pECGreview: Using “Not-to-Miss” ECG diagnoses to assess competency of trainees and faculty in ECG interpretation via a weekly online ECG education program

<h2>Abstract</h2><h3>Background</h3> Trainees and practitioners are suboptimal in their pediatric ECG (pECG) interpretation skills. We have previously validated a pECG educational tool, pECGreview, as a weekly, web-based program inviting free-text pECG interpretations and providing direct feedback. <h3>Objective</h3> We expanded the pECG education tool by introducing a novel approach to pECG interpretation: "Not-to-Miss" (NTM). We evaluated the validity of this approach as an assessment of pECG reading accuracy and competency for a cohort of pediatric cardiology fellows and faculty. <h3>Methods</h3> A multiple-choice selection of 18 critical NTM diagnoses was added to our existing pECGreview program. We retrospectively reviewed 2846 responses, before (Group 1) and after (Group 2) the addition of the NTM approach ("NTM group"). Group 1 contained previously assigned accuracy scores based on free text responses ("Accuracy group") and retrospectively assigned NTM diagnoses for comparison ("assignedNTM group"). Each response was classified as correct, underdiagnosis, or overdiagnosis. <h3>Results</h3> Frequency of underdiagnosing was similar in NTM and assignedNTM responses (17% and 18%). Over diagnosing was more common in the NTM group versus assignedNTM and Accuracy groups (31% vs 21% and 10%). Frequency of Correct scores was 62% for assignedNTM and 50% for NTM, versus 84% for Accuracy scores. Atrial fibrillation, high grade AV block, and SVT were most frequently underdiagnosed in the NTM groups but less frequently missed in the Accuracy group. <h3>Conclusions</h3> The NTM approach is a feasible way to emphasize critical diagnoses and automate assessment of competency in pECG interpretation. Cognitive bias leads to over-diagnosing when prompted with multiple-choice.