Elina E Pliakos, Kaitlyn Shultz, Kriyana P Reddy, Lauren A Eberly, Sameed A M Khatana, Alexander C Fanaroff, Peter W Groeneveld, Jay Giri, Ashwin S Nathan
Incorporating SDOH variables into an already high-performing clinical risk model for in-hospital STEMI death did not meaningfully enhance prediction. These findings indicate that short-term, in-hospital survival is primarily driven by immediate clinical factors rather than upstream socioenvironmental risks.
BACKGROUND: Social determinants of health (SDOH) are increasingly recognized as contributors to disparities in cardiovascular outcomes. However, their incremental value in risk prediction models for ST-segment-elevation myocardial infarction (STEMI) death remains uncertain.
METHODS: We analyzed hospital admissions for STEMI between 2018 and 2020 using the American Heart Association's GWTG-CAD (Get With The Guidelines-Coronary Artery Disease) registry. Logistic regression models were constructed: a clinical model including demographic, clinical, and medical history variables, and a full SDOH model that additionally incorporated zip code-level SDOH (college graduation, high school graduation, per-capita income, poverty, foreign-born status, and median household income). Discrimination was assessed using the concordance statistic/area under the receiver operating characteristic curve, and incremental value was evaluated with DeLong's test. Missing data were addressed with multiple imputation by chained equations.
RESULTS: The imputed cohort included 749 410 STEMI admissions (median age, 62 years, 72% men). Both the clinical and SDOH models demonstrated excellent discrimination (concordance statistic, 0.95; area under the receiver operating characteristic curve, 0.85), with no meaningful difference in overall predictive performance. In exploratory analyses, the addition of high school graduation, college graduation, per-capita income, poverty, and foreign-born status each had a modest statistical improvement in model discrimination, whereas median household income did not.
CONCLUSIONS: Incorporating SDOH variables into an already high-performing clinical risk model for in-hospital STEMI death did not meaningfully enhance prediction. These findings indicate that short-term, in-hospital survival is primarily driven by immediate clinical factors rather than upstream socioenvironmental risks.