Resident physician well-being and assessments of their knowledge and clinical performance

BACKGROUND: Medical knowledge and clinical performance ratings are major criteria for assessing the competence of resident physicians. However, these assessments may be influenced by residents' mental health. The relationship between residents' well-being and empathy and assessments of their global performance remains unclear. OBJECTIVE: To determine whether resident well-being and empathy are associated with assessments of their medical knowledge and clinical performance. DESIGN, SETTING, AND PARTICIPANTS: We studied 730 clinical performance assessments completed by peers, supervising residents, and allied health professionals; 193 mini-clinical evaluation exercise (mini-CEX) evaluations; and 260 in-training examinations (ITE) of Mayo Clinic internalmedicine residents between January 2009 and August 2010. Multivariate generalized estimating equations were used to evaluate associations between residents' well-being and empathy and assessments of their knowledge and clinical performance. MEASUREMENTS: Independent variables were empathy using the Interpersonal Reactivity Index (IRI), burnout using the Maslach Burnout Inventory, depression using a standardized two-question screening instrument, and quality of life using a Linear Analog Self-Assessment item and the Medical Outcomes Study Short Form (SF-8). Dependent variables were mini-CEX, ITE, and the validated six-item Mayo clinical performance assessment. RESULTS: 202 residents (64.7%) provided both wellbeing and at least one category of assessment data. In multivariate models, residents' scores on the IRI empathy measure of "the tendency to adopt the psychological view of others" were associated with higher peer ratings on "desirability as a physician for a family member" (beta=0.023, 95% CI=0.007-0.039, p=0.004). Additionally, burnout was associated with higher supervisor ratings of communication (beta=0.309, 95% CI=0.100-0.517, p=0.004). There were no observed associations between ITE or mini-CEX scores and resident quality of life, burnout, fatigue, depression, or empathy. CONCLUSIONS: Most dimensions of resident well-being were not associated with residents' knowledge scores and assessments of their clinical performance by other members of the health care team, which supports the trustworthiness of these measures. Nonetheless, correlations of resident empathy and burnout with assessments completed by peers and supervising residents suggest that some ratings of residents may be influenced by interpersonal factors.