Purpose: To develop diagnostic competence, students should practice with many examples of clinical problems to build rich mental representations of diseases. How to enhance learning from practice remains unknown. This study investigated the effects of reflection on cases compared with generating a single or differential diagnosis.
Method: In 2012, during the learning phase, 110 fourth-year medical students diagnosed four cases of two criterion diseases under three different experimental conditions: structured reflection, single-diagnosis, or differential-diagnosis. One week later, they diagnosed two novel exemplars of each criterion disease and four cases of new diseases that were not among the cases of the learning phase but were plausible alternative diagnoses.
Results: Diagnostic performance did not differ among the groups in the learning phase. One week later, the reflection group obtained higher mean diagnostic accuracy scores (range: 0-1) than the other groups when diagnosing new exemplars of criterion diseases (reflection: 0.67; single-diagnosis: 0.36, P < .001; differential-diagnosis: 0.51, P = .014) and cases of new diseases (reflection: 0.44; single-diagnosis: 0.32, P = .010; differential-diagnosis: 0.33, P = .015). No difference was found between the single-diagnosis and the differential-diagnosis conditions.
Conclusions: Structured reflection while practicing with cases enhanced learning of diagnosis both of the diseases practiced and of their alternative diagnoses, suggesting that reflection not only enriched mental representations of diseases practiced relative to more conventional approaches to clinical learning but also influenced the representations of adjacent but different diseases. Structured reflection seems a useful addition to the existing clinical teaching methods.