Can ChatGPT Answer Patient Questions Regarding Ankle Fractures & Fixation?

Category: Trauma; Ankle Introduction/Purpose: The contemporary patient has access to numerous resources on common orthopaedic procedures. Recently, artificial intelligence (AI) driven chatbots have become mainstream, allowing patients to interact with interfaces which supply convincing, human-like a...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:Foot & ankle orthopaedics 2024-12, Vol.9 (4)
Hauptverfasser: Kelley, Raegan, Mika, Aleksander, Luther, Lauren, Schultz, Jacob, Mitchell, Phillip, Strasser, Nicholas L.
Format: Artikel
Sprache:eng
Schlagworte:
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:Category: Trauma; Ankle Introduction/Purpose: The contemporary patient has access to numerous resources on common orthopaedic procedures. Recently, artificial intelligence (AI) driven chatbots have become mainstream, allowing patients to interact with interfaces which supply convincing, human-like answers to posed questions. ChatGPT, a recently developed AI-based chat application, is one such application that has garnered rapid growth and popularity and exhibited high functionality in answering common patient questions regarding a gambit of orthopaedic procedures. While ChatGPT has previously demonstrated reliability in answering questions regarding elective orthopaedic procedures, there is little evidence of its effectiveness in the acute, traumatic setting. Given the likelihood that patients may call on this technology for preoperative education, we sought to determine whether ChatGPT could appropriately answer frequently asked questions (FAQs) regarding traumatic ankle fractures. Methods: Ten FAQs regarding ankle fractures were posed to the chatbot during a conversation thread with no follow up questions or repetition. Each FAQ was analyzed for accuracy using an evidence-based approach. Answers were rated as “Excellent requiring no clarification”, “Satisfactory requiring minimal clarification”, Satisfactory requiring moderate clarification” or “Unsatisfactory response requiring significant clarification.” Results: Of the answers given by the chatbot none received an “unsatisfactory” rating with four not requiring any correction and the majority either requiring minimal (4/10) or moderate (2/10) clarification. Although several answers require nuanced clarification, the chatbot's responses were generally unbiased and evidence based, even for controversial topics. Conclusion: The chatbot effectively provides evidence-based responses to common patient FAQs regarding ankle fractures and fixation, presenting the information in a way that most patients can understand. This resource may serve as a valuable clinical tool in the future especially in the acute, traumatic setting.
ISSN:2473-0114
2473-0114
DOI:10.1177/2473011424S00171