Correlation between trainee candidate selection criteria and subsequent performance.

[BACKGROUND] The objective of trainee recruitment is to identify candidates likely to perform well as trainees and subsequent faculty. The effectiveness of this process has not been established. The goal of this study was to identify trainee selection criteria predictive of excellent performance.

[STUDY DESIGN] Twenty-nine microsurgery fellows were enrolled from 2008 to 2012. Each candidate was interviewed and rated based on presentation, plastic surgery (PS) training experience, academic potential, personality, social skills, communication skills, and ability to be a team player. An unadjusted rank list was generated based on weighted averages, and an adjusted rank list was then generated at a faculty meeting. At the conclusion of fellowship, each fellow was rated based on the ACGME core competencies. Spearman correlation coefficients (r) were used to measure the correlations between fellow selection criteria and fellow performance.

[RESULTS] Plastic surgery training and academic potential had, by far, the strongest correlation to overall performance (r: 0.678, p < 0.001 and r: 0.56, p < 0.002), and to all ACGME competencies. When reformulated to weight PS training and academic potential more heavily than subjective criteria, the scoring system was significantly more predictive of excellent performance (r: 0.49 vs 0.70). The unadjusted rank list was more predictive of excellent performance than the adjusted rank list (r: 0.45 vs 0.65).

[CONCLUSIONS] Plastic surgery training experience and academic potential were better predictors of performance than any subjective information ascertained during the interview. Adjustments to the rank list based on faculty discussion resulted in lower performance candidates moving up in ranking. Ranking criteria and interview techniques must be refined to improve predictive power. It may be beneficial for semi-objective criteria to carry more weight than subjective criteria and raw scores to remain unadjusted by extraneous information.