Abstracts

Rationale: EEG interpretation requires significant practice and training, which is often beyond the scope of residency education due to time and resource constraints. Neurologists with insuffiencient EEG training are prone to over-interpretation resulting in unnecessary treatment and excessive costs. We outline the utility of a novel EEG teaching program that is widely distributable, low or no cost, and time saving. This program uses ubiquitous computer resources and can be administered without significant epilepsy knowledge or programming skill.Methods: We developed a program for teaching EEG interpretation, MR EEG. The program quantizes EEG interpretation into discrete topics (for example, background organization, background symmetry, PDR, etc) and organizes teaching EEGs into this framework. Trainees using the program enter their own interpretations, which are graded by comparing with the instructor entries. The program can also test resident knowledge with a bank of EEG related questions. Results are provided in real-time feedback, including peer matched performance. For the past year we had neurology residents from two different neurology programs use this program during their 3 week epilepsy clinic rotation at our institution. Residents were given a pre- and post-test to assess knowledge, and EEGs were scored in comparison to the joint opinion of three fellowship trained epileptologists.Results: 22/34 residents were able to complete a minimum of 20 EEGs. Residents were surveyed on their perceived EEG abilities before and after use of MR EEG. The majority of residents rated their interest in reviewing EEG data for patient care (Table 1) as “fair” or “poor” prior to using MR EEG, subsequently shifting towards “good” (p=0.01, chi square test). Residents rated their EEG interpretation confidence (Table 2) as “fair” or “poor” prior to using MR EEG, but subsequently shifted towards “good” (p=0.002). See Tables 1 and 2 for definitions. In terms of EEG interpretation ability, we assessed trainee ability to interpret normal EEG and ability to identify epileptiform discharges. EEGs 1 and 21 in our set were similar and normal, and were completed by 13 residents in our sample. These residents scored 66% on the first normal EEG (as compared to the author’s EEG interpretation) but improved to 82% by EEG 21, demonstrating statistically significant improvement (p=0.02, two-tailed paired Student’s t test). Twelve residents completed at least 24 modules, and modules 3 and 24 demonstrated similar generalized spike and wave discharges. These residents scored an average of 56% on EEG 3 but 70% on EEG 24, again demonstrating significant improvement (p=0.015).Conclusions: We demonstrate the efficacy of the MR EEG program in EEG education. Our residents demonstrated a statistically significant improvement in their objective interpretation skills of normal and epileptiform EEGs, and developed improved confidence. This training required minimal attending oversight, and could be completed between other clinical duties. We believe this would be a valuable adjunct to conventional EEG teaching methods.