Effectiveness of Generative AI-Enabled Teaching for Clinical Reasoning in Health Care Education: Protocol for a Systematic Review and Meta-Analysis.

BACKGROUND: Clinical reasoning skills are key competencies required by health care professionals. Generative AI (GenAI)-enabled teaching may be able to offer personalized learning; however, its effectiveness compared with traditional teaching methods for developing clinical reasoning capabilities remains underexplored. OBJECTIVE: This systematic review and meta-analysis will evaluate, synthesize, and compare the evidence on effectiveness in clinical reasoning capabilities between GenAI-enabled teaching and traditional teaching methods in health care education. METHODS: This review protocol will adhere to the PRISMA-P (Preferred Reporting Items for Systematic Reviews and Meta-Analysis Protocols) guidelines. The population will include health care professionals. The intervention will be the use of GenAI-enhanced teaching methods, with traditional teaching methods as the comparator. The primary outcome will be objective clinical reasoning performance, with relevant validity for the construct, population, and intended purpose. A database search will be conducted across Embase, MEDLINE, CINAHL, Scopus, and Cochrane CENTRAL, supplemented by Google Scholar, reference list, and citation searches. Study designs will include randomized controlled trials (RCTs), non-RCTs, quasiexperimental studies, and cohort studies with suitable controls. Title and abstract screening and full-text screening will be completed by 2 independent reviewers, with arbitration as required. IBM SPSS will be used to combine and calculate the overall effect. A random-effects model will be deployed, with the Hartung-Knapp-Sidik-Jonkman method used, as appropriate, to calculate the CI around the estimated pooled effect. Study heterogeneity will be assessed using I2 and τ2, and where studies are judged to be too diverse, results will be presented narratively. Prediction intervals will be used, if appropriate, to show how the true effect may vary between similar studies. Risk of bias will be reviewed by 2 independent reviewers, with arbitration, as required, using the Risk of Bias 2 (ROB 2) tool for RCTs and the Risk of Bias in Non-Randomized Studies of Interventions (ROBINS-I) tool for non-RCTs, quasiexperimental studies, and interventional cohort studies. Sensitivity analysis will be used to explore possible sources of heterogeneity. A subgroup analysis may include the type and functionality of GenAI used and/or health care professional background. RESULTS: The review was funded in September 2025. A trial search was completed on MEDLINE in December 2025, with final database searches completed in February 2026. A total of 4831 studies were retrieved, with 2083 duplicates removed. A further 2558 studies were removed at title and abstract screening, with 190 studies undergoing full-text screening and 42 studies included in the final review. Expected completion date of the final review is autumn or winter 2026. CONCLUSIONS: Health care education is rapidly adopting GenAI-enhanced teaching methods, despite inconclusive evidence on their effectiveness. The findings from this review will offer essential insight into the optimization of GenAI-enhanced teaching methods for clinical reasoning within current and future curriculum design. TRIAL REGISTRATION: PROSPERO CRD420261284944; https://www.crd.york.ac.uk/PROSPERO/view/CRD420261284944. INTERNATIONAL REGISTERED REPORT IDENTIFIER (IRRID): DERR1-10.2196/95709.