In an RCT reviewed by this study, an LLM-based digital patient improved ophthalmology medical history-taking skills relative to control¶

Claim · All claims
Evidence · 1 study · review r3 · q2
Currency · newest study 2026 · checked against its sources 2026-10-31 (automated, 1 of 1 entry) · not reviewed by a person

Subclaims¶

q2 i? Medical students using the LLM-based digital patient scored significantly higher on history-taking than controls (mean difference 10.50 points, P<.001). → Jiang Juming 2026

Evidence¶

Jiang Juming 2026¶

Jiang Juming, Ye Megan Zichen, Kwok Tyrone Tai-On, Wong Janet Yuen Ha. (2026). GenAI-Supported Virtual Patients in Health Care Education: Systematic Review. Journal of Medical Internet Research. https://doi.org/10.2196/82756

q2 · i? · review · r3

RCT reported in the review's Table 3 (Luo et al, n=13 per group, 1-hour session): the LLM-based digital patient group outperformed controls on history-taking, with the review noting "The LLMDP group showed significantly higher MHTAn scores". No standardized effect size is printed.

"The LLMDP group showed significantly higher MHTAn scores (mean 64.62, SD 9.52) vs control (54.12, SD 8.80), mean difference 10.50 points (95% CI 4.66‐16.33, P<.001)."

Discussion¶

Related Claims¶

• Controlled GenAI virtual-patient and communication studies reported short-term skill gains, but positive findings were not universal — related
• Multi-agent AI standardized patient training improved final OSCE-aligned examination scores compared with a structured non-LLM control condition — related
• In an RCT reviewed here, adding AI-generated feedback to a GenAI virtual patient improved medical students' clinical decision-making relative to simulation alone — related
• Clearest LLM-teaching signals appear in settings with clearly defined tasks, immediate feedback, and repeated practice, such as simulated patients and structured feedback — a broader claim this one bears on
• History-taking was the most frequently assessed competency in AI-SSP studies (52%), and assessed domains shifted over time toward interactional competencies — related
• LLMs, particularly GPT-4, can match or exceed medical students' examination performance — related
• A social robotic LLM-enhanced virtual patient was rated higher than a conventional computer-based platform for authenticity and learning effect — related
• Near-peer feedback improved clinical note scores compared to no feedback in one included study — related
• Software-based platforms generally improve examination scores, diagnostic accuracy, and decision-making performance — a broader claim this one bears on
• The one pretest-post-test study (Li B et al.) found significant score improvement after ophthalmology teaching (p < 0.01) — related
• Competency level (β = 0.50) and training duration (β = 0.42) are the strongest predictors of ophthalmology education outcomes — related