Evaluating ChatGPTs Responses to Chiropractic Patient Questions: A Comparative Study

ChatGPT (GPT-5.2) responses to 30 chiropractic patient questions scored highly on relevance (4.72 ± 0.56) and clarity (4.76 ± 0.50) versus chiropractic guideline concordance.

• Why it matters: Prior evaluations of LLMs in musculoskeletal and spine topics did not assess chiropractic-relevant content against chiropractic clinical practice guidelines.
• What they did: Thirty low back pain, neck pain, and chronic musculoskeletal pain questions were answered by ChatGPT in isolated sessions, then five licensed chiropractors rated relevance, clarity, completeness, and accuracy on 5-point Likert scales.
• The result: Readability was at approximately a college-level reading grade, and inter-rater reliability was poor (Krippendorffs alpha ordinal -0.023–0.214), implying guideline-concordant content with interpretive variability and patient-education readability concerns.