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

Официальное название

When AI Is the First Clinician: Impact of Pre-Visit AI Use on Presentation, Diagnostic Expectations, and Shared Decision-Making in Spine Physical Therapy

Обзор

Artificial intelligence (AI) symptom-checking tools, including large language models such as ChatGPT, are increasingly used by patients before they seek care. These tools may shape patients' beliefs about their diagnosis, how serious they think their condition is, and when they decide to seek treatment. It is not yet known how this pre-visit AI use affects the initial physical therapy encounter for spine-related problems. This prospective observational cohort study examines whether prior use of AI symptom-checking tools influences the first physical therapy evaluation in adults presenting with spine-related musculoskeletal complaints (neck, thoracic, or low back pain, with or without radicular symptoms). Consecutive patients attending an outpatient physical therapy clinic for a new evaluation are grouped as AI users or non-AI users based on whether they used such a tool for their current complaint in the previous 30 days. The primary outcome is shared decision-making, measured with the SDM-Q-9 immediately after the initial evaluation. Secondary outcomes include stage of presentation, agreement between the patient's expected diagnosis and the clinician's classification, baseline pain and disability, functional performance, and clinical outcomes at 2 and 6 weeks. The investigators hypothesize that prior AI use is associated with differences in shared decision-making and in how patients present for care.

Подробное описание

Background and Rationale AI-based symptom-checking tools, including large language models, are increasingly used by patients as a first point of clinical interpretation before seeking care. Available evidence suggests these tools show moderate and variable diagnostic accuracy and can influence patients' diagnostic beliefs, perceived symptom severity, and healthcare-seeking behavior, raising concerns about pre-diagnostic anchoring. In spine and musculoskeletal care, early presentation, accurate classification, and patient engagement are important determinants of outcomes, and current practice relies on history, physical examination, and shared decision-making under the assumption that patients present without strongly preformed diagnostic expectations. The growing use of AI may alter this dynamic. Evidence on the real-world impact of pre-visit AI use on the physical therapy encounter is currently limited.

Objective The primary objective is to determine whether prior use of AI symptom-checking tools influences shared decision-making during the initial physical therapy evaluation in patients presenting with spine-related musculoskeletal conditions. Secondary objectives examine associations between prior AI use and stage of presentation, agreement between patient-expected and clinician diagnosis, baseline clinical status, functional performance, health-seeking behavior, and short-term clinical outcomes.

Design and Setting This is a prospective observational cohort study conducted in outpatient physical therapy clinics. No study-specific interventions are introduced; all participants receive standard physical therapy care determined by their treating clinician.

Participants and Grouping Consecutive adults (≥18 years) presenting for a new evaluation for a spine-related musculoskeletal complaint (neck, thoracic, or low back pain, with or without radicular symptoms) who can provide informed consent and complete study questionnaires in English are eligible. Patients are excluded for recent spinal surgery within the past 3 months, serious spinal pathology under active medical management (e.g., malignancy, spinal infection, acute fracture), cognitive impairment limiting consent or reliable completion of measures, or current enrollment in another study that may influence clinical decision-making or spine-related outcomes. Participants are classified into two groups based on whether they used an AI symptom-checking tool for their current complaint within the prior 30 days: AI users and non-AI users.

Study Instruments and Measures AI exposure is captured at baseline via a questionnaire assessing AI use (yes/no), tool type, frequency, degree of personalization, and reported influence on care-seeking. Clinical measures include symptom duration, pain intensity (Numeric Pain Rating Scale, 0-10), region-specific disability (Oswestry Disability Index for low back pain or Neck Disability Index for neck pain), and the Five Times Sit-to-Stand test. Diagnostic expectation is assessed by comparing the patient-expected diagnosis with the clinician's classification, categorized as match, partial match, or mismatch. Shared decision-making is measured with the SDM-Q-9 (score range 0-45; higher scores indicate greater patient involvement) immediately after the initial evaluation. Follow-up outcomes at 2 and 6 weeks include pain, disability (ODI/NDI), Global Rating of Change, self-reported adherence, and healthcare utilization.

Procedures At baseline (pre-visit), AI exposure, symptom duration, pain, and disability are collected. During the initial evaluation, the clinician performs a standard physical therapy assessment and classification, and patient expectations are recorded. Immediately post-visit, participants complete the SDM-Q-9. At 2-week and 6-week follow-up, clinical outcomes and adherence are assessed and AI use is re-evaluated. Data are recorded on standardized forms and stored securely using coded identifiers.

Sample Size The primary outcome is SDM-Q-9 following the initial evaluation. Using G\*Power (v3.1.9.7), a sample of 128 participants (64 per group) provides 80% power to detect a moderate between-group difference (Cohen's d = 0.5) at a two-sided alpha of 0.05. To account for potential group imbalance and an anticipated attrition of 15-20%, approximately 200 consecutive patients will be recruited.

Statistical Analysis Continuous variables are summarized as mean ± standard deviation and categorical variables as frequencies and percentages. AI users and non-AI users are compared using independent-samples t-tests for continuous variables and chi-square tests for categorical variables. Statistical significance is set at p ≤ 0.05.

Вмешательства

• Другое re-visit AI symptom-checker use
  Self-reported use of an AI-based symptom-checking tool (e.g., a large language model such as ChatGPT) for the current spine-related complaint during the 30 days before the initial physical therapy evaluation. This exposure occurs naturally prior to presentation and is not assigned by the investigator. Exposure status is ascertained at baseline via a questionnaire capturing whether AI was used (yes/no), the type of tool, frequency of use, degree of personalization, and the reported influence of A

Первичные конечные точки

• Shared Decision-Making (SDM-Q-9) [Срок оценки: Immediately after the initial physical therapy evaluation (Day 0)]

Вторичные конечные точки (6)

• Stage of Presentation (Symptom Duration) [Срок оценки: At initial evaluation (Day 0)]
• Diagnostic Agreement (Patient-Clinician Concordance) [Срок оценки: At initial evaluation (Day 0)]
• Pain Intensity (Numeric Pain Rating Scale) [Срок оценки: Day 0, 2 weeks, 6 weeks]
• Disability (ODI or NDI) [Срок оценки: Day 0, 2 weeks, 6 weeks]
• Health-Seeking Behavior (AI Influence on Care Timing) [Срок оценки: At initial evaluation (Day 0)]
• Adherence and Healthcare Utilization [Срок оценки: 2 weeks, 6 weeks]

Критерии участия

Критерии включения

• Age: 18 years or older
• Presenting for a new evaluation at an outpatient physical therapy clinic for a spine-related musculoskeletal complaint, including Neck, Thoracic, and Low back pain With or without radicular symptoms
• Able to provide informed consent.
• Able to read, understand, and complete study questionnaires

Критерии исключения

• Recent spinal surgery within the past 3 months, due to differing clinical pathways and management strategies.
• Presence of serious spinal pathology under active medical management, such as:

Malignancy (e.g., metastatic disease) Spinal infection Acute fracture

• Cognitive impairment or other conditions that limit the ability to provide informed consent or reliably complete study measures.
• Patients currently enrolled in another study that may influence clinical decision-making or outcomes related to spine care.

Критерии приведены из реестра в оригинале (на английском). Окончательную оценку соответствия проводит исследовательский центр.

Здоровые добровольцы: Нет

Дизайн исследования

Модель наблюдения

Когортное

Центры проведения

Египет · 1 центр

• Faculty of Medicine, Assiut University, Egypt — Asyut

Публикации

• Harvey D, White S, Reid D, Cook C. Patient perspectives of process variables in musculoskeletal care pathways. Musculoskelet Sci Pract. 2025 Apr;76:103287. doi: 10.1016/j.msksp.2025.103287. Epub 2025 Feb 14. PMID 39978029
• Meyer AND, Giardina TD, Spitzmueller C, Shahid U, Scott TMT, Singh H. Patient Perspectives on the Usefulness of an Artificial Intelligence-Assisted Symptom Checker: Cross-Sectional Survey Study. J Med Internet Res. 2020 Jan 30;22(1):e14679. doi: 10.2196/14679. PMID 32012052
• Kuroiwa T, Sarcon A, Ibara T, Yamada E, Yamamoto A, Tsukamoto K, Fujita K. The Potential of ChatGPT as a Self-Diagnostic Tool in Common Orthopedic Diseases: Exploratory Study. J Med Internet Res. 2023 Sep 15;25:e47621. doi: 10.2196/47621. PMID 37713254
• Kumar R, Dougherty C, Sporn K, Khanna A, Ravi P, Prabhakar P, Zaman N. Intelligence Architectures and Machine Learning Applications in Contemporary Spine Care. Bioengineering (Basel). 2025 Sep 9;12(9):967. doi: 10.3390/bioengineering12090967. PMID 41007212
• Bensel VA, Habeck A, Brunot MH, Becton EJ, Ray M, Brackett AL, Lisi AJ. Artificial intelligence in spine care: A scoping review of treatment applications. N Am Spine Soc J. 2025 Nov 20;25:100827. doi: 10.1016/j.xnsj.2025.100827. eCollection 2026 Mar. PMID 41536317
• Muelbauer EJ, Alvi MA, Kennedy DJ, Fehlings MG. The future is now: How AI is reshaping spine care. N Am Spine Soc J. 2025 Nov 14;24:100825. doi: 10.1016/j.xnsj.2025.100825. eCollection 2025 Dec. PMID 41458002
• Rossettini G, Bargeri S, Cook C, Guida S, Palese A, Rodeghiero L, Pillastrini P, Turolla A, Castellini G, Gianola S. Accuracy of ChatGPT-3.5, ChatGPT-4o, Copilot, Gemini, Claude, and Perplexity in advising on lumbosacral radicular pain against clinical practice guidelines: cross-sectional study. Front Digit Health. 2025 Jun 27;7:1574287. doi: 10.3389/fdgth.2025.1574287. eCollection 2025. PMID 40657647
• Basharat A, Shah R, Wilcox N, Tur G, Tripati S, Kansal P, Gandhi N, Pokuri S, Chong G, Odonkor CA, Varhabhatla N, Chow R. ChatGPT and low back pain - Evaluating AI-driven patient education in the context of interventional pain medicine. Interv Pain Med. 2025 Sep 2;4(3):100636. doi: 10.1016/j.inpm.2025.100636. eCollection 2025 Sep. PMID 40978326

Идентификаторы

NCT: NCT07733752 · 04-2026-300876

Первоисточники (государственные реестры)

• ClinicalTrials.gov (NIH, США)

Открыть это исследование на ClinicalTrials.gov ↗