Actionable companion · Retention & transfer

Learning with agency: Effectiveness of Artificial Intelligence–Assisted Peer Teaching in Orthopedic Clinical Education: Historical Cohort Study

AI-assisted editorial implementation guide, separate from the original research. Proposed actions are not claims or recommendations attributed to the source authors. No independent peer review of this guide is asserted.

Read original paper Search paper agent CC BY 4.0

Evidence and context

Title and abstract screening; stage labels describe reported settings, not verified numerical ages. Design labels require full-text appraisal. Reported stage(s): Higher education. Health-professions learning must be distinguished from K–12 instruction.

Original authors’ abstract, reproduced under the linked source license:

Abstract Background Peer teaching is an established pedagogical approach in medical education; yet, traditional methods face challenges including inconsistent knowledge support, variable teaching quality, and limited scalability. Artificial intelligence (AI) large language models offer potential to augment peer teaching by providing on-demand access to medical knowledge and clinical reasoning support. However, AI integration within structured peer teaching has not been systematically evaluated in clinical education. Objective This study aims to evaluate the effectiveness of AI-assisted peer teaching compared to traditional peer teaching in orthopedic clinical education, with respect to knowledge acquisition, clinical skills development (particularly clinical reasoning), student engagement, and 3-month knowledge retention. Methods This historical cohort study compared 2 consecutive cohorts of medical students (aged 20-27 years, 108/190, 56.8% male) at the Second Affiliated Hospital, Zhejiang University School of Medicine, Hangzhou, China. All eligible students from each cohort were enrolled. The control group (2021 cohort, n=96, taught in 2024) received traditional peer teaching; the intervention group (2022 cohort, n=94, taught in 2025) received AI-assisted peer teaching with access to DeepSeek-V3. Primary outcomes were assessed using a validated 50-item multiple-choice examination (0-100 points) and a 4-station Objective Structured Clinical Examination (OSCE; 0-100 points) with standardized rubrics (intraclass correlation coefficient>0.85). Secondary outcomes included student engagement and satisfaction (5-point Likert scales) and AI usage metrics. Assessments were conducted at baseline, postintervention (8 weeks), and 3-month follow-up. Analysis of covariance adjusted for baseline knowledge, prior AI experience, and learning interest to address observed baseline imbalances. Results Using independent samples t tests (α=.05, 2-tailed), the AI-assisted group demonstrated significantly higher postintervention knowledge scores (mean 79.69, SD 8.41 vs mean 75.33, SD 9.26; mean difference=4.36, 95% CI 1.84-6.87; P<.001; Cohen d=0.49). OSCE total scores were significantly higher (mean 80.95, SD 7.57 vs mean 76.24, SD 9.23; mean difference=4.71, 95% CI 2.31-7.11; P<.001; d=0.56), with clinical reasoning showing the largest effect (mean difference=2.22, 95% CI 1.18-3.25; P<.001; d=0.61). Analysis of covariance adjusted results remained significant for all primary outcomes (adjusted knowledge difference=3.52, P=.002; adjusted OSCE difference=4.52, P<.001). At 3-month follow-up (174/190, 91.6%), the AI-assisted group maintained higher knowledge scores (mean 77.36, SD 8.60 vs mean 72.84, SD 10.42; mean difference=4.52, 95% CI 1.68-7.36; P=.002; d=0.47), with similar knowledge decay rates between groups. Conclusions This study provides the first systematic evidence that integrating AI tools within structured peer teaching enhances orthopedic clinical education across multiple domains, including knowledge acquisition, OSCE performance, and student engagement. Unlike prior studies examining AI as a stand-alone learning tool, this work demonstrates the synergistic potential of combining AI knowledge support with peer teaching’s social learning benefits, with particularly strong effects on clinical reasoning. These findings support scalable, cost-effective implementation of AI-augmented peer teaching, though randomized controlled trials are needed to confirm causality and determine optimal implementation strategies. Keywords: artificial intelligence, peer teaching, medical education, orthopedic surgery, clinical reasoning, OSCE, Objective Structured Clinical Examination, educational technology, DeepSeek

What kind of evidence is this?

Supporting AI learning science · Randomized study reported

Stage: Higher education. Screening labels describe reported context and study-design language. They are not a risk-of-bias rating or proof of effectiveness for an age band.

Participant ages · source passage

Methods This historical cohort study compared 2 consecutive cohorts of medical students (aged 20-27 years, 108/190, 56.8% male) at the Second Affiliated Hospital, Zhejiang University School of Medicine, Hangzhou, China.

Study methods · source passage

This historical cohort study compared 2 consecutive cohorts of medical students receiving orthopedic clinical education at Zhejiang University School of Medicine Affiliated Second Hospital. The study followed the STROBE (Strengthening the Reporting of Observational Studies in Epidemiology) guidelines for reporting observational studies [23] and the JARS-Quant (Journal Article Reporting Standards for Quantitative Research) checklist [24].

Limitations and uncertainty · source passage

This study has several limitations. First, the historical cohort design precludes definitive causal inference. Despite identical assessment instruments, curriculum content, and faculty assignments, temporal confounders, including evolving institutional emphasis on AI literacy, curriculum drift, and societal AI awareness, cannot be fully excluded. Causal conclusions cannot be definitively drawn; these findings should be interpreted as suggestive evidence requiring confirmation through randomized controlled trials.

Design the system before the pilot

Use this study to inform a learning mechanism or implementation question. It is not direct evidence that autonomous agents improve learning. Keep the proposed agent behavior separate from what the source actually tested.

Include a delayed assessment and an unfamiliar transfer problem without AI. Immediate supported performance cannot establish durable learning.
Measure engagement separately from achievement. Check whether novelty-driven participation persists and whether it translates to independent performance.
Use common rubrics, blind scoring where feasible and a fresh unaided reasoning task. Document permissible assistance and avoid treating AI-detector scores as proof of misconduct.

A 90-day district pilot

Days 1–15 · Appraise. Have the curriculum lead and a teacher read the methods and limitations. Record the study design, sample, comparator, outcomes and evidence gaps. Define a local question that the study can help inform.
Days 16–30 · Design. Choose one disciplinary task. Require a first attempt, a documented AI consultation, a critique of the output and an independent explanation. Align the comparison group and scoring rubric with the same objective. Name an accountable owner and set the baseline, success threshold and budget before implementation.
Days 31–60 · Test. Start with a small voluntary cohort, preserve an accessible non-AI route and require teacher review. Use synthetic or de-identified examples; approve data handling and tool terms through district processes.
Days 61–90 · Decide. Blind-score reasoning and accuracy on a new task without AI. Distinguish model examination scores, student perceptions and actual learner gains; only the last directly tests educational benefit. Compare with baseline, document uncertainty and stop or revise if learning worsens, access is unequal or critical errors persist.

The board decision brief

Report the learning problem, evidence quality, participating population, total cost including review time, measured benefit, errors, access differences and remaining uncertainties. Make a dated recommendation to stop, revise or expand. A positive perception survey alone does not establish learning improvement.

Attribution and reuse

Chengcheng Yu; Fangcai Li; Ning Zhang; Hejia Hu; Hanxu Huang; Jingkai Wang; Yiqing Tao; Yinan Wu. Effectiveness of Artificial Intelligence–Assisted Peer Teaching in Orthopedic Clinical Education: Historical Cohort Study. JMIR Medical Education (2026-06-24). https://doi.org/10.2196/87959

Companion prepared 2026-10-30. Original authorship and DOI are preserved. The abstract is an attributed source extract; the implementation plan is new AI-assisted editorial material, offered under the same CC BY 4.0 license. This guide has not undergone independent peer review.