Abstracts

Rationale: Neurology residents are often involved in the care of patients undergoing long-term EEG monitoring, but protected time for residents to practice reading EEGs and writing EEG reports is often limited. The Mount Sinai Hospital’s adult neurology residency recently underwent a reform of its EEG curriculum to increase protected time for EEG reading while on a dedicated epilepsy monitoring unit (EMU) service block. In this study, we aimed to measure the impact of this protected time on residents’ performance on an EEG quiz.

Methods: We distributed a survey to adult neurology residents in postgraduate years (PGY) 2-4, of whom some had been exposed to the new EMU service block (EMU residents) and some had not (non-EMU residents). Residents were presented with 20 EEG samples (10-15 second epochs presented in longitudinal bipolar montage) and were asked whether the EEG showed normal activity (including benign variants/artifact). If they scored the EEG as normal, this became their final diagnosis. If they scored the EEG as abnormal, they were asked whether the EEG was ictal or on the ictal-interictal continuum (IIC). Responses were scored against an attending epileptologist’s diagnosis (normal activity or benign variants, n=4; slowing, n=5; sporadic epileptiform discharges, n=1; IIC, n=4; ictal, n=6). For each sample, residents were asked whether they would want to call the EEG fellow/attending (which could suggest either uncertainty or urgency regarding the EEG findings). We calculated descriptive statistics and compared responses between EMU and non-EMU residents, as well as between PGY levels (no significance testing due to low sample size).

Results: Thirteen residents completed the survey (10 EMU [6 PGY2, 4 PGY4]; 3 non-EMU [2 PGY3, 1 PGY4]). Residents correctly recognized normal EEGs as normal with 55% accuracy, compared to 96% accuracy in identifying ictal/IIC EEGs as abnormal. If EEGs were correctly determined to be abnormal, residents identified ictal/IIC activity with 77% accuracy; even if normal EEGs were incorrectly determined to be abnormal, these were not misinterpreted as ictal/IIC patterns 95% of the time. Increased PGY level was broadly associated with increased accuracy in identifying normal activity (PGY2: 74%, PGY3: 88%, PGY4: 89%) and ictal/IIC activity (PGY2: 75%, PGY3: 71%, PGY4: 85%). Compared to non-EMU residents, EMU residents had lower accuracy for identifying normal activity (81% vs. 89%) and higher accuracy for identifying ictal/IIC activity (82% vs. 74%). While similar proportions of EMU and non-EMU residents would opt to call the EEG fellow/attending (52% vs. 56%), EMU residents were less likely to call for slow EEGs (31% vs. 58%) and more likely to call for ictal EEGs (90% vs. 77%).

Conclusions: Residents with more protected EEG reading time, despite being mostly junior residents, displayed similar accuracy for identifying normal and ictal/IIC patterns to senior residents who did not have this protected time. Future prospective comparisons of resident EEG knowledge before and after EMU service time are underway.

Funding: None