Truace
Subscribe

Expert evaluation of ChatGPT-generated responses to patient questions in conventional and fast-track total knee arthroplasty

Background: Fast-track total knee arthroplasty (TKA) asks patients to understand early mobilization, early discharge, opioid-sparing analgesia, and selective use of drains and catheters. We evaluated expert-rated quality of ChatGPT responses to patient questions about these pathway-specific concepts alongside conventional TKA questions. Materials and methods: In this cross-sectional evaluation, three arthroplasty surgeons independently scored responses to 50 conventional TKA and 20 investigator-developed fast-tr…

By the Truvace desk2 min read1 sourcePeer-reviewedTRV-2026-1547· cite & verify
Expert evaluation of ChatGPT-generated responses to patient questions in conventional and fast-track total knee arthroplasty
Physical-Therapy-Dept,-Deshon-General-Hospital,-West-end-whirlpool-room by The US government, most likely the US National Institutes of Health .. Public domain

In brief

Researchers had three arthroplasty surgeons independently score ChatGPT's answers to 70 patient questions: 50 on conventional total knee replacement and 20 on fast-track care, such as early mobilization and early discharge. The 70 responses produced 210 expert assessments across four domains on 5-point scales.

Patients on fast-track pathways must grasp several protocol-specific ideas, so reliable answers could ease information needs. However, the ratings piled up at the maximum, the two question sets were built differently, and the authors say the fast-track ratings need source-data verification, so the study cannot show that the chatbot performs equally on both sets or is safe for unsupervised use.

Main points

  1. Seventy responses received 210 assessments from three arthroplasty surgeons.
  2. Mean scores were 4.99 for accuracy, 4.98 for completeness, 4.98 for safety and 5.00 for readability.
  3. All 20 fast-track responses got a 5 from every rater in every domain.
  4. The conventional questions were chosen by a multidisciplinary panel, while the fast-track questions did not go through that selection process.
  5. The authors say clinician review remains appropriate for protocol-dependent and safety-related advice.

The gain

Three arthroplasty surgeons rated ChatGPT's answers to patient questions about knee replacement, including fast-track pathway concepts, at or near the top of the scale for accuracy, completeness, safety and readability.

The rundown

The study was a cross-sectional evaluation. Fifty conventional knee replacement questions were selected by a multidisciplinary panel, and 20 investigator-developed questions covered fast-track topics: early mobilization, early discharge, opioid-sparing analgesia, and selective use of drains and catheters.

Scoring covered four domains on 5-point scales. Response-level means were compared descriptively and with exploratory Mann-Whitney U tests. Recalculated means across individual ratings were close to the 5-point ceiling in every domain.

The authors conclude that the data do not establish equivalence between the two question sets. They cite differing question construction, the need to verify the fast-track source data, and score clustering at the maximum, and they recommend continued clinician review for protocol-dependent and safety-related advice.

Sources

  1. Peer-reviewedArchives of Orthopaedic and Trauma Surgery2026-10-15

The debate