ChatGPT-3.5 and -4 provide mostly accurate information when answering patients’ questions relating to femoroacetabular impingement syndrome and arthroscopic hip surgery
Bibliographic record
Abstract
OBJECTIVES: This study aimed to evaluate the accuracy of ChatGPT in answering patient questions about femoroacetabular impingement (FAI) and arthroscopic hip surgery, comparing the performance of versions ChatGPT-3.5 (free) and ChatGPT-4 (paid). METHODS: Twelve frequently asked questions (FAQs) relating to FAI were selected and posed to ChatGPT-3.5 and ChatGPT-4. The responses were assessed for accuracy by three hip arthroscopy surgeons using a four-tier grading system. Statistical analyses included Wilcoxon signed-rank tests and Gwet's AC2 coefficient for interrater agreement corrected for chance and employing quadratic weights. RESULTS: The median ratings for responses ranged from "excellent not requiring clarification" to "satisfactory requiring moderate clarification." No responses were rated as "unsatisfactory requiring substantial clarification." The median accuracy scores were 2 (range 1-3) for ChatGPT-3.5 and 1.5 (range 1-3) for ChatGPT-4, with 25 % of ChatGPT-3.5's responses and 50 % of ChatGPT-4's responses rated as "excellent." There was no statistical difference in performance between the two versions (p = 0.279) although ChatGPT-4 showed a tendency towards higher accuracy in some areas. Interrater agreement was substantial for ChatGPT-3.5 (Gwet's AC2 = 0.79 [95% confidence interval (CI) = 0.6-0.94]) and moderate to substantial for ChatGPT-4 (Gwet's AC2 = 0.65 [95% CI = 0.43-0.87]). CONCLUSION: Both versions of ChatGPT provided mostly accurate responses to FAQs on FAI and arthroscopic surgery, with no significant difference between the versions. The findings suggest potential utility of ChatGPT in patient education, though cautious implementation and further evaluation are recommended due to variability in response accuracy and low power of the study. LEVEL OF EVIDENCE: IV.
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 | 0.002 | 0.001 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.000 |
| Bibliometrics | 0.001 | 0.000 |
| 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".