Abstract
While divorced or living alone, patients with stable cardiovascular disease are at increased risk for adverse cardiovascular events. The importance of marital status following a myocardial infarction (MI) is less clear. We hypothesized that marital status may affect cardiovascular outcomes following MI. We analyzed outcomes among patients with MI who underwent percutaneous coronary intervention from the Canadian Observational Antiplatelet Study (COAPT). Marital status was categorized into 3 groups: married/common‐law patients living together; never married; and divorced, separated, or widowed patients. Patients were followed for 15 months and our primary outcome was the occurrence of a major adverse cardiovascular event (MACE), defined as a composite of mortality, repeat acute MI, stroke, or urgent coronary revascularization. Multivariable logistic regression models were performed, with married/common‐law patients living together considered the reference group. Among 2100 patients included in analyses, 1519 (72.3%) were married/common‐law patients living together, 358 (17.1%) were separated/divorced/widowed, and 223 (10.6%) patients were never married. Dual antiplatelet therapy use after 15 months was similar across groups (75.4%, 77.8%, and 73.6%, respectively). The risk of MACE after 15 months was similar among married patients living together (12.7%; referent) compared with patients who were never married (13.9%; adjusted odds ratio: 1.09, 95% confidence interval: 0.58–2.07, P = 0.79) and patients separated/divorced/widowed (14.3%; adjusted odds ratio: 0.71, 95% confidence interval: 0.40–1.25, P = 0.23). Similarly, the risk of individual endpoints, including mortality, was similar across the 3 groups. Among patients stabilized following an MI, we found no association between marital status and 15‐month outcomes.
Keywords: Cardiovascular Disease, Marital Status, Myocardial Infarction, Screening Measures
1. INTRODUCTION
Approximately 17 million deaths globally are caused by cardiovascular disease (CVD), with 2.8 million hospitalizations annually related to cardiovascular (CV) events in Canada alone.1, 2 CVD morbidity and mortality have also been associated with nontraditional risk factors such as living arrangement and marital status. For example, individuals living alone demonstrated an elevated risk of CV death in an international cohort of middle‐aged outpatients with or at risk of CVD.3 In comparison with those who are married, accumulated divorces over an individual's lifetime have also been associated with an elevated risk of incident myocardial infarction (MI).4 Other studies suggest that marital quality may play a greater role than marital status in CV risk.5, 6, 7 One aspect of such risk may be related to the duration or early discontinuation of dual antiplatelet therapy (DAPT). Some studies have found that marital status/living arrangement is associated with early discontinuation of DAPT, whereas others have not found this relationship, but rather observed a significant decline in adherence after 6 months.8 However, these studies have predominantly been conducted in stable outpatients; the impact of marital status on CV outcomes in a contemporary population at the time of MI presentation may have prognostic importance in identifying at‐risk patients at higher risk for adverse CV outcomes.9, 10
Conceptually, we hypothesize that marital status may be a surrogate measure of social support present at the time of an MI that may impact subsequent outcomes during this vulnerable period. We therefore explored the prognostic importance of marital status on subsequent CV risk in a real‐world cohort of MI patients treated with percutaneous coronary intervention (PCI) in the Canadian Observational Antiplatelet Study (COAPT).
2. METHODS
We conducted a retrospective cohort analysis of MI patients undergoing PCI in the COAPT study. The design, baseline characteristics, and primary results of COAPT have been reported previously.11, 12, 13, 14 Briefly, COAPT was an observational study of 2179 adult MI patients who were eligible and recruited from 26 Canadian PCI‐capable hospitals. Index MI hospitalization details were collected and data extracted from medical‐record review by trained abstractors. Patient care was provided according to standard practice. Post‐discharge study follow‐up was conducted via telephone by trained personnel at 6 weeks and 6, 12, and 15 months following admission for the qualifying index MI. Data were collected related to events of interest, including major adverse cardiac events (MACE) and bleeding events. Based on patient reporting during follow‐up and review of available medical records, if a MACE or bleeding event was suspected, additional collection of relevant information was obtained and verified by the investigator. Institutional review boards at each site approved this study and all patients provided written informed consent to participate. The investigations were in accordance with the Declaration of Helsinki.
Patients were categorized into 3 groups according to marital status: (1) married/common‐law patients living together (referent); (2) never married; and (3) separated, divorced, or widowed. We elected to compare results with the married/common‐law patients living together group based on prior data suggesting this group had the lowest adjusted CV risk.4, 9, 10, 15, 16, 17, 18, 19, 20 We grouped patients who were separated, divorced, or widowed based on publications suggesting that this group was often at a higher risk for CV events.9, 19, 20, 21, 22, 23, 24
The primary outcome of this analysis was MACE 4 at 15 months, defined as a composite of all‐cause death, repeat MI, stroke, or urgent coronary revascularization (UCR). The outcomes for MI and stroke were reported as first fatal or nonfatal events. Secondary outcomes included an expanded MACE endpoint (MACE+) that additionally included newly diagnosed or worsening heart failure (HF) or cardiogenic shock and stent thrombosis. We also studied individual components of the primary outcome, bleeding events, and all‐cause hospitalizations. Finally, we explored self‐reported adherence with DAPT.
2.1. Statistical analysis
Baseline clinical characteristics, treatments, and clinical outcomes are presented with numbers and percentages for categorical variables, and median and interquartile range (IQR) for continuous variables. Comparison groups were performed using Mantel–Haenszel χ2, Cochran‐Armitage trend, and Jonckheere‐Terpstra tests where appropriate. The risk outcome associated with marital status was assessed using logistic regression models to derive odds ratios (OR) and 95% confidence intervals (CI). A sensitivity time‐to‐event Cox proportional hazard analysis was conducted for the composite of death, re‐MI, or UCR among patients with known event dates, omitting stroke. Multivariable models were adjusted for sex, age, body mass index, systolic blood pressure, creatinine, race/ethnicity, education (highest level achieved of ≤eighth grade, high school [some or graduated], or university [some or baccalaureate/bachelor degree, or postgraduate education]), insurance coverage (none/private/public), and history of smoking, hypertension, dyslipidemia, prior PCI/coronary bypass graft surgery, diabetes mellitus, MI, stroke, HF, and aspirin use at baseline. Models were constructed considering marital status as an ordinal variable (in order of married patients living together; never married; and separated, divorced, or widowed) based on prior data suggesting an adverse marital event may confer poorer health outcomes.5, 6, 7 We also modeled marital status as a categorical variable in a sensitivity test. We explored a subgroup analysis stratified by sex using unadjusted relative risk regression. Interaction terms were introduced into the models to test for potential effect modification of marital status by sex.
Because 40% of covariate data were lost in the complete case analysis, we further performed multiple imputations to confirm the robustness of our findings. Missing data were imputed using the Markov chain Monte Carlo method to derive 5 complete datasets, which were then combined to generate valid inferences. The Firth bias correction for penalized likelihood estimation method was used to address the quasi‐complete separation of data points where required. Statistical analyses were conducted using SAS version 9.4 (SAS Institute, Inc., Cary, NC), and all tests were 2‐sided with P < 0.05 considered significant.
3. RESULTS
3.1. Participant characteristics by marital status/living arrangement
A total of 2179 patients presenting with MI were enrolled in the overall study, of whom 2100 (median age, 61 years; 78.7% male) had complete data on marital status and were included in this analysis (Figure 1). The majority of patients (n = 1519; 72.3%) were married or common‐law and living together, whereas 358 (17.1%) were separated, divorced, or widowed, and 223 (10.6%) patients were never married. Baseline characteristics of the patient population categorized by marital status are presented in Table 1. Compared with married/common‐law patients living together, the never‐married group tended to be younger and predominantly male. Separated, divorced, or widowed patients were frequently older and female and had more comorbidities such as active tobacco use, hypertension, prior HF, and stroke; these patients also were less frequently prescribed aspirin treatment at presentation and discharge.
Figure 1.
Study flow diagram of eligible patients categorized by marital status
Table 1.
Baseline characteristics of patients by marital status
| Characteristics (%, Mean, Median) | Married/Common Law and Living Together, n = 1519 | Never Married, n = 223 | Separated, Divorced, or Widowed, n = 358 | P for Trend |
|---|---|---|---|---|
| Age, y | 61 (54–68) | 55 (49–64) | 63 (55–73) | 0.40 |
| Male sex | 1252 (82.4) | 189 (84.8) | 212 (59.2) | <0.0001 |
| Caucasian race | 1360 (90.0) | 198 (90.8) | 327 (92.1) | 0.22 |
| BMI, kg/m2 | 28.4 (25.5–31.6) | 28.2 (25.4–31.9) | 28.1 (24.5–31.5) | 0.21 |
| SBP, mm Hg | 138 (120–157) | 138 (120–157) | 138 (120–157) | 0.54 |
| Heart rate, bpm | 74 (64–86) | 78 (68–91) | 75 (65–89) | 0.01 |
| Cr, μmol/L | 82 (70–96) | 83 (71–95) | 80 (66.5–96) | 0.20 |
| Insurance, n/N | 0.75 | |||
| None | 126/1324 (9.5) | 21/202 (10.4) | 39/321 (12.1) | |
| Private | 223/1324 (16.8) | 21/202 (10.4) | 36/321 (11.2) | |
| Provincial/federal | 975/1324 (73.6) | 160/202 (79.2) | 246/321 (76.6) | |
| Education status, n/N | <0.0001 | |||
| ≤8th grade | 84/1251 (6.7) | 11/191 (5.8) | 33/317 (10.4) | |
| High school (some or graduate) | 606/1251 (48.4) | 101/191 (52.9) | 192/317 (60.6) | |
| University (some) | 310/1251 (24.8) | 46/191 (24.1) | 61/317 (19.2) | |
| Baccalaureate/bachelor degree | 156/1251 (12.5) | 19/191 (9.9) | 25/317 (7.9) | |
| Postgraduate | 95/1251 (7.6) | 14/191 (7.3) | 6/317 (1.9) | |
| MI type | 0.28 | |||
| NSTEMI | 577 (38.1) | 86 (38.6) | 148 (41.3) | |
| STEMI | 938 (61.9) | 137 (61.4) | 210 (58.7) | |
| Medical history | ||||
| History of smoking | 918 (70.3) | 144 (71.6) | 238 (78.0) | 0.01 |
| DM | 276 (18.2) | 34 (15.2) | 78 (21.8) | 0.23 |
| HF | 26 (1.7) | 4 (1.8) | 14 (3.9) | 0.01 |
| HTN | 762 (50.2) | 98 (43.9) | 202 (56.4) | 0.12 |
| Prior MI | 527 (34.7) | 83 (37.2) | 141 (39.4) | 0.08 |
| Dyslipidemia | 716 (47.1) | 91 (40.8) | 172 (48.0) | 0.88 |
| AF | 38 (2.5) | 9 (4.0) | 14 (3.9) | 0.10 |
| Prior stroke | 22 (1.4) | 6 (2.7) | 12 (3.4) | 0.01 |
| TIA | 20 (1.3) | 3 (1.3) | 15 (4.2) | 0.00 |
| Prior PCI/CABG | 423 (27.8) | 66 (29.6) | 113 (31.6) | 0.15 |
| ASA index | 1419 (93.4) | 216 (96.9) | 328 (91.6) | 0.51 |
| Antithrombotic therapy at discharge | ||||
| ASA | 1416 (93.4) | 216 (96.9) | 326 (91.6) | 0.40 |
| OAC | 195 (12.9) | 35 (15.7) | 50 (14.0) | 0.41 |
| Discharged on ADP inhibitors | 0.32 | |||
| Clopidogrel | 1106 (73.0) | 153 (68.6) | 274 (76.8) | |
| Ticagrelor/prasugrel | 410 (27.0) | 70 (31.4) | 83 (23.2) | |
| DAPT duration, d | 405 (365–463) | 407.5 (366–468.5) | 407 (357–464) | 0.87 |
| DAPT completion ≥12 mo | 1068 (75.4) | 168 (77.8) | 240 (73.6) | 0.54 |
Abbreviations: ADP, adenosine diphosphate; AF, atrial fibrillation; ASA, acetylsalicylic acid (aspirin); BMI, body mass index; CABG, coronary artery bypass grafting; Cr, creatinine; DAPT, dual antiplatelet therapy; DM, diabetes mellitus; HF, heart failure; HTN, hypertension; IQR, interquartile range; MI, myocardial infarction; NSTEMI, non–ST‐segment elevation myocardial infarction; OAC, oral anticoagulant; PCI, percutaneous coronary intervention; SBP, systolic blood pressure; STEMI, ST‐segment elevation myocardial infarction; TIA, transient ischemic attack.
Data are presented as n (%) or median (IQR).
Duration of DAPT was similar between each group of patients according to marital status and living arrangement, with an overall median duration of therapy of 406 days (P = 0.87). Across all 3 marital‐status groups, adherence with DAPT after 12 monthsnthsnths remained high at a rate of 75.4% for the group of married patients living together, 77.8% for the never‐married group, and 73.6% for the separated, divorced, or widowed group (P = 0.54).
By 15 months, 8.1% of patients were lost to follow‐up. Among patients with complete ascertainment of vital status, a total of 52 deaths were observed, with a significant increase in mortality across groups of marital status (married/common law, 2.0% [referent]; never married, 3.1%; divorced/separated/widowed, 4.2%; P = 0.04). However, after multivariable adjustment, the differences were no longer significant among never‐married patients (adjusted OR: 1.22, 95% CI: 0.33–4.50, P = 0.76) and separated, divorced, or widowed patients (adjusted OR: 0.70, 95% CI: 0.26–1.90, P = 0.49; Table 2).
Table 2.
Rates of CV outcomes of patients during 15 month follow‐up period by marital status
| Outcome | Event Rate | ||||||
|---|---|---|---|---|---|---|---|
| (1) Married/ Common Law and Living Together, n = 1519 | (2) Never Married, n = 223 | (3) Separated, Divorced/Widowed, n = 358 | Unadjusted OR (2) vs (1), (3) vs (1), OR (95% CI) | P Value | Adjusted OR (2) vs (1), (3) vs (1), OR (95% CI) | P Value | |
| Death | 30 (2.0) | 7 (3.1) | 15 (4.2) | 1.61 (0.70–3.71) | 0.26 | 1.22 (0.33–4.50) | 0.76 |
| 2.17 (1.16–4.08) | 0.02 | 0.70 (0.26–1.90) | 0.49 | ||||
| MACE, 4 eventsa | 193 (12.7) | 31 (13.9) | 51 (14.3) | 1.11 (0.74–1.67) | 0.62 | 1.09 (0.58–2.07) | 0.79 |
| 1.14 (0.82–1.59) | 0.44 | 0.71 (0.40–1.25) | 0.23 | ||||
| MACE, 7 eventsb | 249 (16.4) | 44 (19.7) | 63 (17.6) | 1.25 (0.84–1.79) | 0.21 | 1.24 (0.73–2.11) | 0.43 |
| 1.09 (0.80–1.48) | 0.58 | 0.67 (0.41–1.10) | 0.11 | ||||
| MI | 126 (8.3) | 20 (9.0) | 31 (8.7) | 1.09 (0.66–1.79) | 0.73 | 1.48 (0.63–3.49) | 0.37 |
| 1.05 (0.70–1.58) | 0.82 | 0.92 (0.41–2.10) | 0.85 | ||||
| Stroke | 12 (0.8) | 4 (1.8) | 2 (0.6) | 2.29 (0.73–7.18) | 0.15 | 3.47 (1.04–11.57) | 0.04 |
| 0.71 (0.16–3.17) | 0.65 | 0.74 (0.19–2.88) | 0.66 | ||||
| UCR | 117 (7.7) | 15 (6.7) | 27 (7.5) | 0.86 (0.50–1.51) | 0.61 | 0.59 (0.22–1.58) | 0.30 |
| 0.98 (0.63–1.51) | 0.92 | 0.55 (0.25–1.22) | 0.14 | ||||
| HF/shock | 79 (5.2) | 18 (8.1) | 22 (6.1) | 1.60 (0.94–2.73) | 0.08 | 1.65 (0.80–3.40) | 0.17 |
| 1.19 (0.73–1.94) | 0.48 | 0.75 (0.37–1.52) | 0.42 | ||||
| Stent thrombosis | 30 (2.0) | 3 (1.3) | 3 (0.8) | 0.68 (0.21–2.24) | 0.52 | 0.39 (0.05–3.29) | 0.39 |
| 0.42 (0.13–1.38) | 0.15 | 0.62 (0.16–2.46) | 0.50 | ||||
| Major bleeding | 15 (1.0) | 3 (1.3) | 4 (1.1) | 1.37 (0.39–4.76) | 0.62 | 1.65 (0.46–5.60) | 0.45 |
| 1.13 (0.37–3.43) | 0.83 | 1.61 (0.54–4.81) | 0.39 | ||||
Abbreviations: ASA, acetylsalicylic acid (aspirin); BMI, body mass index; CABG, coronary artery bypass grafting; CI, confidence interval; Cr, creatinine; CV, cardiovascular; DM, diabetes mellitus; HF, heart failure; HTN, hypertension; MACE, major adverse cardiovascular events; MI, myocardial infarction; OR, odds ratio; PCI, percutaneous coronary intervention; SBP, systolic blood pressure; UCR, urgent coronary revascularization.
Data are presented as n (%).
Multivariable models were adjusted for sex, age, BMI, SBP, Cr, race/ethnicity, education, insurance, and history of smoking, HTN, dyslipidemia, prior PCI/CABG, DM, MI, stroke, HF, and ASA use at baseline.
MACE 4 includes death, MI, UCR, and stroke.
MACE 7 includes death, MI, UCR, and stroke plus HF/shock and stent thrombosis.
Considering marital status as an ordinal variable, there was no significant increase in MACE across marital groups, with 193 (12.7%) events in the married/common‐law group; 31 events (13.9%) in the never‐married group; and 51 (14.3%) events in the separated, divorced, and widowed group (P for trend = 0.40). After multivariable adjustment, we observed no independent increased risk of MACE or the expanded MACE definition associated with marital status (Table 2). Multiple imputation provided similar results. In women, a potential of divergent risk for HF/shock was indicated, with a greater risk in separated, divorced, or widowed women when compared with men following sex‐stratified analyses (P for interaction = 0.03; Table 3). The survival sensitivity analysis suggested no difference in risk estimates by marital status for the composite of death, re‐MI, or UCR. Results were similar when marital status was modeled as a categorical variable (data not shown).
Table 3.
Sex‐stratified rates of CV outcomes of patients during 15‐month follow‐up period by marital status
| Outcome | (1) Married/Common Law and Living Together, n = 1519 | (2) Never Married, n = 223 | (3) Separated, Divorced, or Widowed, n = 358 | Unadjusted RR (2) vs (1), RR (95% CI) | P Value | Unadjusted RR (3) vs (1), RR (95% CI) | P Value | P for Interaction (Sex* Groups) |
|---|---|---|---|---|---|---|---|---|
| Death | 0.12 | |||||||
| M | 24 (1.9) | 6 (3.2) | 4 (1.9) | 1.66 (0.69–4.00) | 0.26 | 0.98 (0.35–2.80) | 0.98 | |
| F | 6 (2.3) | 1 (2.9) | 11 (7.5) | 1.31 (0.16–10.55) | 0.80 | 3.35 (1.27–8.88) | 0.02 | |
| MACE, 4 eventsa | 0.23 | |||||||
| M | 152 (12.1) | 29 (15.3) | 21 (9.9) | 1.26 (0.88–1.82) | 0.21 | 0.82 (0.53–1.26) | 0.36 | |
| F | 41 (15.4) | 2 (5.9) | 30 (20.6) | 0.38 (0.10–1.51) | 0.17 | 1.34 (0.87–2.05) | 0.18 | |
| MACE, 7 eventsb | 0.13 | |||||||
| M | 200 (16.0) | 38 (20.1) | 27 (12.7) | 1.26 (0.92–1.72) | 0.15 | 0.80 (0.55–1.16) | 0.24 | |
| F | 49 (18.4) | 6 (17.7) | 36 (24.7) | 0.96 (0.45–2.07) | 0.92 | 1.34 (0.92–1.96) | 0.13 | |
| MI | 0.41 | |||||||
| M | 104 (8.3) | 19 (10.1) | 15 (7.1) | 1.21 (0.76–1.93) | 0.42 | 0.85 (0.51–1.43) | 0.55 | |
| F | 22 (8.2) | 1 (2.9) | 16 (11.0) | 0.36 (0.05–2.56) | 0.31 | 1.33 (0.72–2.45) | 0.36 | |
| Stroke | 0.53 | |||||||
| M | 9 (0.7) | 3 (1.6) | 0 (0.0) | 2.21 (0.60–8.08) | 0.23 | 0.00 | 1.00 | |
| F | 3 (1.1) | 1 (2.9) | 2 (1.4) | 2.62 (0.28–24.46) | 0.40 | 1.22 (0.21–7.21) | 0.83 | |
| UCR | 0.24 | |||||||
| M | 86 (6.9) | 15 (7.9) | 15 (7.1) | 1.16 (0.68–1.96) | 0.59 | 1.03 (0.61–1.75) | 0.91 | |
| F | 31 (11.6) | 0 (0.0) | 12 (8.2) | 0.00 | 1.00 | 0.71 (0.38–1.34) | 0.29 | |
| HF/shock | 0.03 | |||||||
| M | 67 (5.4) | 14 (7.4) | 7 (3.3) | 1.38 (0.79–2.41) | 0.25 | 0.62 (0.29–1.33) | 0.22 | |
| F | 12 (4.5) | 4 (11.8) | 15 (10.3) | 2.62 (0.89–7.66) | 0.08 | 2.29 (1.10–4.75) | 0.03 | |
| Stent thrombosis | 0.97 | |||||||
| M | 22 (1.8) | 3 (1.6) | 1 (0.5) | 0.90 (0.27–3.00) | 0.87 | 0.27 (0.04–1.98) | 0.20 | |
| F | 8 (3.0) | 0 (0.0) | 2 (1.4) | 0.00 | 1.00 | 0.46 (0.10–2.12) | 0.32 | |
| Major bleeding | 0.76 | |||||||
| M | 14 (1.1) | 2 (1.1) | 3 (1.4) | 0.95 (0.21–4.13) | 0.94 | 1.27 (0.37–4.37) | 0.71 | |
| F | 1 (0.4) | 1 (2.9) | 1 (0.7) | 7.85 (0.50–122.68) | 0.14 | 1.83 (0.11–29.02) | 0.67 |
Abbreviations: CI, confidence interval; CV, cardiovascular; F, female; HF, heart failure; M, male; MACE, major adverse cardiovascular events; MI, myocardial infarction; RR, relative risk; UCR, urgent coronary revascularization.
Data are presented as n (%).
MACE 4 includes death, MI, UCR, and stroke.
MACE 7 includes death, MI, UCR, and stroke plus HF/shock and stent thrombosis.
4. DISCUSSION
This study aimed to identify whether patients who were never married or divorced, separated, or widowed at the time of presenting with an MI were at higher risk of recurrent CV outcomes compared with married/common‐law patients living together. In our study of patients presenting with MI treated with PCI, there did not appear to be an independent association between marital status and recurrent CV events at 15 months. Though previous studies suggest that early post‐MI may be a period of particular vulnerability during which social supports, including marital status, may influence subsequent risk of CV events, our data did not support this hypothesis. Marital status also showed no differential association with DAPT adherence following MI. Following logistic regression and Cox proportional hazards model comparison, no observed differences in outcomes were seen between marital‐status groups. Nevertheless, there was a significant interaction between marital status and sex on subsequent CV outcomes. Women who were separated, divorced, or widowed were at greater risk for HF/shock events compared with married women, whereas no differences in risk were observed among the men in COAPT. However, this finding requires further study and replication.
Prior literature has primarily focused on the impact of marital status on the risk of an incident MI and mortality rather than observing various CV events following nonfatal MI.9, 10, 25 For instance, in a Finnish population, those living alone without a partner were at 30% to 50% higher odds for first‐day fatality after hospitalization for MI compared with those who were married.17 In an analysis of stable patients with CVD followed in the Reduction of Atherothrombosis for Continued Health (REACH) registry, the risk associated with living alone and CV risk was dependent on age, possibly because of what “living arrangement” represents at different stages of life.3 Beyond impacting prognosis, marital status may be associated with the duration of time that individuals wait to seek hospital care for acute chest pain.15 Single men, single women, and married women tend to delay seeking hospital care longer than married men.16, 26 Furthermore, Coyne et al. showed that the quality of a marital relationship was independently predictive of survival after HF.5, 27 In an attempt to highlight the importance of life stressors on CV health, Dupre et al identified that an individual's current marital status was not associated with the incidence of MI, but rather changes in marital status, such as an accumulated number of divorces, were strongly associated with risk for MI.4 Additionally, in a study analyzing MI patients of the MONICA/KORA data registry, it was observed that being married did not provide protective effects for CV events in patients post‐MI.10 However, it was found that in sex‐stratified groups, married patients with hyperlipidemia age < 60 years had reduced long‐term all‐cause mortality rates. In concordance with our findings, when considering 1 year post‐MI, they observed minimal event rates in all marital‐status groups.10 In multiple studies looking at low perceived social support between patients married and unmarried following MI, it was found that the absence of a partner at baseline was a predictor for death, specifically in‐hospital mortality, and recurrent MI during long‐term follow‐up.17, 18, 28, 29, 30, 31
Several factors may explain our discrepant findings in comparison with these prior results. First, our analyses likely had limited statistical power due to the modest cohort size and relatively short follow‐up period post‐MI. Indeed, the wide CIs indicate that we cannot draw definitive conclusions. Consequently, because patients were predominantly male (82%) and ~60% of the separated, divorced, or widowed participants were also male, we decided to sex‐stratify our results to identify any association between marital status and CV outcomes by sex. In married vs unmarried men and women, past studies highlighted differences in outcomes following an MI.21, 24, 25, 29, 30, 31 It was observed that women receive less informal caregiver support than do men, with worse outcomes, 1 to 12 months following an MI.25 However, in contrast with many of the prior studies that controlled simply for certain socioeconomic determinants of health, such as income and education, we additionally adjusted for differences in baseline comorbidities that may have attenuated any association with marital status. Income was unavailable for analysis; instead, insurance coverage was used as an acceptable marker for socioeconomic status.32, 33 Another consideration is that marital status may have disparate impact in a primary‐prevention setting compared with a secondary‐prevention setting. Furthermore, following an MI, patients' marital status may influence motivation to engage in healthy behaviors and to adhere to medical therapy to a greater extent compared with primary prevention.26, 34, 35
It should not go unnoticed that more preexisting risk factors were present among patients who were never married or were separated, divorced, or widowed, which may place them at higher risk for CVD. Unmarried patients (patients never married, separated, divorced, or widowed) were a particularly vulnerable group, as they tended to be older and had a prior history of smoking, HF, and stroke. Compared with married individuals, the mental and physical health of separated, divorced, or widowed individuals may be poorer. As a result, both widowed and divorced patients are reported to utilize more healthcare because they suffer from chronic conditions.26, 27, 28 Also, risk associated with being widowed, divorced, or separated may be time‐dependent from the point of change in marital status. For instance, data suggest that the negative effects of divorce tend to last longer than those of widowhood.24, 25
4.1. Study limitations
There are other limitations worth noting. Categorization of marital status was made only at baseline and we cannot comment on the association of marital history within each group and potential changes to status with outcomes over time. Beyond marital status, we do not have sufficient detail to further describe patients' social support where added care‐provider networks may exist.30, 31, 35 Considering follow‐up, 15 months may not be sufficient time to observe significant effects of marital status on CV risk following MI, with other reports noting significant effects at 3 years among patients with CVD and 10 years among patients with hyperlipidemia, respectively.9, 10 Moreover, we did not account for recurrent CV events in our analysis, and this may underestimate the total burden of MI and stroke. Additionally, some fatal MI outcomes may have been underdiagnosed, particularly without central adjudication of events. There was also a modest degree of missing data (40%) on covariates and vital status that limited the statistical power of exploratory subgroup analyses. Finally, consecutive enrollment was encouraged but informed consent was required, and we did not use a screening log to determine the number of patients who were eligible but declined to participate. Enrollment focused on patients expected to be discharged, and only after PCI, such that the COAPT cohort likely reflects a selected, lower‐risk population. Because we analyzed a select group of patients who underwent intervention post‐MI, our cohort may not be representative of the general population of post‐MI patients.
Our study helps to address an important gap in the area of social factors and quality of care. We hope our findings strengthen future research in this field, including particularly those undertaking systematic reviews and meta‐analyses of this clinically relevant topic. Moreover, screening patients at baseline based on their marital status may provide information on targeted support systems for at‐risk patients.
5. CONCLUSION
Among Canadian post‐MI patients, marital status was not associated with near‐term adverse CV events. Nevertheless, patients who are never married, separated, divorced, or widowed appear to have a higher burden of comorbid risk factors influencing their prognosis that may benefit from targeted support.
ACKNOWLEDGEMENTS
The authors thank Sue Francis of St. Michael's Hospital and Susan Webber, RN, of the Women's College Research Institute for their administrative contributions to this manuscript.
Conflicts of interest
Akshay Bagai has received speaker/consulting honoraria and/or research grant support from AstraZeneca. Andrew T. Yan has received speaker/consulting honoraria and/or research grant support from Sanofi‐Aventis. Shaun G. Goodman has received speaker/consulting honoraria and/or research grant support from Eli Lilly Canada, AstraZeneca, Bristol‐Myers Squibb, and Sanofi‐Aventis. Shamir R. Mehta has received research grant support from Eli Lilly Canada. Harold N. Fisher is a medical officer and receives salary from Eli Lilly Canada. Eric A. Cohen has received speaker/consulting honoraria and/or research grant support from Abbott, Roche, Sanofi‐Aventis, Eli Lilly, and Medtronic. Thao Huynh has received speaker/consulting honoraria and/or research grant support from Sanofi, AstraZeneca, Bristol‐Myers Squibb, Merck, Boehringer‐Ingelheim, Bayer, Janssen, Eli Lilly, and Amgen. Warren J. Cantor has received speaker/consulting honoraria and/or research grant support from Roche and Sanofi. Jean‐Pierre Déry has received speaker/consulting honoraria and/or research grant support from Eli Lilly Canada, AstraZeneca, Sanofi, and Bristol‐Myers Squibb. Robert C. Welsh has received speaker/consulting honoraria and/or research grant support from Eli Lilly Canada and AstraZeneca. Jacob A. Udell has received speaker/consulting honoraria from Boehringer‐Ingelheim, Janssen, Merck, Novartis, and Sanofi Pasteur and research grants from AstraZeneca, Novartis and Sanofi. The authors declare no other potential conflicts of interest.
Ghosh‐Swaby OR, Tan M, Bagai A, et al. Marital status and outcomes after myocardial infarction: Observations from the Canadian Observational Antiplatelet Study (COAPT). Clin Cardiol. 2018;41:285–292. 10.1002/clc.22901
Funding information This study was supported by a grant from the Heart and Stroke Foundation of Canada (#G‐15‐0009034) and by a Heart and Stroke Foundation of Ontario/University of Toronto Polo Chair in Cardiology Young Investigator Award. Ms. Ghosh‐Swaby was supported by a Women's College Research Institute Summer Research Studentship Award. Dr. Udell was supported in part by a Heart and Stroke Foundation of Canada National New Investigator/Ontario Clinician Scientist Award, Women's College Research Institute and the Department of Medicine, Women's College Hospital; Peter Munk Cardiac Centre, University Health Network; and Department of Medicine and Heart and Stroke Richard Lewar Centre of Excellence in Cardiovascular Research, University of Toronto. Dr. Goodman was supported by the Heart and Stroke Foundation of Ontario in his role as Heart and Stroke Foundation (Polo) Chair at the University of Toronto. The Canadian Observational Antiplatelet Study (COAPT) was sponsored by Eli Lilly and Co. and Daiichi‐Sankyo. Site management was performed by the Canadian Heart Research Centre. The results and conclusions of this study are those of the authors and should not be attributed to any of the sponsoring agencies. The study's sponsors had no role in the design and conduct of this study, in the analysis and interpretation of the data, or in the preparation or approval of the manuscript for publication.
REFERENCES
- 1. Canadian Institute for Health Information . Highlights of 2009–2010: Inpatient hospitalizations and emergency department visits. https://secure.cihi.ca/estore/productFamily.htm?pf=PFC1626. Published May 12, 2011. Last accessed March 12, 2018.
- 2. World Health Organization . The atlas of heart disease and stroke. http://www.who.int/cardiovascular_diseases/resources/atlas/en/. Published September 23, 2014. Last accessed March 12, 2018.
- 3. Udell JA, Steg PG, Scirica BM, et al; Reduction of Atherothrombosis for Continued Health (REACH) Registry Investigators. Living alone and cardiovascular risk in outpatients at risk of or with atherothrombosis. Arch Intern Med. 2012;172:1086–1095. [DOI] [PubMed] [Google Scholar]
- 4. Dupre ME, George LK, Liu G, et al. Association between divorce and risks for acute myocardial infarction. Circ Cardiovasc Qual Outcomes. 2015;8:244–251. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5. Coyne JC, Rohrbaugh MJ, Shoham V, et al. Prognostic importance of marital quality for survival of congestive heart failure. Am J Cardiol. 2001;88:526–529. [DOI] [PubMed] [Google Scholar]
- 6. Liu H, Waite L. Bad marriage, broken heart? Age and gender differences in the link between marital quality and cardiovascular risks among older adults. J Health Soc Behav. 2014;55:403–423. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7. Bennett‐Britton I, Teyhan A, Macleod J, et al. Changes in marital quality over 6 years and its association with cardiovascular disease risk factors in men: findings from the ALSPAC prospective cohort study. J Epidemiol Community Heath. 2017;71:1094–1100. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8. Czarny MJ, Nathan AS, Yeh RW, et al. Adherence to dual antiplatelet therapy after coronary stenting: a systematic review. Clin Cardiol. 2014;37:505–513. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9. Shultz WM, Hayek SS, Tahhan AS, et al. Marital status and outcomes in patients with cardiovascular disease. J Am Heart Assoc. 2017;6:1–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10. Quinones PA, Kirchberger I, Heier M, et al. Marital status shows a strong protective effect on long‐term mortality among first acute myocardial infarction survivors with diagnosed hyperlipidemia—findings from the MONICA/KORA myocardial infarction registry. BMC Public Health. 2014;14:98. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11. Bagai A, Wang TY, Goodman SG, et al; Canadian Observational Antiplatelet Study (COAPT) Investigators . Longitudinal treatment patterns with ADP receptor inhibitors after myocardial infarction: insights from the Canadian Observational Antiplatelet Study. Int J Cardiol. 2017;228:459–464. [DOI] [PubMed] [Google Scholar]
- 12. Russo JJ, Goodman SG, Bagai A, et al. Duration of dual antiplatelet therapy and associated outcomes following percutaneous coronary intervention for acute myocardial infarction: contemporary practice insights from the Canadian Observational Antiplatelet Study. Eur Heart J Qual Care Clin Outcomes. 2017;3:303–311. [DOI] [PubMed] [Google Scholar]
- 13. Déry JP, Mehta SR, Fisher HN, et al; Canadian Observational Antiplatelet Study (COAPT) Investigators . Baseline characteristics, adenosine diphosphate receptor inhibitor treatment patterns, and in‐hospital outcomes of myocardial infarction patients undergoing percutaneous coronary intervention in the prospective Canadian Observational Antiplatelet Study (COAPT). Am Heart J. 2016;181:26–34. [DOI] [PubMed] [Google Scholar]
- 14. Sra S, Tan MK, Mehta SR, et al; Canadian Observational Antiplatelet Study (COAPT) Investigators . Ischemic and bleeding events in patients with myocardial infarction undergoing percutaneous coronary intervention who require oral anticoagulation: insights from the Canadian Observational Antiplatelet Study. Am Heart J. 2016;180:82–89. [DOI] [PubMed] [Google Scholar]
- 15. Atzema CL, Austin PC, Huynh T, et al. Effect of marriage on duration of chest pain associated with acute myocardial infarction before seeking care. CMAJ. 2011;183:1482–1491. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16. Austin D, Yan AT, Spratt JC, et al. Patient characteristics associated with self‐presentation, treatment delay and survival following primary percutaneous coronary intervention. Eur Heart J Acute Cardiovasc Care. 2014;3:214–222. [DOI] [PubMed] [Google Scholar]
- 17. Kilpi F, Konttinen H, Silventoinen K, et al. Living arrangements as determinants of myocardial infarction incidence and survival: a prospective register study of over 300 000 Finnish men and women. Soc Sci Med. 2015;133:93–100. [DOI] [PubMed] [Google Scholar]
- 18. Koskinen S, Joutsenniemi K, Martelin T, et al. Mortality differences according to living arrangements. Int J Epidemiol. 2007;36:1255–1264. [DOI] [PubMed] [Google Scholar]
- 19. Bulanda JR, Brown S, Yamashita T. Marital quality, marital dissolution, and mortality risk during the later life course. Soc Sci Med. 2016;165:119–127. [DOI] [PubMed] [Google Scholar]
- 20. Tanno K, Ohsawa M, Itai K. Associations of marital status with mortality from all causes and mortality from cardiovascular disease in Japanese haemodialysis patients. Nephrol Dial Transplant. 2013;28:1013–1020. [DOI] [PubMed] [Google Scholar]
- 21. Floud S, Balkwill A, Canoy D, et al. Marital status and ischemic heart disease incidence and mortality in women: a large prospective study. BMC Med. 2014;12:42. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22. Lynch JW, Kaplan GA, Cohen RD, et al. Do cardiovascular risk factors explain the relation between socioeconomic status, risk of all‐cause mortality, cardiovascular mortality, and acute myocardial infarction? Am J Epidemiol. 1996;144:934–942. [DOI] [PubMed] [Google Scholar]
- 23. Malyutina S, Bobak M, Simonova G, et al. Education, marital status, and total and cardiovascular mortality in Novosibirsk, Russia: a prospective cohort study. Ann Epidemiol. 2004;14:244–249. [DOI] [PubMed] [Google Scholar]
- 24. Molloy GJ, Stamatakis E, Randall G, et al. Marital status, gender and cardiovascular mortality: behavioural, psychological distress and metabolic explanations. Soc Sci Med. 2009;69:223–228. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25. Burg MM, Barefoot J, Berkman L, et al; ENRICHD Investigators . Low perceived social support and post–myocardial infarction prognosis in the enhancing recovery in coronary heart disease clinical trial: the effects of treatment. Psychosom Med. 2005;67:879–888. [DOI] [PubMed] [Google Scholar]
- 26. Joung IM, van der Meer JB, Mackenbach JP. Marital status and health care utilization. Int J Epidemiol. 1995;24:569–575. [DOI] [PubMed] [Google Scholar]
- 27. Wade TJ, Pevalin DJ. Marital transitions and mental health. J Health Soc Behav. 2004;45:155–170. [DOI] [PubMed] [Google Scholar]
- 28. Williams K, Umberson D. Marital status, marital transitions, and health: a gendered life course perspective. J Health Soc Behav. 2004;45:81–98. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29. Chandra V, Szklo M, Goldberg R, et al. The impact of marital status on survival after an acute myocardial infarction: a population‐based study. Am J Epidemiol. 1983;117:320–325. [DOI] [PubMed] [Google Scholar]
- 30. Kristofferzon ML, Löfmark R, Carlsson M. Perceived coping, social support, and quality of life 1 month after myocardial infarction: a comparison between Swedish women and men. Heart Lung. 2005;34:39–50. [DOI] [PubMed] [Google Scholar]
- 31. Schmaltz HN, Southern D, Ghali WA, et al. Living alone, patient sex and mortality after acute myocardial infarction. J Gen Intern Med. 2007;22:572–578. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32. DeSantis C, Jemal A, Ward E. Disparities in breast cancer prognostic factors by race, insurance status, and education. Cancer Causes Control. 2010;21:1445–1450. [DOI] [PubMed] [Google Scholar]
- 33. Fang J, Alderman MH. Does supplemental private insurance affect care of Medicare recipients hospitalized for myocardial infarction? Am J Public Health. 2004;94:778–782. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34. Wu JR, Lennie TA, Chung ML, et al. Medication adherence mediates the relationship between marital status and cardiac event‐free survival in patients with heart failure. Heart Lung. 2012;41:107–114. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 35. Trivedi RB, Ayotte B, Edelman D, et al. The association of emotional well‐being and marital status with treatment adherence among patients with hypertension. J Behav Med. 2008;31:489–497. [DOI] [PMC free article] [PubMed] [Google Scholar]