Effectiveness of AI-Assisted Orthopedic Nursing Education: A Mixed-Methods Randomized Controlled

<title>Abstract</title> <p>Background Orthopedic nursing education requires learners to convert imaging cues into patient assessment, risk identification, and care planning, but traditional teaching provides limited visualization and feedback. This study aimed to evaluate and interpret the effectiveness of an AI-assisted orthopedic nursing education using DDH as a teaching case. Methods Sixty nursing interns were randomized, and 56 participants with complete post-training data were included in the final analysis. Both groups received the same DDH educational content, and the intervention group additionally used an AI-assisted tool. We analyzed the quantitative data using IBM SPSS Statistics 27.0, including chi-square or Fisher’s exact tests, independent-samples t tests, repeated-measures analysis of variance, Mann–Whitney U tests, and Wilcoxon signed-rank tests. Interviews with 10 intervention participants were analyzed using Colaizzi’s method. Results The intervention group had higher post-training scores on DDH-related nursing assessment (MD = 3.72, 95% CI [0.96, 6.48], d = 0.72, p = 0.009), DDH classification (MD = 4.07, 95% CI [0.17, 7.97], d = 0.56, p = 0.041), and deep learning (MD = 5.64, 95% CI [0.25, 11.03], d = 0.56, p = 0.040). However, the time-by-group interactions for deep learning and DDH-related nursing assessment were not statistically significant, so the evidence for differential pre-post change between groups was limited. In exploratory analyses, several AI learning intention factors showed within-group increases in the intervention group. Qualitative findings indicated that AI-assisted imaging education helped interns integrate image interpretation with patient assessment, nursing decision-making, and care planning. The findings also highlighted potential limitations, including the risk of undermining independent thinking and insufficient consideration of students' baseline knowledge and abilities. Conclusions AI-assisted orthopedic nursing education was associated with higher post-training scores in DDH classification and nursing assessment, by helping interns translate imaging findings into clinically relevant information. However, several secondary outcomes did not show significant group-by-time interaction effects. Therefore, these findings should be interpreted as preliminary. Future implementation should preserve opportunities for independent thinking and incorporate differentiated instruction tailored to students' baseline knowledge and abilities. Trial registration The study was prospectively registered with the Chinese Clinical Trial Registry (ChiCTR2600119719) on March 3, 2026.</p>