Cardiovascular disease is a hypernym for various types of heart and blood vessel pathologies, including ischemic heart disease, cerebrovascular disease, peripheral vascular disease, and heart failure (Fisher and Scheidt, 2012). Cardiovascular disease is a leading cause of mortality across most Western nations, accounting for 25% of annual deaths in the United States and resulting in $252 billion USD in healthcare expenditures in 2019–2020 (Centers for Disease Control and Prevention, 2024). In Canada, one in 12 adults 20 years of age or older have a diagnosis of heart disease, which represents 2.6 million people (Public Health Agency of Canada, 2024). Ischemic heart disease, which is the weakening of the heart due to decreased blood flow to it, is the most common cardiac affliction in Canada, affecting roughly 8.5% of all adults (Public Health Agency of Canada, 2024). The onset of cardiovascular disease may cause significant functional impairment and psychological distress, including reductions in mobility and symptoms of depression and anxiety. Symptoms such as chest pain and breathlessness can recur or deteriorate, and patients may require secondary preventive interventions like cardiac rehabilitation. In addition to high rates of mortality, rehospitalization, and morbidity, the onset of a cardiovascular condition can prompt social challenges that affect the patient and their family members (Helgeson et al., 2018).

Romantic partners of patients with cardiovascular disease, for example, encounter a plethora of shared stressors, including drastic changes in domestic roles, lifestyle, responsibilities, goals, and life plans (Dalteg et al., 2011; Tulloch et al., 2020). The stress that accompanies the threat of a potentially life-threatening and chronic illness like cardiovascular disease may ultimately endanger the integrity of the couple relationship (Smith and Baucom, 2017; Tulloch and Greenman, 2018). Indeed, approximately 30% of patients with cardiovascular disease report marital discord (Cameron et al., 2017). Reductions in relationship quality in the context of chronic illness has been concurrently and longitudinally linked to poor adherence to treatment regimes, such as pharmaceutical treatments and psychological interventions, and increased symptoms of mental distress (Martire and Helgeson, 2017). On the other hand, supportive relationships may be directly and indirectly (e.g., via reductions in mental health symptoms) linked to reduced cardiovascular disease risk and to more optimal recovery (Smith, 2022; Tulloch and Greenman, 2018).

Attachment theory helps explain why supportive relationships appear to be related to health outcomes, including cardiovascular disease. According to several decades of social psychological research, couple relationships constitute attachment bonds (Zeifman, 2019). Attachment theory posits that human beings are hardwired through evolution to form and to maintain affectional bonds with significant people across the lifespan, including parents during childhood and romantic partners in adulthood, because the establishment and preservation of these emotional ties are crucial to the survival of the human race (Bowlby, 1969; Zeifman, 2019). The assumed function of the attachment system is to promote survival by seeking proximity to these “attachment figures” for comfort in moments of perceived threat or danger (Bowlby, 1973). Thus, “activation” of the attachment system refers to active proximity-seeking with the aim of receiving emotional comfort that will restore a sense of safety and equilibrium.

According to attachment theory, the quality of early attachment experiences shapes emerging expectations, attitudes, and beliefs about oneself, others, and the larger social world (referred to hereafter as “internal working models;” Bowlby, 1973). A reliable pattern of interactions in which attachment figures behave with responsiveness, availability, and emotional engagement to bids for proximity is known to soothe a person’s distress and deactivate the attachment system, which in turn allows for the pursuit of growth-oriented activities such as exploration and affiliation (Cassidy, 2016). Over time, these repeated experiences of security attainment are thought to be stored at the level of schematic memory and promote the development of a secure attachment bond (Cassidy, 2016). In this way, “securely attached” individuals learn that proximity seeking is an effective emotion regulation strategy that can be confidently and reliably used to relieve distress in times of stress (Simpson and Rholes, 2017).

Conversely, a recurring pattern of interactions in which primary attachment figures lack these characteristics either maintains or heightens a person’s distress and can keep the attachment system activated (Mikulincer and Shaver, 2016). Repeated experiences of security nonfulfillment are also stored at the level of schematic memory, but in this case they foster the development of an “insecure” attachment bond (Mikulincer and Shaver, 2016). Insecurely attached individuals learn that proximity seeking is an ineffective regulation strategy in times of threat or danger and that alternative strategies are needed to relieve distress. These alternative strategies may involve hyperactivation of the attachment system or attempts to deactivate it (Cassidy, 2016). As a result, early attachment experiences provide salient procedural knowledge about distress management, laying the foundation for later intrapersonal development and interpersonal functioning (Mikulincer and Shaver, 2016; Bodenmann, 2008).

The internal dynamics of the behavioral attachment system have been found to carry forward into adulthood and to exert a continuing influence in a broad range of social contexts, especially in romantic love (Zeifman and Hazan, 2016). Individuals within a romantic dyad tend to mutually become each other’s central source of support, comfort, and security in times of need (i.e., primary attachment figures), thus making the emotional bond that develops between romantic partners one of the most significant relationships in adulthood. Like the child’s emotional ties to its mother, romantic bonds between adults also appear to have a biological basis and are linked to the survival of the species (Zeifman, 2019).

Health problems like cardiovascular disease tend to activate the attachment system (Simpson and Rholes, 2012) and, depending on their severity and chronicity, the activation of the attachment system may extend over a period of time (Meredith and Strong, 2019). This means that people with chronic illnesses tend to remain in a distressed state, with a salient need for proximity to an attachment figure. This can lead to higher levels of psychological distress, particularly in people with insecure attachment orientations. As previously mentioned, individuals with insecure attachments to their partners are likely to employ hyperactivating or deactivating emotion-regulation strategies in response to stressors, which can affect their responses to physical disease and their mental health.

The results of recent research have indeed borne out the notion of insecure attachment bonds as indicative of ineffective emotion-regulation strategies in the face of stressors. For example, in a recent systematic review of 37 studies of responses to attachment related stressors (e.g., pictures of one’s children with sad faces) with over 2,000 participants, Eilert and Buchheim (2023) found that both attachment anxiety and avoidance were significantly related to difficulties in emotion regulation. People in the studies reviewed who were high in attachment anxiety tended to use fewer cognitive reappraisal strategies, had more impulse regulation problems, and relied more heavily on interpersonal strategies like reassurance-seeking and venting to regulate emotions. Conversely, individuals high in attachment avoidance showed a stronger tendency toward suppression, lower use of reappraisal, and generally less use of interpersonal emotion regulation (Eilert and Buchheim, 2023). Secure attachment, on the other hand, involved more balanced emotion regulation strategies like cognitive reappraisal and effective co-regulation (Eilert and Buchheim, 2023).

Similarly, Messina et al. (2023) found in a recent study of 630 adults in Italy that both anxious attachment to romantic partners was related to impairment in the ability to engage in cognitive appraisal as an emotion-regulation strategy and attachment avoidance was related to attempts to suppress negative emotions rather than to deal with and work though them. Finally, Mosannenzadeh et al. (2024) noted in their study of “inflexible over-reliance on either intrapersonal (self-directed, e.g., suppression) or interpersonal (involving others, e.g., sharing)” (p. 1) emotion regulation, which combined survey and experience-sampling measures, that adults in the Netherlands who demonstrated anxious attachment to their relationship partners tended to turn to their partner for comfort, but they did not adjust that tendency depending on whether their partner was actually available. Even when their partner was not responsive, they still tended to want or to try to seek support, rather than switching to self-soothing strategies. On the other hand, those with avoidant attachment to partners exhibited a tendency to rely excessively on intrapersonal strategies, generally looking solely to themselves for emotional support. All these results demonstrate that ineffective emotion regulation strategies are characteristic of insecure attachment, which can influence the ways in which individuals and their partners cope with the stress generated by chronic illnesses such as cardiac disease.

There is mounting evidence that insecure attachment to relationship partners and the challenges to potent emotion regulation that it poses can affect how people with cardiac disease and their partners cope with the stress of the illness, which can, in turn have a negative impact on their mental health. This reflects the general finding in the literature that insecure attachment to romantic partners is linked with depression, anxiety, PTSD, and a host of other mental health problems (see Mikulincer and Shaver, 2016, for a review). Meredith and Strong (2019) reviewed the literature on attachment in chronic illness and concluded that the enduring stress and dependency associated with conditions such as diabetes, arthritis, and cardiovascular disease create contexts in which attachment needs become particularly salient, as attachment theory would predict. More recent empirical work has shown that, in cardiac populations, attachment insecurity undermines both patients’ and partners’ adjustment. For example, George-Levi et al. (2020) demonstrated in their study of 114 couples that cardiac patients with higher attachment anxiety showed greater difficulties trusting care, which may have contributed the maintenance of symptoms of anxiety over time following their diagnosis. Avoidant patients in their sample did not appear to seek or respond to partner support (George-Levi et al., 2020). Laflamme et al. (2022) found that couple relationship partners who were providing care for individuals with heart disease reported higher caregiver burden and psychological distress when they were insecurely attached to their partner. In this investigation of 181 patients with heart disease and their life partners, anxiously attached caregivers tended to become overinvolved and distressed by worry, whereas avoidantly attached caregivers appeared to struggle with the sustained intimacy and dependency demands of caregiving, which overwhelmed their usual distancing strategies (Laflamme et al., 2022).

Over time, the continuous activation of the attachment system and difficulties responding effectively to their own and to their partner’s distress may result in an elevated risk for symptoms of depression and anxiety (Mikulincer and Shaver, 2016). To our knowledge, there have been seven studies of the links between attachment orientations and psychological distress in couples facing cardiovascular disease. In four of them (Kidd et al., 2016; Vilchinsky et al., 2010; Heenan et al., 2020; Söllner et al., 2018), attachment was examined as a predictor of psychological distress among patients with cardiovascular disease over the short term (2–3 months), long term (6–12 months), or in response to psychotherapy. The results indicated that attachment anxiety was a strong predictor of depression, anxiety (Heenan et al., 2020; Kidd et al., 2016), and post-traumatic stress (Heenan et al., 2020) in people with cardiovascular disease, and that secure attachment was linked to reductions in depression (Söllner et al., 2018). Attachment avoidance did not appear to be related to mental-health outcomes in these investigations, despite meta-analytic findings in previous studies of significant associations between attachment avoidance and depression (Zhang et al., 2022), anxiety (van Leeuwen et al., 2020), and post-traumatic stress disorder (PTSD) (Ogle et al., 2015).

There are also two additional studies of the role of attachment orientations in psychological distress among spouses of patients with cardiovascular disease. In Laflamme et al.’s (2022) cross-sectional investigation, attachment anxiety and attachment avoidance predicted depression in life partners of patients with cardiovascular disease; this link was mediated by caregiver burden in both cases. In a longitudinal investigation of female partners of male patients with cardiovascular disease, Vilchinsky et al. (2015) determined that attachment-related anxiety moderated the relationship between caregiver burden at one-month post- hospitalization and spouses’ depressive symptoms at six-months follow up. Insights from these studies, however, were gleaned from just one member of the couple, which precludes the possibility of evaluating effects that take place within the dyad.

Interdependent effects of attachment and psychological distress among couples in which one or both partners have cardiovascular disease have been articulated in only one study (Kafescioglu et al., 2010). In it, dyadic coping, defined as the ways in which romantic partners cope together as a unit when dealing with a stressor, was found to be a partial mediator of attachment insecurity and mental health outcomes. Only actor effects were observed in this study. This means that one’s own attachment insecurity and perceptions of dyadic coping were not related to one’s partner’s mental health. The interpretation of these results, however, is limited by a small sample size (N = 72 couples), the use of measures with poor psychometric properties, and the exclusion of common dyadic coping, which might be a catalyst for the effects of attachment on mental health in people with cardiovascular disease.

In sum, it appears that attachment bonds are important predictors of mental health trajectories. Internalizing symptoms such as depression and anxiety tend to co-occur and are robustly linked to recurrent cardiovascular events and to all-cause mortality among those with established cardiovascular disease (Chen et al., 2019; Meijer et al., 2011). There is also evidence that dyadic coping may be involved in the relation between attachment and mental-health outcomes in patients and partners facing cardiovascular disease (Kafescioglu et al., 2010). It is therefore crucial to identify potentially modifiable factors, including relational ones, that can mitigate the worsening of patients’ and partners’ mental health. Dyadic coping is one such factor.

Cardiovascular disease is a type of “dyadic stress,” which has been defined as any form of stressful life encounter directly affecting the couple as a unit that elicits common concerns and joint appraisals in both romantic partners (Bodenmann, 1997, 2005). Research has found that dyadic stress, from daily hassles to chronic illness such as cardiovascular disease, can affect the mental health and relationship satisfaction of romantic partners, and that effective dyadic coping may buffer against these negative outcomes (Aldwin and Revenson, 1987; Papp and Witt, 2010; Randall and Bodenmann, 2009; Rusu et al., 2016), similar to the way that secure attachment can.

The Systemic Transactional Model (STM) (Bodenmann, 1995, 2005) can be understood as an extension of the Transactional Theory of Stress and Coping (Lazarus and Folkman, 1984) within a romantic perspective. The STM is based upon the postulate that stress has crossover effects from one romantic partner to the other due to the interdependent nature of couple relationships (Bodenmann, 1995). Theoretically, the STM suggests that the resources of a supportive partner may augment the available resources of the stressed partner, thereby reshaping the appraisal of external and/or internal demands as less threatening (Bodenmann, 1995). Dyadic coping is conceptualized in this model as an interpersonal stress regulation process involving both romantic partners dealing jointly with stressors that affect each partner, either directly or indirectly (Bodenmann, 2005).

Emerging research has found that common dyadic coping (CDC), which is a specific form of dyadic coping that occurs when both partners conjointly work together towards mitigating or resolving stressors experienced as a dyad by focusing on specific problems (e.g., joint problem-solving) or by engaging in a conjoint effort to regulate emotions (e.g., joint relaxation), is the most salient form of dyadic coping for couples facing stressors (Falconier et al., 2015; Falconier and Kuhn, 2019). Common dyadic coping is used in response to a stressful situation that is initially related to one member of the dyad, such as the onset of cardiovascular disease in an individual (Rapelli et al., 2021), but is experienced as a situation affecting both partners (i.e., a “we-disease”). Other positive forms include supportive dyadic coping (i.e., when one partner helps the other to deal with stress, and delegated dyadic coping (i.e., when one partner undertakes the responsibilities usually assumed by the other to help relieve the experienced stress) (Bodenmann, 2005). There are also negative forms including hostile dyadic coping (i.e., when one partner disparages or criticizes the other), superficial dyadic coping (i.e., when one partner offers support in an insincere or inattentive manner), and ambivalent dyadic coping (i.e., when one partner offers support reluctantly or grudgingly) (Bodenmann, 2005). There is evidence of links between the various types of dyadic coping and attachment to romantic partners.

Attachment researchers (e.g., Cassidy, 2016; Mikulincer and Shaver, 2016; Simpson and Rholes, 2017) have found that internal working models and strategies for responding to the activation of the attachment system not only shape the ways in which individuals in romantic relationships seek and perceive support from their partner, but they also influence how people in a relationship provide support to their distressed partner. Given that insecure romantic attachment has been found to compound emotion regulation and individual coping difficulties, a growing body of empirical evidence suggests that insecure romantic attachment may also hinder effective dyadic coping (Alves et al., 2019; Batinic and Kamenov, 2017; Gagliardi et al., 2013; Levesque et al., 2017). For instance, Batinic and Kamenov (2017) examined actor and partner effects of insecure romantic attachment on dyadic coping in a sample of 223 opposite-sex, married couples through dyadic analyses. They found significant actor and partner effects (ranging from β = −0.11 to β = −0.46), indicating that men and women with high levels of attachment anxiety and attachment avoidance were both less prone to engage in CDC themselves and less prone to have a partner who engaged in CDC.

In the same vein, Alves et al. (2019) recently conducted a longitudinal study investigating the mediating effect of CDC between insecure romantic attachment and parental adjustment in a sample of 92 opposite-sex couples. Significant actor and partner effects were found, suggesting that fathers and mothers with high levels of attachment anxiety and attachment avoidance were both less likely to engage in CDC themselves and less likely to have a partner that engaged in CDC. Interestingly, their findings revealed that the associations between attachment anxiety and CDC were broadly weaker in comparison to the associations between attachment avoidance and CDC, suggesting that the excessive worries experienced by individuals with high levels of attachment anxiety may inhibit their ability to engage in joint coping efforts in times of stress.

In light of theoretical and empirical developments from the past few years, researchers have increasingly started exploring whether dyadic coping may serve as a pathway through which romantic attachment affects intrapersonal and interpersonal outcomes. Although the dyadic coping literature indicates that dyadic coping may be more closely related to relational outcomes than individual outcomes (Bodenmann et al., 2011), dyadic coping has nonetheless been found to mediate the relationship between romantic attachment and intrapersonal outcomes, such as mental health (Kafescioglu et al., 2010), non-suicidal self-injury (Levesque et al., 2017), parental adjustment (Alves et al., 2019), and quality of life (Crangle et al., 2020).

To our knowledge, there has been only one study to date of the ways in which dyadic coping (i.e., supportive and negative dyadic coping exclusively) may be a mediating factor between insecure romantic attachment and mental health. Among a sample of 63 couples in which one partner had received a cardiac diagnosis, Kafescioglu et al. (2010) conducted Actor-Partner Interdependence Model (APIM) analyses through the use of a pooled regression approach. They found two significant mediating effects. On one hand, patients with higher levels of attachment avoidance were more likely to engage in more negative dyadic coping, which in turn appeared to negatively influence their own mental health (b = −28.72, t(1) = −3.16, p < 0.05). On the other hand, spouses with higher levels of attachment anxiety were more likely to engage in more supportive dyadic coping, which in turn, positively affected their own mental health (b = −1.85, t(1) = −2.45, p < 0.05). However, the small sample size in this study, the use of measures with poor psychometric properties, the use of pooled regressions-APIM analyses, and the exclusion of CDC considerably limit our understanding of dyadic coping in a cardiac context.

We have established that attachment to romantic partners is related to mental-health outcomes and psychological distress in people with heart disease (e.g., George-Levi et al., 2020; Laflamme et al., 2022; Meredith and Strong, 2019), that secure attachment in couple relationships is likely to coincide with common dyadic coping whereas insecure attachment has been linked to a lesser likelihood of engaging in effective common dyadic coping (e.g., Alves et al., 2019; Batinic and Kamenov, 2017), and that both supportive and negative dyadic coping have been found to mediate the link between insecure romantic attachment and mental health. Since CDC has been found to have predominantly beneficial effects from an individual and relationship perspective and it is the most prominent form of dyadic coping (Falconier et al., 2015; Rusu et al., 2020), we focus in the present paper specifically on CDC as a mediator, using a dyadic approach as a way to account for actor and partner effects.

The primary objective of this study was to investigate the link between attachment (attachment avoidance and attachment anxiety) and mental health indicators of anxiety and depression among patients with cardiovascular disease and their spouses. In light of previous findings and the tenets of both attachment theory and the STM, we formulated the following hypotheses:

The objectives of the present study were (1) to examine the relationship between insecure romantic attachment and mental health among patients with cardiovascular disease and their spouses, and (2) to explore the potential mediating effect of common dyadic coping. We hypothesized that their own romantic attachment insecurity would be related to lower mental health outcomes in cardiac patients and their partners (H1), that dyadic coping would mediate the path from one’s own romantic attachment security to one’s own mental health (H2), that own insecure romantic attachment would be associated with partners’ lower mental health outcomes (H3), and that dyadic coping would mediate the link from one’s own romantic attachment insecurity to partners’ mental health (H4). The first two hypotheses represent actor effects. The second two represent partner effects.

Our results provide partial support for H1, H2, and H4. As predicted in H1, we found that greater romantic attachment anxiety was related to more symptoms of depression and anxiety in both patients and partners. Specifically, participants’ (patients and partners) own romantic attachment anxiety predicted their own self-reported struggles with their mental health. Contrary to our predictions, we did not detect any effects for attachment avoidance. Concerning H2, we noted that patients’ and spouses’ own romantic attachment avoidance was associated with their own dyadic coping, such that greater avoidance was linked to less CDC. However, we did not detect any relation between attachment anxiety and dyadic coping nor did we uncover any mediating effect of dyadic coping. H3 was not supported: There was no link between individuals’ (patients and spouses) insecure romantic attachment and their partners’ mental health. Finally, there was some support for H4: Spouses’ higher attachment avoidance was related to patients’ lower dyadic coping, but spouses’ attachment anxiety was not related to patients’ dyadic coping. There was also no link between dyadic coping and mental health in any of the analyses.

There are several strengths to this study, including the measurement of both patients’ and spouses’ perceptions of their romantic attachment, common dyadic coping, and mental health; the use of validated questionnaires with robust psychometric properties; the addition of common dyadic coping as a potential mediating construct; and the analysis of patient and partner effects using as sophisticated model that accounts for interdependent relationships. Although this research addresses gaps in the literature and has theoretical and methodological strengths, there are limitations that should be noted. First, the cross-sectional nature of this study precludes any observations of causality, although it is still advantageous preliminary work that is necessary before embarking on more resource-intensive experimental research (Czajkowski et al., 2015). Second, despite our best attempts to recruit a diverse sample, the characteristics of the participants in this study were relatively homogenous, including mostly white, well-educated, male patients and female spouses aged 60 and over, with an average relationship length of more than 30 years. It is possible that people in this age group perceive and experience attachment and mental health differently than do people who are younger or older, just as it is possible, even likely, that people of sociocultural or ethnic backgrounds that do not represent the majority might have different experiences with attachment and mental health (Guillory, 2021). Our understanding of the associations between attachment and mental health would benefit greatly from samples that are more diverse in terms of socioeconomic status, race/ethnicity, education, sex, and age, as these factors are all implicated in the association between relationship quality and mental health (Lincoln and Chae, 2010; Yu and Zhang, 2020).

Furthermore, the participant population was recruited from a cardiac rehabilitation program. Participation rates for cardiac rehabilitation at our center are approximately 50% of those diagnosed with cardiovascular disease, and these participants typically have fewer comorbidities than do non-participants (Peters and Keeley, 2018). In this study, only a small subset of participants (13.5%) reported poor mental health, which may partially explain some of the null findings. Low levels of anxiety and depression may also be a result of participation in a cardiac rehabilitation program that included elements of stress management; therefore, it is plausible that participants’ anxiety and depression symptoms would be lower than those who did not participate in cardiac rehabilitation (who were not participants in this study). Finally, although the CDC subscale has been found to have good psychometric properties, it is plausible that its brevity may not have entirely captured how relational coping efforts possibly influence mental health, hence limiting the measured variance. The same holds true for our measures of attachment and mental health: all were brief and might not provide a completely accurate snapshot of the experiences of participants. Future research would do well to include measures from multiple sources (e.g., diagnoses from mental health professionals, observational measures of couple interactions) beyond the self-report measures included here. It would also be good to compare couples facing heart disease with couples in good physical health to determine which aspects of dyadic coping, attachment and mental health might be specific to the context of heart disease.

The effects in this cross-sectional study, although only small to moderate, indicate that insecure attachment, particularly attachment anxiety, may be linked to symptoms of depression and anxiety following a cardiovascular event or diagnosis. This provides scientific justification for further examination of attachment and mental health in clinical samples using longitudinal designs. Our results additionally revealed that common dyadic coping was not a significant mediator of the relationship between insecure romantic attachment and depression and anxiety. Further research investigating common dyadic coping and other relational factors (e.g., relationship satisfaction, intimacy, mutual disclosure) is now warranted. Relationship quality and coping skills are modifiable factors that are sensitive to intervention (Berry et al., 2017). They therefore represent an excellent opportunity for secondary prevention.

Although no partner effects for our main objective were observed in this study, decades of research have illuminated the potential “spill over” effects of an individual’s health on family members and shown that these effects are reciprocal (Wittenberg and Prosser, 2016). Thus, there is still a continued need for research to assess and analyze patient and spousal effects interdependently. In a cardiovascular disease context, this shift from individual-focused to a dyadic- or family-focused health management has led to increased understanding of the ways that spouses’ behaviors and outcomes influence those of patients and vice-versa (Bidwell et al., 2023). Accumulating research also seems to indicate that couples-or family-based interventions (that target both the patient and a loved one) yield sustainable effects on mental health, quality of life, and cardiometabolic risk factors (Helgeson et al., 2018). To further inform such interventions, future research, particularly longitudinal studies, should aim to clarify the processes and potential mechanisms that explain the interconnections between couples’ physical and mental health, in which relational factors appear to figure prominently.