Generative AI–Assisted Microlearning for Erectile Dysfunction Myth Reduction: Single-Center Pre–Post Quasi-Experimental Study

Le résumé fourni par la source

Background: Erectile dysfunction (ED) is strongly influenced by persistent misconceptions that delay help-seeking and limit engagement with effective care. Patient-centered digital strategies, including generative artificial intelligence (AI) microlearning, may improve sexual-health knowledge; however, real-world evidence in urological practice remains sparse. Objective: This study aimed to evaluate whether a clinician-supervised generative AI microlearning video improves ED-related knowledge in adult men attending routine outpatient care. Methods: This single-center pre-post quasi-experimental study included 200 adult men in a university urology clinic. Participants completed an 8-item ED myth questionnaire immediately before and after watching a 3-minute educational video. The narration script was drafted using a large language model (ChatGPT) and iteratively reviewed by urologists for accuracy and cultural appropriateness. The primary outcome was the within-participant change in total correct responses (0-8). Subgroup analyses assessed effects across age (<40 years vs ≥40 years), education level, and self-reported ED. Paired analyses and multivariable logistic regression were used (α=.05). Results: All participants completed the intervention (mean age 44.0, SD 11.6 years). Total mean correct responses increased from 3.77 to 6.56 (mean difference 2.79; P<.001), indicating a large effect (Cohen d=1.52). Knowledge gains were consistent across subgroups, with greater improvements among those with lower education. Self-reported ED was independently associated with lower odds of achieving ≥2-point improvement (odds ratio 0.46, 95% CI 0.26-0.81; P=.01). No adverse events or technical difficulties occurred. Conclusions: A brief clinician-supervised generative AI microlearning video was associated with substantial short-term improvements in ED myth-related knowledge in routine outpatient care. AI-assisted microlearning may represent a scalable adjunct to patient education during urological consultations. Future studies should evaluate long-term retention and behavioral outcomes.

Ce résumé expose les affirmations des auteurs. BNTIC ne l’interprète pas comme une validation indépendante des résultats.