Synapse
⌘+K
Synapse
PulseExploreClubsResearchersJournals
Instagram
HomeClubsExplore
July 21, 2025DiagnosticsOpen Access

ChatGPT 4.0 provided mostly 'adequate' or 'excellent' responses to 26 questions on musculoskeletal disorders, though diagnosis ratings varied significantly by area of interest (p=0.007).

View Full Paper
Ask AI
Bookmark
Share

Why the study?

Patients increasingly attempt self-diagnosis and self-treatment, prompting evaluation of the reliability, comprehensibility, and readability of ChatGPT 4.0 responses for common orthopaedic problems.

Population

26 open-ended questions about the most prevalent orthopaedic problems

Design

Two-observer evaluation study

Key result

ChatGPT 4.0 provided mostly 'adequate' or 'excellent' responses to 26 questions on musculoskeletal disorders, though diagnosis ratings varied significantly by area of interest (p=0.007).

Authors

UAUfuk ArzuBGBatuhan Gencer

Discussion

Loading...

Member takes

Overview

Clinicians should advise against sole reliance on ChatGPT for common orthopaedic self-care; leaves open whether formal audits mitigate misguidance.

Study Design

Type

Cross-Sectional (n=26)

Structured PICO

P
Population
Evaluation of ChatGPT 4.0 responses to 26 open-ended questions concerning the most common musculoskeletal disorders.
E
Exposure
ChatGPT 4.0
O
Outcome
Reliability, comprehensibility, and readability of responses (evaluated using Likert scale in diagnosis, recommendation, and referral; FRES and FKGL)

Main Result

p-value: p=0.007

ChatGPT 4.0 provides generally adequate or excellent advice for musculoskeletal disorders with a high school reading level, though caution is needed for trauma-related conditions.

Cite This Study

Arzu et al. (2025) conducted a cross-sectional in Musculoskeletal disorders (n=26). ChatGPT 4.0 was evaluated on Reliability, comprehensibility, and readability of responses in categories of diagnosis, recommendation, and referral (p=0.007). ChatGPT 4.0 provided mostly 'adequate' or 'excellent' responses to 26 questions on musculoskeletal disorders, though diagnosis ratings varied significantly by area of interest (p=0.007).

synapsesocial.com/papers/6a8074fdfcb29c286731dc6chttps://doi.org/10.3390/diagnostics15141834
View Full Paper
Ask AI
Bookmark
Share

Also Consider

Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context:

  1. 1Is Readability a Proxy for Reliability? A Qualitative Evaluation of ChatGPT‐4.0 in Orthopaedic Trauma Communication2025 · 3 citations
  2. 2Pandora’s Box of Pediatric Sports Medicine: Assessing ChatGPT Responses to Patient-Centered Questions in Pediatric Sports Medicine2026
  3. 3ChatGPT is capable of providing satisfactory responses to frequently asked questions regarding total shoulder arthroplasty2024 · 7 citations
  4. 4Assessing ChatGPT Responses to Common Patient Questions on Knee Osteoarthritis2024 · 5 citations
  5. 5Assessing the Usability of ChatGPT Responses Compared to Other Online Information in Hand Surgery2025 · 1 citations