PHENOMENON: Changes in the medical education milieu have led away from the apprenticeship model resulting in shorter physician-student interactions. Faculty and student feedback suggests that supervisor/student interactions may now be more cursory with increasing numbers of supervisors per student, and shorter duration of interaction. This may affect both education and student assessment.
Approach: We compared inpatient attending and resident daily schedules with those of 3rd- and 4th-year medical students rotating on medicine clerkships at Brigham and Women's Hospital during academic years 2009-11 to determine the number of days of overlap. We used evaluation forms to determine the extent of evaluator's self-reported knowledge of the student.
Findings: We correlated the daily schedules of 199 students and 204 resident and 187 attending physicians, which resulted in 558 resident-student pairings and 680 attending-student pairings over 2 years. During a 4-week block, students averaged 3.7 attending physicians (M = 4, range = 2-7), with 49.7% supervised by 4 or more. Attending-student overlap averaged 9 days (M = 9, range = 2-23), though 40% were 7 days or less. Students overlapped with an average 3.4 residents (M = 3, range = 1-6). Resident-student overlap averaged 12 days (M = 11, range = 3-26). There were 824 student assessment forms analyzed. Resident and attending physician supervisors describing knowledge of their student as "good/average" overlapped with students for 14 and 11 days respectively compared to resident and physician supervisors who described their knowledge as "poor" (11 days, p < .01; 6 days, p < .01). Insights: On the inpatient medicine clerkship, students have multiple supervising physicians with wide variability in the period of overlap. This leads to a disrupted apprenticeship model with fragmentation of supervision and concomitant effects on assessment, feedback, role modeling, and clerkship education.
Keywords: apprenticeship; clerkship; evaluation; interaction; medical school; team structure.