Behavioral risk factors and socioeconomic inequalities in ischemic heart disease mortality in the United States: A causal mediation analysis using record linkage data

Abstract

Background Ischemic heart disease (IHD) is a major cause of death in the United States (US), with marked mortality inequalities. Previous studies have reported inconsistent findings regarding the contributions of behavioral risk factors (BRFs) to socioeconomic inequalities in IHD mortality. To our knowledge, no nationwide study has been conducted on this topic in the US. Methods and findings In this cohort study, we obtained data from the 1997 to 2018 National Health Interview Survey with mortality follow-up until December 31, 2019 from the National Death Index. A total of 524,035 people aged 25 years and older were followed up for 10.3 years on average (SD: 6.1 years), during which 13,256 IHD deaths occurred. Counterfactual-based causal mediation analyses with Cox proportional hazards models were performed to quantify the contributions of 4 BRFs (smoking, alcohol use, physical inactivity, and BMI) to socioeconomic inequalities in IHD mortality. Education was used as the primary indicator for socioeconomic status (SES). Analyses were performed stratified by sex and adjusted for marital status, race and ethnicity, and survey year. In both males and females, clear socioeconomic gradients in IHD mortality were observed, with low- and middle-education people bearing statistically significantly higher risks compared to high-education people. We found statistically significant natural direct effects of SES (HR = 1.16, 95% CI: 1.06, 1.27 in males; HR = 1.28, 95% CI: 1.10, 1.49 in females) on IHD mortality and natural indirect effects through the causal pathways of smoking (HR = 1.18, 95% CI: 1.15, 1.20 in males; HR = 1.11, 95% CI: 1.08, 1.13 in females), physical inactivity (HR = 1.16, 95% CI: 1.14, 1.19 in males; HR = 1.18, 95% CI: 1.15, 1.20 in females), alcohol use (HR = 1.07, 95% CI: 1.06, 1.09 in males; HR = 1.09, 95% CI: 1.08, 1.11 in females), and BMI (HR = 1.03, 95% CI: 1.02, 1.04 in males; HR = 1.03, 95% CI: 1.02, 1.04 in females). Smoking, physical inactivity, alcohol use, and BMI mediated 29% (95% CI, 24%, 35%), 27% (95% CI, 22%, 33%), 12% (95% CI, 10%, 16%), and 5% (95% CI, 4%, 7%) of the inequalities in IHD mortality between low- and high-education males, respectively; the corresponding proportions mediated were 16% (95% CI, 11%, 23%), 26% (95% CI, 20%, 34%), 14% (95% CI, 11%, 19%), and 5% (95% CI, 3%, 7%) in females. Proportions mediated were slightly lower with family income used as the secondary indicator for SES. The main limitation of the methodology is that we could not rule out residual exposure-mediator, exposure-outcome, and mediator-outcome confounding. Conclusions In this study, BRFs explained more than half of the educational differences in IHD mortality, with some variations by sex. Public health interventions to reduce intermediate risk factors are crucial to reduce the socioeconomic disparities and burden of IHD mortality in the general US population.