Accuracy and reproducibility of ChatGPT responses to real‐world drug information questions
Open Access
- 22 April 2025
- journal article
- Published by Wiley in JACCP: Journal of the American College of Clinical Pharmacy
- Vol. 8 (6) , 432-438
Abstract
Introduction: The expanding use of Chat Generative Pre‐Trained Transformer (ChatGPT, OpenAI, San Francisco, CA) for drug information may enhance access to information. However, it is crucial to assess the accuracy and reproducibility of ChatGPT responses to drug information questions, examining its utility and limitations in clinical decision‐making.Objective: To evaluate the accuracy and reproducibility of ChatGPT‐3.5 and ChatGPT‐4 in responding to clinician drug information questions compared with a commonly accepted resource, Lexicomp®(Wolters Kluwer Health, Philadelphia, PA).Methods: A serial cross‐sectional study was conducted on ChatGPT from March 5 to 12, 2024 in the United States. ChatGPT‐3.5 is a free, artificial intelligence (AI) chatbot trained up to January 2022; ChatGPT‐4 is a paid‐subscription AI chatbot with internet access and more data. For trial 1 (day 0) we input 30 real‐world questions (10 drug information categories) into both ChatGPT‐3.5 and ChatGPT‐4. For trial 2 (day 1) and 3 (day 7), 10 randomly selected questions were re‐input into ChatGPT. The primary outcome evaluated the accuracy of ChatGPT‐3.5 responses versus (vs.) Lexicomp® using a 4‐point Likert scale. Secondary outcomes included assessing the accuracy of ChatGPT‐4 responses vs. Lexicomp, comparing the accuracy of both ChatGPT versions' responses, and comparing reproducibility of ChatGPT responses over time. Cohen's Kappa and Cochran's Q assessed reproducibility.Results: ChatGPT‐3.5 demonstrated 30% accuracy (9/30), while ChatGPT‐4 had 40% (12/30) accuracy (p = 0.51). Neither ChatGPT versions accurately answered all the questions in any category. ChatGPT‐3.5's agreement between trials 1 vs. 2, 1 vs. 3, and 2 vs. 3 had fair (k = 0.21), moderate (k = 0.41), and substantial agreement (k = 0.62), respectively. ChatGPT‐4 trials 1 vs. 2, 1 vs. 3, and 2 vs. 3 had fair (k = 0.23), substantial (k = 0.80), and fair agreement (0.40). The accuracy of ChatGPT‐3.5 vs. ChatGPT‐4 for the 10 questions across the three trials was 30%, 20%, and 10% (p = 0.78), and 60%, 40%, and 50% (p = 0.82).Conclusions: Both ChatGPT versions demonstrated limited accuracy and reproducibility in answering drug information questions, suggesting that health care professionals should exercise caution when using ChatGPT for drug information.Keywords
This publication has 24 references indexed in Scilit:
- Patient Safety and Artificial Intelligence in Clinical CareJAMA Health Forum, 2024
- Health Disinformation Use Case Highlighting the Urgent Need for Artificial Intelligence VigilanceJAMA Internal Medicine, 2024
- Effectiveness of ChatGPT in clinical pharmacy and the role of artificial intelligence in medication therapy managementJournal of the American Pharmacists Association, 2023
- Examining the Potential of ChatGPT on Biomedical Information Retrieval: Fact-Checking Drug-Disease AssociationsAnnals of Biomedical Engineering, 2023
- Assessing the accuracy and consistency of ChatGPT in clinical pharmacy management: A preliminary analysis with clinical pharmacy experts worldwideResearch in Social and Administrative Pharmacy, 2023
- Evaluation of community pharmacists’ perceptions and willingness to integrate ChatGPT into their pharmacy practice: A study from JordanJournal of the American Pharmacists Association, 2023
- Impact of artificial intelligence: A perspective on human life and pharmacy practiceJACCP: Journal of the American College of Clinical Pharmacy, 2023
- Efficacy of personalized models in discriminating high cognitive demand conditions using text-based interactionsInternational Journal of Human-Computer Studies, 2017
- The problem of information overload in business organisations: a review of the literatureInternational Journal of Information Management, 2000
- A capacity theory of comprehension: Individual differences in working memory.Psychological Review, 1992