Test-enhanced learning of clinical reasoning: a crossover randomised trial

CONTEXT: Clinical reasoning is an essential skill, the foundations of which should be acquired during undergraduate medical education. Student performance in clinical reasoning can be assessed using key feature examinations. However, within a paradigm of test-enhanced learning, such examinations may also be used to enhance long-term retention of procedural knowledge relevant to clinical reasoning.

OBJECTIVES: This study tested the hypothesis that repeated testing with key feature questions is more effective than repeated case-based learning in fostering clinical reasoning.

METHODS: In this randomised crossover trial, Year 4 medical students attended 10 weekly computer-based seminars during which patient case histories covering general medical conditions were displayed. The presentation format was switched between groups every week. In the control condition, students studied long case narratives. The intervention condition used the same content but augmented case presentation with a sequence of key feature questions. Using a within-subjects design, student performance on intervention and control items was assessed at 13 weeks (exit examination) and 9 months (retention test) after the first day of term.

RESULTS: A total of 87 of 124 eligible students provided complete data for the longitudinal analysis (response rate: 70.2%). In the retention test, mean ± standard deviation student scores on intervention items were significantly higher than those on control items (56.0 ± 25.8% versus 48.8 ± 24.7%; p < 0.001). The results remained unchanged after accounting for exposure time in a linear regression analysis that also adjusted for sex and general student performance levels.

CONCLUSIONS: This is the first study to demonstrate an effect of test-enhanced learning on clinical reasoning as assessed with key feature questions. In this randomised trial, repeated testing was more effective than repeated case-based learning alone. Curricular implementation of longitudinal key feature testing may considerably enhance student learning outcomes in relevant aspects of clinical medicine.