Evidence for AI-powered PBL/CBL is limited: small study count, some concerns risk of bias, and no long-term effectiveness data¶

Claim · All claims
Evidence · 1 study · quant-synthesis r2 · q2
Currency · newest study 2025 · not yet checked against its sources · not reviewed by a person

Subclaims¶

q2 i? The review reports that only six RCTs were included, all judged at overall risk of bias of 'some concerns', and that 4 of 6 studies evaluated outcomes immediately after a single session, leaving long-term effectiveness untested. → Wei H 2025

Evidence¶

Wei H 2025¶

Wei H, Dai Y, Yuan K, Li KY, Hung KF, Hu EM, Lee AHC, Chang JWW, Zhang C, Li X. (2025). AI-Powered Problem- and Case-based Learning in Medical and Dental Education: A Systematic Review and Meta-analysis. International Dental Journal. https://doi.org/10.1016/j.identj.2025.100858

q2 · i? · quant-synthesis · r2

The review's own quality appraisal of the six included RCTs using RoB 2; all studies were judged 'some concerns' overall, and the review states there is "insufficient evidence supporting the long-term effectiveness" of AI-powered learning. No effect size applies to this methodological finding.

"It should also be noted that the overall risk of bias for all included studies was judged to have ‘some concerns’. Furthermore, there is insufficient evidence supporting the long-term effectiveness of AI-powered learning."

Discussion¶

Related Claims¶

• Evidence quality is limited: high risk of bias in allocation concealment and blinding, small samples, subjective outcome measures, and limited generalizability — a broader claim this one bears on
• AI-powered PBL/CBL significantly improves students' knowledge acquisition compared with PBL/CBL without AI or traditional lectures — related
• Students report generally high satisfaction with AI-powered PBL/CBL — related
• The RCTs showed some concern or high overall risk of bias and the non-randomized studies moderate to serious risk, making the evidence base uncertain — related