Impact of Using AI Symptom-Checking Tools Before Physical Therapy on Shared Decision-Making and Diagnosis in Adults with Spine Pain

What this Trial Is About

Researchers are examining how the use of artificial intelligence AI symptom-checking tools, including large language models like ChatGPT, before seeking medical care affects the initial physical therapy evaluation for adults with spine-related musculoskeletal complaints. This observational study focuses on adults with neck, thoracic, or low back pain, with or without radiating symptoms, comparing patients who used AI tools in the 30 days before their first physical therapy visit to those who did not. The goal is to understand how prior AI use influences shared decision-making and patient presentation during care. Participants are grouped based on whether they used an AI symptom-checking tool for their current spine complaint before their visit. All receive standard outpatient physical therapy care determined by their clinician, with no additional interventions introduced by the study. Data collection includes baseline questionnaires on AI use, symptom duration, pain, and disability, followed by a standard physical therapy assessment and classification during the initial visit. Follow-up assessments occur at 2 and 6 weeks to measure clinical outcomes, adherence, and healthcare utilization. During the study, participants complete surveys on their diagnostic expectations and shared decision-making immediately after the initial evaluation. Researchers collect data on pain intensity, disability levels, functional performance tests, and patient-clinician diagnosis agreement. Follow-up visits assess changes in symptoms, adherence to therapy, and any healthcare use. All information is recorded securely, with a total recruitment target of approximately 200 patients to ensure sufficient data for analysis.

Pre-Visit AI Symptom-Checking and Shared Decision-Making in Spine Physical Therapy