Impact of artificial intelligence in managing musculoskeletal pathologies in physiatry: a qualitative observational study evaluating the potential use of ChatGPT versus Copilot for patient information and clinical advice on low back pain
Bibliographic record
Abstract
BACKGROUND: The self-management of low back pain (LBP) through patient information interventions offers significant benefits in terms of cost, reduced work absenteeism, and overall healthcare utilization. Using a large language model (LLM), such as ChatGPT (OpenAI) or Copilot (Microsoft), could potentially enhance these outcomes further. Thus, it is important to evaluate the LLMs ChatGPT and Copilot in providing medical advice for LBP and assessing the impact of clinical context on the quality of responses. METHODS: This was a qualitative comparative observational study. It was conducted within the Department of Physical Medicine and Rehabilitation, University of Montreal in Montreal, QC, Canada. ChatGPT and Copilot were used to answer 27 common questions related to LBP, with and without a specific clinical context. The responses were evaluated by physiatrists for validity, safety, and usefulness using a 4-point Likert scale (4, most favorable). RESULTS: Both ChatGPT and Copilot demonstrated good performance across all measures. Validity scores were 3.33 for ChatGPT and 3.18 for Copilot, safety scores were 3.19 for ChatGPT and 3.13 for Copilot, and usefulness scores were 3.60 for ChatGPT and 3.57 for Copilot. The inclusion of clinical context did not significantly change the results. CONCLUSION: LLMs, such as ChatGPT and Copilot, can provide reliable medical advice on LBP, irrespective of the detailed clinical context, supporting their potential to aid in patient self-management.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame distilled prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.
Codex and Gemma teacher scores by category
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.013 | 0.020 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.000 | 0.001 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.001 |
| Open science | 0.000 | 0.000 |
| Research integrity | 0.000 | 0.000 |
| Insufficient payload (model declined to judge) | 0.000 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one teacher head, not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".