Oral and maxillofacial surgery residents have poor understanding of biostatistics

The purpose of this study was to evaluate residents' understanding of biostatistics and interpretation of research results.

A questionnaire previously used in internal medicine residents was modified to include oral and maxillofacial surgery (OMS) examples. The survey included sections to identify demographic and educational characteristics of residents, attitudes and confidence, and the primary outcome-knowledge of biostatistics. In 2009 an invitation to the Internet survey was sent to all 106 program directors in the United States, who were requested to forward it to their residents.

One hundred twelve residents responded. The percentage of residents who had taken a course in epidemiology was 53%; biostatistics, 49%; and evidence-based dentistry, 65%. Conversely, 10% of OMS residents had taken none of these classes. Across the 6-item test of knowledge of statistical methods, the mean percentage of correct answers was 38% (SD, 22%). Nearly half of the residents (42%) could not correctly identify continuous, ordinal, or nominal variables. Only 21% correctly identified a case-control study, but 79% correctly identified that the purpose of blinding was to reduce bias. Only 46% correctly interpreted a clinically unimportant and statistically nonsignificant result. None of the demographic or experience factors of OMS residents were related to statistical knowledge. Overall, OMS resident knowledge was below that of internal medicine residents (P<.0001). However, OMS residents were overconfident in their claim to understand most statistical terms.

OMS residents lack knowledge in biostatistics and the interpretation of research and are thus unprepared to interpret the results of published clinical research. Residency programs should include effective biostatistical training in their curricula to prepare residents in evidence-based dentistry.