Abstract
BACKGROUND: Low socioeconomic status is associated with worse secondary prevention use and prognosis after myocardial infarction (MI). Actions for health equity improvements warrant identification of risk mediators. Therefore, we assessed mediators of the association between socioeconomic status and first recurrent atherosclerotic cardiovascular disease event (rASCVD) after MI. METHODS: In this cohort study on 1-year survivors of first-ever MI with Swedish universal health coverage ages 18 to 76 years, individual-level data from SWEDEHEART (Swedish Web System for Enhancement and Development of Evidence-Based Care in Heart Disease Evaluated According to Recommended Therapies) and linked national registries was collected from 2006 through 2020. Exposure was socioeconomic status by disposable income quintile (principal proxy), educational level, and marital status. The primary outcome was rASCVD and secondary outcomes were cardiovascular and all-cause mortality. We initially assessed the incremental attenuation of hazard ratios with 95% CIs in sequential multivariable models adding groups of potential mediators (ie, previous risk factors, acute presentation and infarct severity, initial therapies, and secondary prevention). Thereafter, the proportion of excess rASCVD associated with a low income mediated through nonparticipation in cardiac rehabilitation, suboptimal statin management, a cardiometabolic risk profile, persistent smoking, and blood pressure above target after MI were calculated using causal mediation analysis. RESULTS: Among 68 775 participants (73.8% men), 7064 rASCVD occurred during a mean 5.7-year follow-up. Income, adjusted for age, sex, and calendar year, was associated with rASCVD (hazard ratio, 1.63 [95% CI, 1.51-1.76] in the lowest versus highest income quintile). Risk attenuated most by adjustment for previous risk factors and by adding secondary prevention variables for a final model (hazard ratio, 1.38 [95% CI, 1.26-1.51]) in the lowest versus highest income quintile. The proportions of the excess 15-year rASCVD risk in the lowest income quintile mediated through nonparticipation in cardiac rehabilitation, cardiometabolic risk profile, persistent smoking, and poor blood pressure control were 3.3% (95% CI 2.1-4.8), 3.9% (95% CI, 2.9-5.5), 15.2% (95% 9.1-25.7), and 1.0% (95% CI 0.6-1.5), respectively. Risk mediation through optimal statin management was negligible. CONCLUSIONS: Nonparticipation in cardiac rehabilitation, a cardiometabolic risk profile, and persistent smoking mediate income-dependent prognosis after MI. In the absence of randomized trials, this causal inference approach may guide decisions to improve health equity.