A pilot randomized, controlled trial of a longitudinal residents-as-teachers curriculum

To determine whether a longitudinal residents-as-teachers curriculum improves generalist residents' teaching skills.

From May 2001 to February 2002, 23 second-year generalist residents in four residencies affiliated with the University of California, Irvine, College of Medicine, completed a randomized, controlled trial of a longitudinal residents-as-teachers program. Thirteen intervention residents underwent a 13-hour curriculum during one-hour noon conferences twice monthly for six months, practicing teaching skills and receiving checklist-guided feedback. In a 3.5-hour, eight-station objective structured teaching examination (OSTE) enacted and rated by 15 senior medical students before and after the curriculum, two trained, blinded raters independently assessed each station with detailed, case-specific rating scales (rating scale reliability = 0.96, inter-rater reliability = 0.78).

The intervention and control groups were similar in academic performance, specialty distribution, and gender (chi(2) = 0.434, p =.81). On a five-point Likert scale (5 = best teaching skills), intervention and control residents showed similar mean pretest OSTE scores (2.83 vs. 2.88, p =.736). The intervention group improved their mean overall OSTE scores 22.3% (more than two standard deviations) from 2.83 (pretest) to 3.46 (post-test; p <.0005; 95% CI 0.53 to 0.72). Intervention residents also improved significantly on six of eight OSTE stations. Within the control group, no pretest-to-post-test change achieved statistical significance. Mann-Whitney and Wilcoxon signed-rank tests confirmed these results.

Generalist residents randomly assigned to receive a 13-hour longitudinal residents-as-teachers curriculum consistently showed improved OSTE scores. Future research should clarify which aspects of residents-as-teachers curricula most effectively improve educational outcomes.