The Accuracy and Potential Racial and Ethnic Biases of GPT-4 in the Diagnosis and Triage of Health Conditions: Evaluation Study
Open accessThe DOI matches its Crossref record; 13 of the 13 references it cites were found in a registry or an index.
Abstract
Background: Whether GPT-4, the conversational artificial intelligence, can accurately diagnose and triage health conditions and whether it presents racial and ethnic biases in its decisions remain unclear. Objective: We aim to assess the accuracy of GPT-4 in the diagnosis and triage of health conditions and whether its performance varies by patient race and ethnicity. Methods: We compared the performance of GPT-4 and physicians, using 45 typical clinical vignettes, each with a correct diagnosis and triage level, in February and March 2023. For each of the 45 clinical vignettes, GPT-4 and 3 board-certified physicians provided the most likely primary diagnosis and triage level (emergency, nonemergency, or self-care). Independent reviewers evaluated the diagnoses as “correct” or “incorrect.” Physician diagnosis was defined as the consensus of the 3 physicians. We evaluated whether the performance of GPT-4 varies by patient race and ethnicity, by adding the information on patient race and ethnicity to the clinical vignettes. Results: The accuracy of diagnosis was comparable between GPT-4 and physicians (the percentage of correct diagnosis was 97.8% (44/45; 95% CI 88.2%-99.9%) for GPT-4 and 91.1% (41/45; 95% CI 78.8%-97.5%) for physicians; P=.38). GPT-4 provided appropriate reasoning for 97.8% (44/45) of the vignettes. The appropriateness of triage was comparable between GPT-4 and physicians (GPT-4: 30/45, 66.7%; 95% CI 51.0%-80.0%; physicians: 30/45, 66.7%; 95% CI 51.0%-80.0%; P=.99). The performance of GPT-4 in diagnosing health conditions did not vary among different races and ethnicities (Black, White, Asian, and Hispanic), with an accuracy of 100% (95% CI 78.2%-100%). P values, compared to the GPT-4 output without incorporating race and ethnicity information, were all .99. The accuracy of triage was not significantly different even if patients’ race and ethnicity information was added. The accuracy of triage was 62.2% (95% CI 46.5%-76.2%; P=.50) for Black patients; 66.7% (95% CI 51.0%-80.0%; P=.99) for White patients; 66.7% (95% CI 51.0%-80.0%; P=.99) for Asian patients, and 62.2% (95% CI 46.5%-76.2%; P=.69) for Hispanic patients. P values were calculated by comparing the outputs with and without conditioning on race and ethnicity. Conclusions: GPT-4’s ability to diagnose and triage typical clinical vignettes was comparable to that of board-certified physicians. The performance of GPT-4 did not vary by patient race and ethnicity. These findings should be informative for health systems looking to introduce conversational artificial intelligence to improve the efficiency of patient diagnosis and triage.
Overview
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
Citations
Cites
High-performance medicine: the convergence of human and artificial intelligence
Topol, E. J. (2019) · Nature Medicine · doi:10.1038/s41591-018-0300-7 · Cited by 10,450
The Triple Aim: Care, Health, And Cost
Berwick, D. M., Nolan, T. W., & Whittington, J. (2008) · Health Affairs · doi:10.1377/hlthaff.27.3.759 · Cited by 5,136
Kung, T. H., Cheatham, M., Medenilla, A., Sillos, C., De Leon, L., Elepaño, C., Madriaga, M., Aggabao, R., Diaz-Candido, G., Maningo, J., et al. (2023) · PLOS Digital Health · doi:10.1371/journal.pdig.0000198 · Cited by 4,115 · cited 1×
“Indeed, the latest study has reported that ChatGPT, an older model of GPT, passed the United States Medical Licensing Examination (USMLE) with moderate accuracy and high concordance [14].”
BackgroundKung, T. H., Cheatham, M., ChatGPT, Medenilla, A., Sillos, C., De Leon, L., Elepaño, C., Madriaga, M., Aggabao, R., Diaz-Candido, G., et al. (2022) · doi:10.1101/2022.12.19.22283643 · Cited by 4,115
Key challenges for delivering clinical impact with artificial intelligence
Kelly, C. J., Karthikesalingam, A., Suleyman, M., Corrado, G., & King, D. (2019) · BMC Medicine · doi:10.1186/s12916-019-1426-2 · Cited by 3,136
STARD 2015 guidelines for reporting diagnostic accuracy studies: explanation and elaboration
Cohen, J. F., Korevaar, D. A., Altman, D. G., Bruns, D. E., Gatsonis, C. A., Hooft, L., Irwig, L., Levine, D., Reitsma, J. B., de Vet, H. C. W., et al. (2016) · BMJ Open · doi:10.1136/bmjopen-2016-012799 · Cited by 2,570
ChatGPT: five priorities for research
van Dis, E. A. M., Bollen, J., Zuidema, W., van Rooij, R., & Bockting, C. L. (2023) · Nature · doi:10.1038/d41586-023-00288-7 · Cited by 1,842
Evaluation of symptom checkers for self diagnosis and triage: audit study
Semigran, H. L., Linder, J. A., Gidengil, C., & Mehrotra, A. (2015) · BMJ · doi:10.1136/bmj.h3480 · Cited by 489 · cited 6×
“Second, the clinical vignettes used in this study were publicly available in PDF format [11].”
BackgroundMethodResultInfluentialHuh, S. (2023) · Journal of Educational Evaluation for Health Professions · doi:10.3352/jeehp.2023.20.1 · Cited by 278 · cited 1×
“The potential of LLMs to revolutionize health care through improved diagnostics, personalized treatment plans, and enhanced patient-provider communication is widely recognized, making them a focal point for research and investment [4].”
BackgroundVasey, B., Ursprung, S., Beddoe, B., Taylor, E. H., Marlow, N., Bilbro, N., Watkinson, P., & McCulloch, P. (2021) · JAMA Network Open · doi:10.1001/jamanetworkopen.2021.1276 · Cited by 154
Algorithmic Political Bias in Artificial Intelligence Systems
Peters, U. (2022) · Philosophy & Technology · doi:10.1007/s13347-022-00512-8 · Cited by 111
Clinical Decision Support Systems for Diagnosis in Primary Care: A Scoping Review
Harada, T., Miyagami, T., Kunitomo, K., & Shimizu, T. (2021) · International Journal of Environmental Research and Public Health · doi:10.3390/ijerph18168435 · Cited by 89
Boonstra, A., & Laven, M. (2022) · BMC Health Services Research · doi:10.1186/s12913-022-08070-7 · Cited by 87 · cited 1×
“The use of AI in emergency rooms has already been shown to improve clinical decision-making and reduce physician workload [22].”
Background
Cited by
ChatGPT in healthcare: A taxonomy and systematic review
Li, J., Dada, A., Puladi, B., Kleesiek, J., & Egger, J. (2024) · Computer Methods and Programs in Biomedicine · doi:10.1016/j.cmpb.2024.108013 · Cited by 452
Takita, H., Kabata, D., Walston, S. L., Tatekawa, H., Saito, K., Tsujimoto, Y., Miki, Y., & Ueda, D. (2025) · npj Digital Medicine · doi:10.1038/s41746-025-01543-z · Cited by 203
Omar, M., Sorin, V., Agbareia, R., Apakama, D. U., Soroush, A., Sakhuja, A., Freeman, R., Horowitz, C. R., Richardson, L. D., Nadkarni, G. N., et al. (2025) · International Journal for Equity in Health · doi:10.1186/s12939-025-02419-0 · Cited by 110
Wan, P., Huang, Z., Tang, W., Nie, Y., Pei, D., Deng, S., Chen, J., Zhou, Y., Duan, H., Chen, Q., et al. (2024) · Nature Medicine · doi:10.1038/s41591-024-03148-7 · Cited by 97
Chow, J. C. L., & Li, K. (2025) · Information · doi:10.3390/info16070549 · Cited by 84
Metadata
- Authors
- Ito, N., Kadomatsu, S., Fujisawa, M., Fukaguchi, K., Ishizawa, R., Kanda, N., Kasugai, D., Nakajima, M., Goto, T., & Tsugawa, Y.
- Venue
- JMIR Medical Education
- Publisher
- JMIR Publications Inc.
- Published
- 2023-11-02
- Volume
- 9
- Pages
- e47532
- ISSN
- 2369-3762 (ISSN-L)
- OpenAlex
- W4388237833
- Cited by
- 63
- References
- 13
- FWCI
- 2.18
- Type
- journal article
- Language
- en
- Open access
- gold
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
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.