ChatGPT vs. web search for patient questions: what does ChatGPT do better?
Open accessThe DOI matches its Crossref record; 10 of the 10 references it cites were found in a registry or an index.
Abstract
Chat generative pretrained transformer (ChatGPT) has the potential to significantly impact how patients acquire medical information online. Here, we characterize the readability and appropriateness of ChatGPT responses to a range of patient questions compared to results from traditional web searches. Patient questions related to the published Clinical Practice Guidelines by the American Academy of Otolaryngology-Head and Neck Surgery were sourced from existing online posts. Questions were categorized using a modified Rothwell classification system into (1) fact, (2) policy, and (3) diagnosis and recommendations. These were queried using ChatGPT and traditional web search. All results were evaluated on readability (Flesch Reading Ease and Flesch-Kinkaid Grade Level) and understandability (Patient Education Materials Assessment Tool). Accuracy was assessed by two blinded clinical evaluators using a three-point ordinal scale. 54 questions were organized into fact (37.0%), policy (37.0%), and diagnosis (25.8%). The average readability for ChatGPT responses was lower than traditional web search (FRE: 42.3 ± 13.1 vs. 55.6 ± 10.5, p < 0.001), while the PEMAT understandability was equivalent (93.8% vs. 93.5%, p = 0.17). ChatGPT scored higher than web search for questions the ‘Diagnosis’ category (p < 0.01); there was no difference in questions categorized as ‘Fact’ (p = 0.15) or ‘Policy’ (p = 0.22). Additional prompting improved ChatGPT response readability (FRE 55.6 ± 13.6, p < 0.01). ChatGPT outperforms web search in answering patient questions related to symptom-based diagnoses and is equivalent in providing medical facts and established policy. Appropriate prompting can further improve readability while maintaining accuracy. Further patient education is needed to relay the benefits and limitations of this technology as a source of medial information.
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
Ayers, J. W., Poliak, A., Dredze, M., Leas, E. C., Zhu, Z., Kelley, J. B., Faix, D. J., Goodman, A. M., Longhurst, C. A., Hogarth, M., et al. (2023) · JAMA Internal Medicine · doi:10.1001/jamainternmed.2023.1838 · Cited by 2,613
Gilson, A., Safranek, C. W., Huang, T., Socrates, V., Chi, L., Taylor, R. A., & Chartash, D. (2023) · JMIR Medical Education · doi:10.2196/45312 · Cited by 2,078
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
Eysenbach, G. (2023) · JMIR Medical Education · doi:10.2196/46885 · Cited by 840 · cited 1×
“This functionality allows ChatGPT to provide answers at a wide range of education levels, which may have implications in increasing accessibility to medical information and reducing health care disparities [25].”
Johnson, D., Goodman, R., Patrinely, J., Stone, C., Zimmerman, E., Donald, R., Chang, S., Berkowitz, S., Finn, A., Jahangir, E., et al. (2023) · doi:10.21203/rs.3.rs-2566942/v1 · Cited by 598 · cited 1×
“The scoring was completed using an ordinal three-point scale [18].”
Xu, L., Sanders, L., Li, K., & Chow, J. C. L. (2021) · JMIR Cancer · doi:10.2196/27850 · Cited by 571 · cited 1×
“Similar advantages in other AI conversational agents have previously been reported [26, 27]; however, ChatGPT represents a significant advancement over prior iterations.”
Finney Rutten, L. J., Blake, K. D., Greenberg-Worisek, A. J., Allen, S. V., Moser, R. P., & Hesse, B. W. (2019) · Public Health Reports® · doi:10.1177/0033354919874074 · Cited by 426
Samaan, J. S., Yeo, Y. H., Rajeev, N., Hawley, L., Abel, S., Ng, W. H., Srinivasan, N., Park, J., Burch, M., Watson, R., et al. (2023) · Obesity Surgery · doi:10.1007/s11695-023-06603-5 · Cited by 282 · cited 1×
“It can provide empathetic responses to patient questions [9] and answer queries within a range of surgical subspecialties [10, 11].”
Artificial intelligence and chatbots in psychiatry
Pham, K. T., Nabizadeh, A., & Selek, S. (2022) · Psychiatric Quarterly · doi:10.1007/s11126-022-09973-8 · Cited by 241
Lessons for artificial intelligence from the study of natural stupidity
Rich, A. S., & Gureckis, T. M. (2019) · Nature Machine Intelligence · doi:10.1038/s42256-019-0038-z · Cited by 94
Cited by
Large language models in patient education: a scoping review of applications in medicine
Aydin, S., Karabacak, M., Vlachos, V., & Margetis, K. (2024) · Frontiers in Medicine · doi:10.3389/fmed.2024.1477898 · Cited by 288
Large Language Models for Chatbot Health Advice Studies: A Systematic Review
Huo, B., Boyle, A., Marfo, N., Tangamornsuksan, W., Steen, J. P., McKechnie, T., Lee, Y., Mayol, J., Antoniou, S. A., Thirunavukarasu, A. J., et al. (2025) · JAMA Network Open · doi:10.1001/jamanetworkopen.2024.57879 · Cited by 283
Akdogan, O., Uyar, G. C., Yesilbas, E., Baskurt, K., Malkoc, N. A., Ozdemir, N., Yazici, O., Oksuzoglu, B., Uner, A., Ozet, A., et al. (2025) · European Journal of Cancer · doi:10.1016/j.ejca.2025.115408 · Cited by 45
Mondal, H., Tiu, D. N., Mondal, S., Dutta, R., Naskar, A., & Podder, I. (2025) · Journal of Mid-life Health · doi:10.4103/jmh.jmh_182_24 · Cited by 6
Metadata
- Authors
- Shen, S. A., Perez-Heydrich, C. A., Xie, D. X., & Nellis, J. C.
- Venue
- European Archives of Oto-Rhino-Laryngology
- Published
- 2024-02-28
- Volume
- 281
- Issue
- 6
- Pages
- 3219-3225
- ISSN
- 0937-4477 (ISSN-L)
- 1434-4726
- OpenAlex
- W4392234111
- Cited by
- 52
- References
- 10
- FWCI
- 4.17
- Type
- journal article
- Language
- en
- Open access
- green
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.