Impact of undergraduate medical training on housestaff problem-solving performance: implications for problem-based curricula.

This article reports a study comparing the problem-solving performance of housestaff with undergraduate medical training in either conventional or problem-based schools. After reading two clinical cases, residents were required to write differential diagnoses and pathophysiological explanations. Biomedical and clinical knowledge used and reasoning strategies were identified. The results suggest that housestaff performance is influenced by the nature of instruction. Housestaff trained in a conventional curriculum (CC) focused on patient information, separated biomedical from clinical knowledge, and used data-driven strategies. Housestaff from problem-based learning curricula (PBLC) organized their knowledge around generated inferences, integrated biomedical and clinical knowledge, and used hypothesis-driven strategies. Data-driven reasoning appears to be impeded in PBLC, suggesting that PBLC students have difficulties in acquiring problem schemata. Previous investigations also found this pattern to be true for medical students trained in two different curriculum formats. Although all housestaff generated equal numbers of diagnostic hypotheses during the reasoning process, housestaff from the conventional curriculum generated a greater number of accurate hypotheses than residents in PBLC. These results are discussed in relation to assumptions in health professional curricula about the adequacy of hypothetico-deductive methods of reasoning as teaching mechanisms and the need for clinical and biomedical knowledge integration.