A cross-sectional study on ChatGPT’s alignment with clinical practice guidelines in musculoskeletal rehabilitation

BMC Musculoskeletal Disorders, Journal Year: 2025, Volume and Issue: 26(1)

Published: April 24, 2025

AI models like ChatGPT have the potential to support musculoskeletal rehabilitation by providing clinical insights. However, their alignment with evidence-based guidelines needs evaluation before integration into physiotherapy practice. To evaluate performance of (GPT-4 model) in generating responses queries comparing its recommendations practice (CPGs). This study was designed as a cross-sectional observational study. Twenty questions covering disease information, assessment, and were developed two experienced physiotherapists specializing disorders. The distributed across three anatomical regions: upper extremity (7 questions), lower (9 spine (4 questions). ChatGPT's obtained evaluated independently raters using 5-point Likert scale assessing relevance, accuracy, clarity, completeness, consistency. Weighted kappa values calculated assess inter-rater agreement consistency within each category. received highest average score for clarity (4.85), followed accuracy (4.62), relevance (4.50), completeness (4.20). Consistency lowest (3.85). (weighted = 0.90) observed information category, whereas displayed relatively 0.56). Variability moderate weighted highlighted areas requiring improvement. demonstrates guideline-aligned rehabilitation. due limitations consistency, ability replicate nuanced reasoning, use should remain supplementary rather than primary decision-making tool. While it performed better evidenced higher scores, category comparatively lower, highlighting challenges addressing complex, therapeutic interventions. variability domain-specific reasoning underscores need further refinement ensure reliability complex scenarios. Not applicable.

Language: Английский

Mapping global research on artificial intelligence in physical therapy: a bibliometric analysis from 1990 to 2023

European Journal of Physiotherapy, Journal Year: 2025, Volume and Issue: unknown, P. 1 - 11

Published: May 1, 2025

Language: Английский