Resident selection and predictors of performance: can we be evidence based?

Selection of orthopaedic residents can be a difficult process; we have endeavored to make it more objective by developing a scoring methodology for screening applications. The purpose of this investigation is to determine if an academic score, using objective elements only, will discriminate among applicants and will correlate with outcomes. Applications to our orthopaedic residency program for 2004 and 2005 were assigned an academic score as a screening tool in the residency selection process. Data was analyzed for the entire group both by gender and whether the applicant had interviewed for the program. Additionally, the applications of program graduates over the past 5 years were retrospectively assigned academic scores, which were compared with outcomes of the training program. Academic scores for applicants formed a generally normal distribution, and residents training in the program generally had higher scores. The distribution of scores for female applicants was similar to male applicants; however, a greater percentage of female applicants interviewed for the program. Scores on the OITE and ABOS examinations tended to parallel academic scores, but faculty ratings of performance in the program showed no difference between those with high and low academic scores. Calculating academic scores makes the application screening process more objective but does not appear to correlate with outcomes of the training program.