EUVIMEDOntdek EUVIMED
Systematische ÜbersichtMetaanalyseOpen access

Evaluating the Potential and Accuracy of ChatGPT-3.5 and 4.0 in Medical Licensing and In-Training Examinations: Systematic Review and Meta-Analysis

Evaluierung des Potenzials und der Genauigkeit von ChatGPT-3.5 und 4.0 in medizinischen Zulassungs- und In-Training-Prüfungen: systematische Übersichtsarbeit und Metaanalyse

Anila Jaleel iD, Umair Aziz iD, Ghulam Farid iD, Muhammad Zahid Bashir iD, Tehmasp Rehman Mirza iD, Syed Mohammad Khizar Abbas iD, Shiraz Aslam iD, Rana Muhammad Hassaan Sikander iD

JMIR Medical Education · 2025-09-19 · 11 · e68070

DOI: 10.2196/68070 ↗

Auf einen Blick

Die Metaanalyse von 53 Studien zeigt: ChatGPT-4 schnitt in medizinischen Prüfungen besser ab als ChatGPT-3.5. Die Leistung variierte je nach Prüfungstyp.

Fragestellung

Wie genau sind ChatGPT-3.5 und ChatGPT-4.0 in medizinischen Zulassungs- und In-Training-Prüfungen?

Methode

Systematische Übersichtsarbeit und Metaanalyse nach PRISMA; Datenbanken Scopus, PubMed, JMIR Publications, Elsevier, BMJ, Wiley Online Library; Studien von Januar 2023 bis Juli 2024.

Ergebnisse

53 Studien. In medizinischen Zulassungsprüfungen: gepoolte Genauigkeit 81,8 % für ChatGPT-4 vs. 60,8 % für ChatGPT-3.5. In Residency-Prüfungen: 72,2 % vs. 57,7 %. RR 1,36 (95 %-KI 1,30–1,43).

Schlussfolgerung der Autor:innen

ChatGPT-4.0 und 3.5 zeigen laut Autor:innen vielversprechende Ergebnisse für die medizinische Ausbildung und die klinische Entscheidungsfindung, können aber die erforderlichen umfassenden Fähigkeiten der medizinischen Praxis nicht ersetzen.

Kernaussagen

  • ChatGPT-4 erzielte höhere Genauigkeit als ChatGPT-3.5
  • RR 1,36 (95 %-KI 1,30–1,43)
  • Leistung war je nach Prüfungstyp unterschiedlich

Deutsche Zusammenfassung: KI-gestützt aus dem Original-Abstract erstellt; alle Zahlen wurden automatisch mit dem Original abgeglichen. Maßgeblich ist der Original-Abstract.

At a glance

Systematische Übersicht in JMIR Medical Education, published online on 2025-09-19. Written by Anila Jaleel and 7 co-authors. Institution: Shalamar Medical and Dental College, Lahore, Pakistan. Major MEDLINE topics: Licensure, Medical, Educational Measurement, Artificial Intelligence, Education, Medical. Cited 31 times so far (Crossref); 90 references. The full text is freely available in PubMed Central.

Auf einen BlickSamenvattingSubjectsBibliografische gegevensCiteRelated studies

Abstract (original language)

Background

Artificial intelligence (AI) has significantly impacted health care, medicine, and radiology, offering personalized treatment plans, simplified workflows, and informed clinical decisions. ChatGPT (OpenAI), a conversational AI model, has revolutionized health care and medical education by simulating clinical scenarios and improving communication skills. However, inconsistent performance across medical licensing examinations and variability between countries and specialties highlight the need for further research on contextual factors influencing AI accuracy and exploring its potential to enhance technical proficiency and soft skills, making AI a reliable tool in patient care and medical education.

Objective

This systematic review aims to evaluate and compare the accuracy and potential of ChatGPT-3.5 and 4.0 in medical licensing and in-training residency examinations across various countries and specialties.

Methods

A systematic review and meta-analysis were conducted, adhering to the PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) guidelines. Data were collected from multiple reputable databases (Scopus, PubMed, JMIR Publications, Elsevier, BMJ, and Wiley Online Library), focusing on studies published from January 2023 to July 2024. Analysis specifically targeted research assessing ChatGPT’s efficacy in medical licensing exams, excluding studies not related to this focus or published in languages other than English. Ultimately, 53 studies were included, providing a robust dataset for comparing the accuracy rates of ChatGPT-3.5 and 4.0.

Results

ChatGPT-4 outperformed ChatGPT-3.5 in medical licensing exams, achieving a pooled accuracy of 81.8%, compared to ChatGPT-3.5’s 60.8%. In in-training residency exams, ChatGPT-4 achieved an accuracy rate of 72.2%, compared to 57.7% for ChatGPT-3.5. The forest plot presented a risk ratio of 1.36 (95% CI 1.30-1.43), demonstrating that ChatGPT-4 was 36% more likely to provide correct answers than ChatGPT-3.5 across both medical licensing and residency exams. These results indicate that ChatGPT-4 significantly outperforms ChatGPT-3.5, but the performance advantage varies depending on the exam type. This highlights the importance of targeted improvements and further research to optimize ChatGPT-4’s performance in specific educational and clinical settings.

Conclusions

ChatGPT-4.0 and 3.5 show promising results in enhancing medical education and supporting clinical decision-making, but they cannot replace the comprehensive skill set required for effective medical practice. Future research should focus on improving AI’s capabilities in interpreting complex clinical data and enhancing its reliability as an educational resource.

Subjects

MeSH terms (MEDLINE)

Licensure, MedicalEducational MeasurementArtificial IntelligenceEducation, MedicalHumansInternship and ResidencyClinical CompetenceGenerative Artificial Intelligence

Author keywords

AI in health careChatGPTChatGPT-3.5 performanceChatGPT-4.0 performanceaccuracy of ChatGPTartificial intelligenceclinical decision-makingmedical educationmedical licensing examinations

Authors and institutions

  • Shalamar Medical and Dental College, Lahore, Pakistan.

ORCID profiles: Anila Jaleel, Umair Aziz, Ghulam Farid, Muhammad Zahid Bashir, Tehmasp Rehman Mirza, Syed Mohammad Khizar Abbas, Shiraz Aslam, Rana Muhammad Hassaan Sikander

Cite

APA

Jaleel, A., Aziz, U., Farid, G., Zahid Bashir, M., Mirza, T. R., Khizar Abbas, S. M., Aslam, S., & Sikander, R. M. H. (2025). Evaluating the Potential and Accuracy of ChatGPT-3.5 and 4.0 in Medical Licensing and In-Training Examinations: Systematic Review and Meta-Analysis. JMIR Medical Education, 11, e68070. https://doi.org/10.2196/68070

Vancouver

Jaleel A, Aziz U, Farid G, Zahid Bashir M, Mirza TR, Khizar Abbas SM, et al. Evaluating the Potential and Accuracy of ChatGPT-3.5 and 4.0 in Medical Licensing and In-Training Examinations: Systematic Review and Meta-Analysis. JMIR Medical Education. 2025;11:e68070. doi:10.2196/68070

BibTeX
@article{Jaleel2025,
  author = {Jaleel, Anila and Aziz, Umair and Farid, Ghulam and Zahid Bashir, Muhammad and Mirza, Tehmasp Rehman and Khizar Abbas, Syed Mohammad and Aslam, Shiraz and Sikander, Rana Muhammad Hassaan},
  title = {Evaluating the Potential and Accuracy of ChatGPT-3.5 and 4.0 in Medical Licensing and In-Training Examinations: Systematic Review and Meta-Analysis},
  journal = {JMIR Medical Education},
  year = {2025},
  volume = {11},
  pages = {e68070},
  doi = {10.2196/68070}
}
Start for free →