Clinician attitudes toward biostatistics

OBJECTIVE: To assess clinician attitudes toward biostatistics at an academic medical center. PARTICIPANTS AND METHODS: We conducted a cross-sectional survey of medical students, internal medicine resident physicians, and internal medicine teaching faculty at Mayo Clinic College of Medicine in Rochester, MN, in April 2005. RESULTS: Of 468 eligible participants, 301 (64.3%) responded to the survey. A total of 87.3% of respondents (262/300) believed it would benefit their career to better understand biostatistics, but only 17.6% (53/301) believed their training in biostatistics was adequate for their needs. A total of 23.3% of respondents (70/300) agreed they could identify when correct statistical methods had been applied in a study, 28.0% (84/300) agreed they could design their own research projects with confidence, and 14.6% (44/301) agreed they could conduct their own statistical analyses with confidence. Respondents with the highest self-reported level of statistical education and research experience were more likely to report these skills (all, P<.001). A total of 92.7% of respondents (279/301) believed biostatistics is an important part of evidence-based medicine (EBM), and 88.0% (265/301) believed EBM is important for clinical practice. However, biostatistics was not evaluated as being as important as many other areas of study within medicine. CONCLUSION: Clinicians across levels of training have low perceived knowledge of biostatistical concepts despite a clear recognition of the importance of these issues. An integrated approach to teaching biostatistics that merges biostatistics with clinically relevant medical discussions, such as those that occur in many EBM curricula for epidemiological principles, may promote learning of biostatistics.