The Additive Value of Training Evaluations for Predicting Early-Career Doctor Retention: A Real-World Evidence (Preprint)

Le résumé fourni par la source

BACKGROUND The global shortage of physicians poses a significant threat to healthcare systems and health outcomes. While training and evaluations are identified as potential factors influencing physician retention, few studies have explored their predictive role. OBJECTIVE This study aims to investigate how clinical training evaluations can serve as predictors of early-career physicians’ retention, addressing a gap in the existing literature. METHODS This observational study analyzed data from a medical center in northern Taiwan, including 342 Clinical Competency Committee evaluation records of 128 postgraduate year (PGY) physicians. The evaluation records were collected at three six-month intervals for each trainee. Evaluations included multiple competency assessments, such as Direct Observation of Procedural Skills (DOPS), Mini-Clinical Evaluation Exercise, Case-based Discussion, and 360-degree feedback from peers, patients, and clinical teachers. A Random Forest predictive model was developed to identify key factors associated with physician retention, with a training cohort (70%) and a testing cohort (30%) used for internal validation. RESULTS The analysis revealed that DOPS and multi-source feedback, particularly from peers and patients, were the most important predictors of physicians' retention at the institution. Other factors, such as demographic characteristics and academic performance, showed minimal impact on retention outcomes. The Random Forest model demonstrated high accuracy, with an Area Under the Curve of 93.2% (95% CI: 88.2% to 98.1%), correctly predicting retention in 93.75% of cases in which physicians left and 76.32% of cases in which they stayed (p<0.001). CONCLUSIONS This study underscores the potential of clinical training evaluations as early indicators of physician retention. Procedural competence and feedback from peers and patients emerged as key factors influencing retention. These findings imply that training evaluations can serve as predictive tools for physician retention. By identifying at-risk physicians early, healthcare institutions can implement targeted strategies to improve retention and ensure a sustainable workforce. CLINICALTRIAL N/A

Ce résumé expose les affirmations des auteurs. BNTIC ne l’interprète pas comme une validation indépendante des résultats.