PURPOSE. To compare clinical performances in a third-year medicine clerkship between studies from a problem-based learning (PBL) curriculum and students from a traditional, lecture-based learning (LBL) curriculum. METHOD. The study participants were 88 PBL students and 364 LBL students rotating through a common third-year internal medicine clerkship at the Bowman Gray School of Medicine at Wake Forest University, classes of 1991-1995. Faculty and housestaff assessed the students' performances using four clinical rating scales. The student also completed the medicine student (“shelf”) test of the National Board of Medical Examiners (NBME). RESULTS. On average, the PBL students received significantly higher ratings from housestaff and faculty on all four rating scales. No difference in scores on the NBME medicine shelf test was observed. CONCLUSION. The results support the hypothesis that preclinical PBL curricula as found at the Bowman Gray School of Medicine may enhance third-year students' clinical performances.