CoChat

How Does ChatGPT Perform on the United States Medical Licensing Examination (USMLE)? The Implications of Large Language Models for Medical Education and Knowledge Assessment

Open access

The DOI matches its Crossref record; 26 of the 26 references it cites were found in a registry or an index.

Abstract

Background: Chat Generative Pre-trained Transformer (ChatGPT) is a 175-billion-parameter natural language processing model that can generate conversation-style responses to user input. Objective: This study aimed to evaluate the performance of ChatGPT on questions within the scope of the United States Medical Licensing Examination (USMLE) Step 1 and Step 2 exams, as well as to analyze responses for user interpretability. Methods: We used 2 sets of multiple-choice questions to evaluate ChatGPT’s performance, each with questions pertaining to Step 1 and Step 2. The first set was derived from AMBOSS, a commonly used question bank for medical students, which also provides statistics on question difficulty and the performance on an exam relative to the user base. The second set was the National Board of Medical Examiners (NBME) free 120 questions. ChatGPT’s performance was compared to 2 other large language models, GPT-3 and InstructGPT. The text output of each ChatGPT response was evaluated across 3 qualitative metrics: logical justification of the answer selected, presence of information internal to the question, and presence of information external to the question. Results: Of the 4 data sets, AMBOSS-Step1, AMBOSS-Step2, NBME-Free-Step1, and NBME-Free-Step2, ChatGPT achieved accuracies of 44% (44/100), 42% (42/100), 64.4% (56/87), and 57.8% (59/102), respectively. ChatGPT outperformed InstructGPT by 8.15% on average across all data sets, and GPT-3 performed similarly to random chance. The model demonstrated a significant decrease in performance as question difficulty increased (P=.01) within the AMBOSS-Step1 data set. We found that logical justification for ChatGPT’s answer selection was present in 100% of outputs of the NBME data sets. Internal information to the question was present in 96.8% (183/189) of all questions. The presence of information external to the question was 44.5% and 27% lower for incorrect answers relative to correct answers on the NBME-Free-Step1 (P<.001) and NBME-Free-Step2 (P=.001) data sets, respectively. Conclusions: ChatGPT marks a significant improvement in natural language processing models on the tasks of medical question answering. By performing at a greater than 60% threshold on the NBME-Free-Step-1 data set, we show that the model achieves the equivalent of a passing score for a third-year medical student. Additionally, we highlight ChatGPT’s capacity to provide logic and informational context across the majority of answers. These facts taken together make a compelling case for the potential applications of ChatGPT as an interactive medical education tool to support learning.

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.

A few free checks a day · papers already verified open instantly · results are public

Integrity signals

Registry

Crossref · 10.2196/45312

Cites

26 of 26 references verified0 citation errors · 0 not checkable

Statistics

Figures

Figure 1 Template of question posed to each large language model (LLM), including both AMBOSS Attending Tip and the response from Chat Generative Pre-trained Transformer (ChatGPT). The correct answer to this question is “E. Zidovudine (AZT).” In the case of GPT-3, prompt engineering was necessary, with: "Please answer this multiple choice question:" + question as described previously + "Correct answer is." As GPT-3 is inherently a nondialogic model, this was necessary to reduce model hallucinations and force a clear answer [17].
Figure 1 Template of question posed to each large language model (LLM), including both AMBOSS Attending Tip and the response from Chat Generative Pre-trained Transformer (ChatGPT). The correct answer to this question is “E. Zidovudine (AZT).” In the case of GPT-3, prompt engineering was necessary, with: "Please answer this multiple choice question:" + question as described previously + "Correct answer is." As GPT-3 is inherently a nondialogic model, this was necessary to reduce model hallucinations and force a clear answer [17].CC BY · PubMed Central

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.

A few free checks a day · papers already verified open instantly · results are public

Claims other papers make about this one

Each is checked against this paper’s own text: does it say what the citing paper says it does?

39 more claims about this paper

Create a free account to see every claim other papers make about this one.

See all claims — free account

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

Show all 26 references

Research works printed but not linked: 2

  1. 10.48550/arxiv.2108.01764 · DOI
  2. Wang, DC, 2020, Yale J Biol Med, 93, 441

Cited by

Items 101–200 of 1,047

Metadata

Authors
Gilson, A., Safranek, C. W., Huang, T., Socrates, V., Chi, L., Taylor, R. A., & Chartash, D.
Affiliations
  • Yale University (US)
  • University College Dublin (IE)
Corresponding author
Chartash, David
Venue
JMIR Medical Education
Published
2023-02-08
Volume
9
Pages
e45312
ISSN
2369-3762 (ISSN-L)
OpenAlex
W4319460874
Cited by
2,078
References
21
Citations by year
2026: 351 · 2025: 550 · 2024: 699 · 2023: 474
FWCI
82.58
Type
journal article
Language
en
Open access
Licence
cc-by
Locations
  • JMIR Medical Education (publishedVersion, cc-by)
  • PubMed Central (publishedVersion, cc-by)
  • PubMed (publishedVersion)
  • DOAJ (DOAJ: Directory of Open Access Journals) (submittedVersion, cc-by-sa)
Funders
  • Yale University
  • National Institutes of Health (UL1TR001863)
  • National Heart, Lung, and Blood Institute (T35HL007649)
  • National Institute of Diabetes and Digestive and Kidney Diseases (T35DK104689)
  • National Center for Advancing Translational Sciences
Topics
  • Health Sciences › Medicine › Health Informatics › Artificial Intelligence in Healthcare and Education
  • Social Sciences › Education › Innovations in Medical Education
  • Social Sciences › Education › Artificial Intelligence in Education
Keywords
ChatGPT, large language models, USMLE Step 1, USMLE Step 2, medical licensing examinations, question difficulty, GPT-3, medical education, knowledge assessment, AMBOSS, National Board of Medical Examiners, artificial intelligence in medical education

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.