How and when should clinical reasoning be taught in undergraduate medicine: a systematic review and meta-analyses

Background: Clinical reasoning is a critical aspect of clinical practice, though there is considerable variation regarding how and when to teach this skill. This systematic review and meta-analyses examined the effectiveness of interventions that explicitly taught clinical reasoning in undergraduate medical education and the optimal timing for introducing interventions. Methods: A systematic (PRISMA 2020) search of the SCOPUS, MEDLINE, CINAHL, PsycINFO, ERIC, and Informit databases was conducted from 1 January 2014 to 31 December 2024. The quality of studies was assessed using the Quality Assessment with Diverse Studies tool. Pooled estimates and 95% confidence intervals (CIs) were estimated using both random and fixed effects meta-analyses. Results: The final sample included 50 studies, of which 46 (92%) reported a measurable improvement. Small-group teaching generally achieved better outcomes, with technology and serious game innovations further improving them. Meta-analysis of six randomised control trials using a random effects model showed an overall significant result (MD 2.23; 95% CI: 0.67, 3.80; I² = 88%). A subgroup analysis indicated that interventions undertaken in pre-clinical years (MD 0.32; 95% CI: –3.99, 4.64; I² = 88%) did not result in significant improvements, whereas interventions in the clinical years were significant (MD 1.89; 95% CI: 1.06, 2.72; I² = 88%). A second subgroup analysis showed that interventions based on face-to-face workshops (MD 1.74; CI: 1.19, 2.28; I² = 88%) were significant. Conclusions: The findings suggest that small-group activities, such as interactive online modules, may lay a foundation for early-year students, while skills-based workshops and serious games progressively refine and enhance clinical reasoning. Future research should focus on longitudinal outcomes and standardised assessment measures across diverse educational contexts.