Abstract
BACKGROUND: Depression is recognized as being linked to atherosclerosis and adverse cardiovascular outcomes; however, robust evidence from nationally representative cohorts regarding its association with myocardial infarction (MI) is still insufficient. This study aimed to investigate the relationship between depressive symptoms, measured by the PHQ-9, and the odds of having a history of MI using data from NHANES 2005-2020, while further exploring possible nonlinear patterns and potential modifiers of this association. METHODS: We included 37,139 adults aged 20 years or older (1,574 with MI) and performed weighted analyses accounting for the NHANES complex survey design. Exposure was defined as PHQ-9 scores, analyzed both continuously and by severity categories (none, mild, moderate, severe), with self-reported MI as the outcome. We fitted three progressively adjusted weighted logistic regression models and employed restricted cubic spline (RCS) analysis to evaluate nonlinearity. We further performed prespecified subgroup and interaction analyses, along with a series of sensitivity analyses. RESULTS: Higher depressive burden was significantly associated with higher odds of MI, with MI prevalence rising progressively across depression severity groups (2.89% vs. 4.39% vs. 5.80% vs. 7.08%, p < 0.001). In the unadjusted model, every 1-point increase in PHQ-9 score corresponded to a 6.2% increase in the odds of MI (OR = 1.062, 95% CI: 1.047-1.077; p < 0.001), while severe depression was linked to approximately a 2.56-fold higher odds compared with non-depressed participants. The association remained robust after sequential full adjustment for potential confounders. Restricted cubic spline analysis revealed no evidence of nonlinearity (p > 0.05), supporting an approximately linear association. Subgroup analyses demonstrated consistent associations across most strata; however, significant interactions were detected for CHF, CHD, and higher V/MPA groups (P for interaction < 0.05), indicating attenuated associations in these groups. CONCLUSION: In a nationally representative sample of US adults, depressive burden was strongly associated with prevalent MI, with cardiovascular comorbidities and physical activity acting as potential modifiers of this association. These findings underscore the importance of integrating routine depression screening and holistic mind-body management into both general and high-risk populations, and highlight the necessity of prospective and interventional studies to clarify the temporal sequence and evaluate potential causality.