An effective automated method for teaching EEG interpretation to neurology residents

EEG interpretation is a fundamental procedural skill in the practice of neurology, but there is no standardized method for educating residents. One-to-one instruction is commonly employed, but is time intensive for supervising physicians, provides arbitrary exposure to normal and abnormal EEG patterns, and often lacks immediate and detailed feedback on performance. Here, we investigated the effectiveness of a novel automated program to assist in educating neurology residents in EEG interpretation.

An EEG teaching program was developed to provide neurology residents EEG training less dependent on attending supervision. Residents enter interpretations of full-length pre-selected EEGs and receive immediate feedback based on consensus interpretation of supervising epileptologists. Resident learning was assessed based on performance on matched pre- and post-tests covering common EEG findings including artifacts, normal variants, and abnormalities.

Twenty residents were included in this analysis: 12 post-graduate year (PGY) 3 and eight PGY 4 neurology residents. All residents showed improvement, from a mean score of 42.7% (95% CI 36.9-48.5%) on the pre-test to 75.4% (95% CI 70.7-80.2%) on the post-test (p<0.001). No significant difference was noted between the classes. Residents reported taking 16-30h to complete this teaching module spread over a 3-week rotation.

This pilot study demonstrated the effectiveness of an automated EEG teaching program used by neurology residents in training. This tool could serve as an effective method of supplementing resident education.