Cited by 273Artificial Intelligence in Healthcare and EducationTotal Knee Arthroplasty OutcomesSocial Media in Health Education

Assessing ChatGPT Responses to Common Patient Questions Regarding Total Hip Arthroplasty

Mika, Aleksander P., Martin, J. Ryan, Engstrom, Stephen M., Polkowski, Gregory G., Wilson, Jacob M.·2023·Journal of Bone and Joint Surgery

Abstract

Background: The contemporary patient has access to numerous resources on common orthopaedic procedures before ever presenting for a clinical evaluation. Recently, artificial intelligence (AI)-driven chatbots have become mainstream, allowing patients to engage with interfaces that supply convincing, human-like responses to prompts. ChatGPT (OpenAI), a recently developed AI-based chat technology, is one such application that has garnered rapid growth in popularity. Given the likelihood that patients may soon call on this technology for preoperative education, we sought to determine whether ChatGPT could appropriately answer frequently asked questions regarding total hip arthroplasty (THA). Methods: Ten frequently asked questions regarding total hip arthroplasty were posed to the chatbot during a conversation thread, with no follow-up questions or repetition. Each response was analyzed for accuracy with use of an evidence-based approach. Responses were rated as “excellent response not requiring clarification,” “satisfactory requiring minimal clarification,” “satisfactory requiring moderate clarification,” or “unsatisfactory requiring substantial clarification.” Results: Of the responses given by the chatbot, only 1 received an “unsatisfactory” rating; 2 did not require any correction, and the majority required either minimal (4 of 10) or moderate (3 of 10) clarification. Although several responses required nuanced clarification, the chatbot’s responses were generally unbiased and evidence-based, even for controversial topics. Conclusions: The chatbot effectively provided evidence-based responses to questions commonly asked by patients prior to THA. The chatbot presented information in a way that most patients would be able to understand. This resource may serve as a valuable clinical tool for patient education and understanding prior to orthopaedic consultation in the future.

Integrity signals

Registry

Crossref · 10.2106/jbjs.23.00209

Cites

105 of 105 references verified0 citation errors · 0 not checkable

Statistics

Recomputing p-values needs the paper’s full text.

Claims the paper makes

This paper’s full text isn’t openly available, so its claims and self-check weren’t run. The registry and citation checks still apply.

Claims other papers make about this one

Each is checked against this paper’s own text: does it say what the citing paper says it does?

Cites

Flagged first, then by citations

Cited by

Sorted by citations

Authors
Mika, A. P., Martin, J. R., Engstrom, S. M., Polkowski, G. G., & Wilson, J. M.
Affiliations
Vanderbilt University Medical Center (US)
Venue
Journal of Bone and Joint Surgery
Published
2023-07-17
Volume
105
Issue
19
Pages
1519-1526
ISSN
  • 0021-9355 (ISSN-L)
  • 1535-1386
OpenAlex
W4384499300
Cited by
273
References
105
Citations by year
2026: 66 · 2025: 99 · 2024: 100 · 2023: 7 · 2012: 1
FWCI
10.83
Type
journal article
Language
en
Open access
closed
Locations
  • Journal of Bone and Joint Surgery (publishedVersion)
  • PubMed (publishedVersion)
Topics
  • Health Sciences › Medicine › Health Informatics › Artificial Intelligence in Healthcare and Education
  • Health Sciences › Medicine › Surgery › Total Knee Arthroplasty Outcomes
  • Social Sciences › Health › Social Media in Health Education
Keywords
ChatGPT, total hip arthroplasty, chatbots, patient education, preoperative information, frequently asked questions, artificial intelligence, orthopedic consultation
MeSH terms
Artificial Intelligence, Communication, Humans, Orthopedics, Orthopedic Procedures, Arthroplasty, Replacement, Hip

About these checks

Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.

Statistics and arithmetic are recomputed, not judged.

Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.

What counts as a claim ›How we measure confidence ›How we check this ›

Are you an author, or do you see a problem with this page? Request a review