Education Research: High-Fidelity Simulation to Evaluate Diagnostic Reasoning Reveals Failure to Detect Viral Encephalitis in Medical Trainees

Abstract

Delays in treatment of both herpes simplex virus (HSV) encephalitis and seizures are associated with poor patient outcomes, but many physicians fail to recognize HSV despite classic presenting symptoms. Our goal was to assess trainee performance in a simulation-based case to recognize HSV encephalitis as the underlying etiology of refractory status epilepticus.This is a prospective, observational, single-center simulation-based study of participants ranging from subinterns to attending physicians managing a patient with viral encephalitis complicated by nonconvulsive status epilepticus. Using a modified Delphi approach, we developed a list of critical actions. The primary outcome measure was critical action item sum score. We compared level of training and performance using analysis of variance as validity evidence to support our findings.Fifty-nine trainees completed the simulation. The mean sum of critical actions completed was 13.9/25 (56%). Eighty percent of trainees administered an appropriately dosed benzodiazepine, and 97% administered a second-line agent. Despite 88% of trainees obtaining a lumbar puncture, only 47% recognized viral encephalitis as the most likely diagnosis with 36% starting appropriate treatment. There was significant effect of training level on critical action sum score (level 1 mean score [SD] = 10.8 [1.5] vs level 2 mean score [SD] = 12.2 [2.5] vs level 3 mean score [SD] = 13.9 [3.0] vs level 4 mean score [SD] = 18.2 [3.2],p< 0.001,R2= 0.38).Although initial seizure treatment was sufficient, failure to recognize HSV encephalitis was common with few trainees initiating appropriate treatment potentially leading to poor outcomes in real-life scenarios. High-fidelity simulation holds promise as an assessment tool in identifying trainee knowledge gaps and why classic clinical cases escape trainee diagnosis.