Problem-based learning: an effective educational method for a surgery clerkship.

Problem-based learning (PBL) has been implemented during the clinical years in a few medical schools. The purpose of this study is to determine whether PBL provides a better education than traditional methods. Students in the first and third rotations (n = 42) went through the traditional clerkship, which utilized Socratic teaching (SI), while students in the second and fourth rotations (n = 36) were taught by the PBL method. Two performance measures were used to assess clerkship effectiveness. One was a modified essay examination (MEE) administered as part of the departmental evaluation. The other was the NBME-II exam and its surgery subsection NBME-II-S. The MEE was designed to measure six dimensions of the problem-solving process. The NBME-II was utilized to measure knowledge. Unpaired t tests were used to identify statistically significant group differences. The PBL group performed significantly better on two MEE dimensions: (1) differential diagnosis formation (PBL, 92.5 +/- 0.8; SI, 89.1 +/- 0.5; P < 0.01) and (2) interpretation of clinical data (PBL, 93.3 +/- 0.6; SI, 91.6 +/- 0.4; P < 0.03). A third dimension, ordering appropriate lab and diagnostic studies, approached significance (P = 0.057), and the PBL group performed better. On the NBME-II there was not a significant difference between the two groups. However, the trend (P = 0.059) was for the PBL group to score higher on the NBME-II-S (PBL mean: 502 +/- 15; SI mean: 468 +/- 12). When overall achievement was controlled for, the PBL group performed significantly better than the SI group (P = 0.046) on the NBME-II-S.(ABSTRACT TRUNCATED AT 250 WORDS)