Meta-regression identifies country, outcome type, grade and sample size as significant moderators of GenAI's clinical-skills effect, while intervention type is not¶

Claim · All claims
Evidence · 1 study · quant-synthesis r2 · q2
Currency · newest study 2026 · not yet checked against its sources · not reviewed by a person

Subclaims¶

q2 i? In the final clinical-skills meta-regression model (Adjusted R² = 75.59%), significant moderating effects were found for participants' country (P = 0.001), outcome type (P = 0.009), grade (P = 0.026), and sample size (P = 0.022). → Xie Guanli 2026 q2 i? Neither the experimental group intervention type nor the control group intervention type was significantly associated with effect size (P = 0.883 and P = 0.803, respectively). → Xie Guanli 2026

Evidence¶

Xie Guanli 2026¶

Xie Guanli, Liao Jianglong, Tang Xiaoxia, Fu Han, Liu Duo, Deng Li, Wang Tao, Hong Xin. (2026). The impact of generative artificial intelligence on clinical skills and knowledge acquisition in medical undergraduates: a systematic review and meta-analysis. BMC Medical Education. https://doi.org/10.1186/s12909-026-09458-3

q2 · i? · quant-synthesis · r2

Random-effects meta-regression (REML estimation with Knapp-Hartung modification) of clinical-skills effect sizes across seven covariates, after removing two collinear covariates (mean VIF = 132.79 in the initial model). Residual heterogeneity remained moderate (residual I² = 66.90%).

"these covariates jointly explained a substantial proportion of between-study variance (Adjusted R² = 75.59%), with significant moderating effects identified for participants' country (P = 0.001), outcome type (P = 0.009), grade (P = 0.026), and sample size (P = 0.022)"

Discussion¶

Related Claims¶

• Meta-regression identifies percentage of female participants, sample size, and interdisciplinary participation as moderators of effect size — related
• The effect size does not vary significantly with sample size or publication year in meta-regression — related