Unique Performance Problems in Surgical Residency as Novel Predictors of Negative Postresidency Outcomes

Surgical educators have an interest in ensuring graduates of residency programs are able to pass board examinations and build a surgical practice. We sought to learn if there are indicators in residency that can be used to predict propensity for difficulty in passing boards or in later practice.

We reviewed the residency and postresidency (board results) files of 104 residents in our general surgery program, cataloging performance problems using a checklist we had developed for a previous study. We then used online resources including state licensing boards to assess nonboard postresidency outcomes. We used binary logistic regression to assess which predictors pointed to negative outcomes, and receiver operating characteristic (ROC) analysis to determine if there were cutoff scores for American Board of Surgery In-Training Examination (ABSITE) exams, seriousness of problems, and number of unique problems identified in the file.

This study took place at a general surgery residency program in the Midwest.

Residents who had 5 or more unique problems in our checklist were at statistically significant risk of first-time board failure and negative postresidency outcomes compared to residents with zero documented problems. They were also at statistically significantly higher risk than residents who had 1 to 4 unique performance problems documented in their files. Residents who had 1 to 4 documented performance problems were at statistically significantly higher risk than residents who had zero documented performance problems only for negative online mentions.

Capturing unique performance problems during residency is a novel method to assess risk for negative postresidency outcomes, both in passing board exams and mounting a successful practice.