Knowledge of levels of evidence criteria in orthopedic residents

The purpose of the levels of evidence system is to provide a framework for critical evaluation of orthopedic literature. This rating system is based on guidelines from the Oxford Centre for Evidence-Based Medicine and is currently in use in several orthopedic surgery journals. The purpose of this study was to investigate resident knowledge of the levels of evidence criteria used in classification of clinical articles. Thirty-eight residents from 5 orthopedic surgery training programs, from year-in-training 3 to 5, determined the levels of evidence rating of 10 blinded articles representing all levels of evidence types in the orthopedic literature. Residents were then provided with a levels of evidence information sheet and asked to re-rate each article. The mean percentage correct for the initial rating was 29.5% and for the posteducation rating was 41.3%, with significant improvement after levels of evidence education (P <.001). The year-in-training-3 group had the highest mean percentage correct for the average of both tests (46.7%) compared to year-in-training-4 (34.2%) and year-in-training-5 (25.4%). Residents were significantly more accurate scoring therapeutic (41.1% correct pre-levels of evidence; 51.6% post-levels of evidence) than prognostic studies (6.6% correct pre-levels of evidence; 28.9% post-levels of evidence) (P <.001). Residents graded the level of evidence correctly in fewer than half the papers. These findings indicate that resident knowledge of levels of evidence criteria is limited and suggest a need for more education in this area.