Key result
A Diversity-Sensitive Model-augmented human practice achieved a sensitivity of 91.3% for detecting STEMI compared to 72.8% for human practice alone, improving predictions in all demographic subgroups.
Why the study?
The study sought to improve subgroup performance variability in predictive models used to identify arriving emergency department patients at high risk for ACS who require an ECG within 10 minutes to detect STEMI.
Does a Diversity-Sensitive Model or Model-Augmented Human Practice improve the sensitivity of identifying ED patients at high risk for ACS compared to human practice alone?
Cohort (n=490,805)
No
Does a Diversity-Sensitive Model or Model-Augmented Human Practice improve the sensitivity of identifying ED patients at high risk for ACS compared to human practice alone?
Augmenting human triage with a diversity-sensitive predictive model significantly improves the sensitivity of identifying ED patients needing an immediate ECG for suspected ACS, though subgroup performance variations remain.
We sought to improve the subgroup performance variability of a model that identifies arriving ED patients at high risk for ACS, to receive an ECG within 10 minutes of arrival, to detect STEMI. We compared a Base Model using age, sex, and chief complaint alone to (1) an Interactions Model adding interactions among the 3 variables, and (2) a Diversity-Sensitive Model including race, ethnicity, language, as well as identity interactions. We quantified human performance and combined it with each of the 3 models simulating use as practice augmenting AI predictions. With sensitivity as our primary outcome, we found humans at 72.8% were bested by the Diversity-Sensitive Model at 82.8%, and by the human-augmented Diversity-Sensitive Model at 91.3%, improving ACS predictions in all demographic subgroups. However, there was residual variation among subgroups (range of sensitivity: 62%−98%). Given risk distribution differences, subgroup-specific ECG-testing thresholds may further equitize ACS prediction performance.
No takes yet. Share an insight, caveat, or question.
Bunney et al. (2025) conducted a cohort in Acute Coronary Syndrome (ACS) / STEMI (n=490,805). Diversity-Sensitive Model-augmented human practice vs. Human practice alone was evaluated on Sensitivity for identifying STEMI (95% CI 90.4-92.1). A Diversity-Sensitive Model-augmented human practice achieved a sensitivity of 91.3% for detecting STEMI compared to 72.8% for human practice alone, improving predictions in all demographic subgroups.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: