Machine learning to enhance electronic detection of diagnostic errors.
Machine learning (ML) has the potential to improve diagnosis and reduce adverse events in healthcare. This study evaluated an ML-enhanced electronic trigger to identify possible missed opportunities in diagnosis (MODs) among a large cohort of VA patients discharged from the emergency department (ED) after presenting with dizziness or abdominal pain who were subsequently hospitalized soon thereafter. Both ML-enhanced trigger tools had high positive predictive value, indicating that the tool could reduce the burden of manual medical record review to identify MODs