relationship is a vital part of overall well-being (Heller et al., 2004). Luckily, most couples are relatively happy, but there appears to be a small group or class of couples that represents the extreme ends of unhappiness (Whisman et al., 2008). Modeling techniques that look for discontinuities or “clusters” among the population have consistently found that approximately 20% of the population falls into this “dissatisfied class.” Even individuals outside of this distressed group may experience an intra-individual decrease in satisfaction (i.e., a change from previous levels) that is personally significant even if it is smaller in absolute comparison with that of other individuals (South et al., 2020).
There are clear mental health consequences for unhappily married or partnered individuals (for a recent review, see South, 2021). Individuals who are more distressed and less satisfied in their current romantic relationship have a greater likelihood of meeting a diagnosis for several common forms of psychopathology, including specific phobia (odds ratio [OR] = 1.34), social phobia (OR = 1.93), generalized anxiety disorder (OR = 2.54), posttraumatic stress disorder (OR = 2.30) major depressive disorder (OR = 1.68), bipolar disorder (OR = 3.60), and alcohol use disorder (OR = 2.78; e.g., Whisman, 2007). In three different nationwide community surveys, increased relationship satisfaction was related to decreased likelihood of both broadband classes of disorders (e.g., any anxiety disorder, any mood disorder, any substance use disorder) as well as most of the specific individual disorders (e.g., generalized anxiety disorder); further, these associations did not differ across 11 racial and ethnic categories (McShall & Johnson, 2015). In a study of young male same-sex couples, relationship quality was negatively correlated with depression (Feinstein et al., 2019). The most commonly studied disorder with regard to marital distress is depression, which has a moderate negative relationship with marital satisfaction (meta-analytic r = −.35; Whisman, 2001). As a comparison, data from a large nationwide community survey found associations between risk factors (other than relationship distress) and first episode of major depression that were similar or smaller in magnitude (e.g., sex: OR = 2.13, race: OR = 1.16, education: OR = 0.98, childhood abuse: OR = 3.58, family history: OR = 1.62; Peters et al., 2015).
Much of the early work on relationship distress and psychopathology was cross-sectional, and the direction of the effect was assumed to go from distress to disorder. It is possible that the direction of the effect may flow from increased levels of psychopathology to decreased levels of relationship satisfaction. For instance, the experience of depression can contribute to a worsening of relationship problems (Coyne, 1976). There is also evidence that having a spouse with a history of psychopathology (depression, anxiety, bipolar disorder, or nonaffective psychosis) increases his or her partner’s risk for alcohol use disorder during the marriage (Salvatore et al., 2022).
The longitudinal studies that have been conducted, however, seem to point toward the direction from distress to disorder (Whisman & Bruce, 1999). In one study, individuals who were not depressed at baseline but were in a distressed relationship were 270% more likely to experience a subsequent major depressive episode that individuals who were not in a distressed relationship (Whisman & Bruce, 1999); similar results were found in a study examining new onset of substance use (Whisman et al., 2006). Among a sample of military personnel, the presence of marital distress led to incidence (new cases) of disorder for those not meeting criteria at baseline for major depression (OR = 1.17), generalized anxiety (OR = 1.20), and posttraumatic stress disorder (OR = 1.17;Whisman et al., 2021). In summary, longitudinal studies would support a direction of effect whereby a distressed relationship leads to greater likelihood of mental health problems.
Specificity or Generality of the Effect of Distress on Psychopathology
In the categorical system of the Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition (DSM-5; American Psychiatric Association, 2013), the most widely used classification of psychopathology in the United States, every disorder is presumed to be discrete from every other disorder and is demarcated from nondisorder according to a set threshold of symptoms that are met by the patient. But evidence shows that mental disorders are comorbid (occur together at greater than chance likelihood), and there are strong and significant correlations between disorders (e.g., depression and generalized anxiety disorder; Kotov et al., 2017). In the last few decades, many researchers have argued for a reconceptualization of psychopathology classification. The Hierarchical Taxonomy of Psychopathology (HiTOP; Kotov et al., 2017) consortium proposes a dimensional classification of psychopathology in which disorder severity exists along a continuum and psychopathology is arranged hierarchically (see Fig. 1). The correlations among psychopathology symptoms are explained by six domains or spectra that also explain the covariance between personality and mental disorders. Detachment subsumes extreme extroversion, antagonistic externalizing subsumes conduct disorder and antisocial personality disorder, disinhibited externalizing subsumes substance use disorders, thought disorder subsumes psychosis and related disorders, internalizing subsumes depressive and anxiety disorders, and somatoform subsumes disorders that center around body preoccupations. At the topic of the hierarchy sits a general factor of psychopathology (p factor; Caspi et al., 2014). One implication of research is that individuals may be prone to general forms of psychopathology, such as the higher order spectra, rather than individual types of mental disorders.
A conceptual representation of the Hierarchical Taxonomy of Psychopathology (HiTOP) model. The HiTOP model proposes that features of personality and symptoms of psychopathology are the lowest level in a hierarchy that progresses from phenotypic (i.e., observed) features to latent domains that capture the covariance between the observed indicators. At the highest level sits an overall factor, p, that could represent generally propensity toward psychopathology per se.
Many studies looking at relationship satisfaction and psychopathology fail to account for this type of comorbidity (for an exception, see Whisman, 2007). My colleagues and I examined the generality and specificity of the associations among relationship distress and psychopathology using a sample of long-term heterosexual married couples who were assessed for lifetime psychopathology using structured clinical interviews. Symptom counts of 10 different forms of psychopathology were modeled using factor analysis, along with both husband and wife reports of relationship satisfaction. As expected, the bivariate correlations between overall marital satisfaction and psychopathology were generally small but significant (e.g., satisfaction and adult antisocial behavior: r = −.21 for husbands, r = −.19 for wives). The psychopathology symptom counts were modeled as indicators of latent factors; the subscales of the relationship-adjustment measure were also modeled as indicators of an overall latent factor of marital satisfaction. We found that the latent factor of relationship satisfaction was negatively associated with latent factors of internalizing (depression, generalized anxiety, phobias; r = −.28) and externalizing (substance use, antisocial behavior; r = −.26) psychopathology (South et al., 2011). These associations were the same across husbands and wives, meaning that the correlation between satisfaction and psychopathology did not differ across gender.
After accounting for the association between the psychopathology latent factors and the relationship distress factor, we found no significant effects between individual syndromes and overall relationship satisfaction. That is, the “action” between satisfaction and psychopathology seems to be at the level of the higher order domains, not the specific syndrome that is a time- and context-dependent manifestation of that domain. This is not to suggest that the specific form of psychopathology that an individual experiences in the context of relationship distress is unimportant. How these higher order factors of psychopathology manifest at a given time for a given individual in a specific relationship will depend on several idiosyncratic factors unique to that individual and that relationship. The take-home message from this work is that when one is looking for reasons why marital distress and relationship satisfaction are linked, in order to find a tractable hold on possible mechanisms, it might be necessary to examine the higher order domains of psychopathology rather than the specific form.
Etiology of the Association Between Relationship Distress and Psychopathology
Research has established an association between relationship distress and subsequent psychopathology. With that association established, it is possible to examine if it is attributable to genetic influences, environmental influences, or some combination of both. Several studies have used data on psychopathology and romantic-relationship distress collected from adult twins to estimate whether there are overlapping etiological influences on these constructs. Twins provide a fascinating natural experiment; identical twins share 100% of their genes, and fraternal twins share, on average, 50% of their genes. Thus, comparing the correlation on a phenotype (observed variable) between identical twins with the correlation between fraternal twins leads to a simple estimate of heritability; more complex statistical models allow for estimates of the relative influence of genes, environment shared between family members (common environment), and environment not shared between family members (unique environment) on psychopathology. There are shared genetic influences between romantic conflict and both internalizing (depression; Spotts et al., 2004) and externalizing (Salvatore et al., 2015) forms of psychopathology. This suggests that the same genetic factors that contribute to relationship distress also contribute to symptoms of psychopathology. As McGue and Lykken (1992) theorized, genetically influenced personality traits impact the quality of a romantic relationship, leading, in part, to the finding of genetic influences on phenotypes that one might not necessarily associate with a genetic basis, such as relationship satisfaction (South & Krueger, 2008) and divorce (McGue & Lykken, 1992).
What about the nonshared environment, or those experiences that are unique (i.e., not shared in common) between siblings from the same family? A recent twin analysis of middle-aged adults from Sweden found that nonshared environmental influences explained the largest amount of the covariance between relationship adjustment and both depressive symptoms and anxiety symptoms (Whisman et al., 2018), although genetic factors were still important. The nonshared environment would include a relationship with one’s romantic partner, as this is an environmental influence that an individual does not share in common with his or her sibling (indeed, fascinating research suggests that the spouses of identical twins share little in common; Lykken & Tellegen, 1993). Nonshared environmental influences on the association between two variables come close to providing a quasicausal explanation of one variable (i.e., relationship distress) having a direct causal impact on another (i.e., psychopathology), after accounting for unmeasured genetic and other environmental influences. In summary, the work using genetically informative data suggests a role for both genes and environment in the association between relationship distress and psychopathology.
A Diathesis-Stress Model of Relationship Distress and Psychopathology
A slightly different question is whether the genetic influences on psychopathology vary as a function of one’s level of distress or conflict in their romantic relationship. Following from a diathesis-stress model, marital distress may be the stressor that triggers a diathesis for a general form of psychopathology. A reasonable hypothesis is that the diathesis (the predisposition for psychopathology) might be genetic. Using a nationwide twin sample, South and Krueger (2008) found that the estimated genetic influences (i.e., heritability, or the proportion of variance due to genetic differences between people) on an internalizing factor of psychopathology differed as a function of relationship distress. Heritability was highest among individuals with the most distressed marriages and lowest among those in the most satisfying marriages. This suggests that genetic influences (and the downstream effect, for instance, level of neurotransmitters in the brain) would be expressed only among individuals whose relationship becomes distressed.
The diathesis for psychopathology, however, may not be only genetic (as operationalized by either heritability or measured genetic loci). Other possible diatheses include emotional or cognitive risk factors that are unique to the person. For instance, research shows that female partners show more attention to (male) partners’ faces than men do toward female partners’ faces. In a study of college students in romantic relationships, participants were shown pictures of their romantic partner while electrical activity was recorded. For women in less happy relationships, activity increased in areas of the brain known to be associated with rewarding experiencing, whereas activity decreased among women who were more satisfied; for men, better relationship quality was associated with more reactivity to their partner’s face. These results suggest that women in distressed relationships had heightened sensitivity to any potential threats to the relationship (Burdwood & Simons, 2016). Another possible mechanism is that in a relationship that is distressing or high in conflict, the lack of support from one’s romantic partner (Gariépy et al., 2016) leaves an individual more prone to other stressors in their life. Finally, research suggests that among partnered individuals, emotion regulation strategies mediate the links between relationship distress and symptoms of psychopathology (Marroquín & Nolen-Hoeksema, 2015).
Summary and Conclusions
A committed monogamous marriage or marriage-like relationship is almost ubiquitous in the United States and in many countries around the globe. Unfortunately, in some cases, this relationship can turn distressing and conflicted for at least one if not both partners. As shown in Figure 2, an individual may become increasingly distressed within their romantic relationship until the point that they cross a threshold, after which the distressed relationship can trigger psychopathology. This is most likely among individuals who are vulnerable because of a preexisting diathesis. Some individuals may lack risk factors that would interact with relationship distress (see blue line in Fig. 2), or they may be buffered by the existence of certain protective factors. Future work needs to examine a larger plethora of diatheses that put individuals at risk specifically from distressed relationships. Most importantly, future research needs complex designs and analyses to tease apart the idiographic ways in which individual diatheses are triggered by aspects of the relationship that are uniquely distressing to that individual. For instance, two individuals may both be at risk for psychopathology because of heightened threat sensitivity centered on the relationship; for one person, their partner’s dismissiveness of everyday concerns about the relationship may trigger psychopathology (Fig. 2, red line), whereas the other person is protected against psychopathology by a partner who is consistently reassuring and attentive to the relationship (Fig. 2, blue line). Here, a fruitful avenue for research might be focusing less on broad personality traits that characterize overall ways of behaving and thinking and more on the goals and values and needs that each person brings to the relationship and how those are met (or not) by one’s partner (e.g., Fitzsimons et al., 2015).
Schematic illustrating the associations among relationship distress and psychopathology. An individual’s level of relationship satisfaction may decrease from their individual starting point (which may be lower or higher than the population average). Distress is, in turn, prospectively linked to the occurrence of psychopathology, particularly for individuals who are at risk because of the presence of certain diatheses.
Recommended Reading
Joel, S., Eastwick, P. W., Allison, C. J., Arriaga, X. B., Baker, Z. G., Bar-Kalifa, E., Bergeron, S., Birnbaum, G. E., Brock, R. L., Brumbaugh, C. C., Carmichael, C. L., Chen, S., Clarke, J., Cobb, R. J., Coolsen, M. K., Davis, J., de Jong, D. C., Debrot, A., DeHaas, E. C. . . . Wolf, S. (2020). Machine learning uncovers the most robust self-report predictors of relationship quality across 43 longitudinal couples studies. Proceedings of the National Academy of Sciences, USA, 117(32), 19061–19071. https://doi.org/10.1073/pnas.1917036117. A recent, comprehensive meta-analysis of the longitudinal predictors of relationship quality.
Kotov, R., Krueger, R. F., Watson, D., Cicero, D. C., Conway, C. C., DeYoung, C. G., Eaton, N. R., Forbes, M. K., Hallquist, M. N., Latzman, R. D., Mullins-Sweatt, S. N., Ruggero, C. J., Simms, L. J., Waldman, I. D., Waszczuk, M. A., & Wright, A. G. C. (2021). The Hierarchical Taxonomy of Psychopathology (HiTOP): A quantitative nosology based on consensus of evidence. Annual Review of Clinical Psychology, 17(1), 83–108. https://doi.org/10.1146/annurev-clinpsy-081219-093304. A comprehensive review of the evidence in support of a new dimensional hierarchical conceptualization of psychopathology.
South, S. C. (2021). Pathology in relationships. Annual Review of Clinical Psychology, 17(1), 577–601. https://doi.org/10.1146/annurev-clinpsy-081219-115420. A review of the evidence supporting the links between psychopathology and relationship quality.
Whisman, M. A. (2001). (See References). A meta-analysis focusing on the association between depression and marital dissatisfaction.
