ArticleIndian dermatology online journal2025
A Meta-Analysis of ChatGPT's Performance on Dermatology Specialty-Level (Board-Style) Certification Questions.
Article in Indian dermatology online journal, 2025. The graph could read no effect estimate from its abstract, so it casts no vote on the map. Not yet cited in PubMed.
What it found
Each row is one number read from the abstract, on the scale the paper reported it, with its interval. Left of the dashed line favours the treatment, right favours the comparator. Under each row is the sentence it came from. New to these charts? A ten-minute tutorial.
The abstract states no effect estimate the extractor could read, or names no intervention and outcome on the map, so this paper lights no cell and moves no belief. It is still indexed, cited and linked below.
The trial behind it
Trials whose registry record cites this paper, or whose number appears in the abstract. A trial that started after this paper was published is citing it as background, not reporting it.
Neither the registry nor the abstract names a trial number. If this is a trial report, that itself is worth knowing.
Who cites it
0 citing papers in PubMed.
No citing paper in PubMed yet.
Corrections and comments
PubMed lists nothing against this paper. Absence here is not a guarantee, only a check that was made.
Authors and funding
1 author.
Funding
No grant is acknowledged in the PubMed record.
Abstract
BACKGROUND AND
objectiveChatGPT, an advanced artificial intelligence chatbot, has shown promise in various medical applications, including dermatology. This meta-analysis evaluates its performance on dermatology specialty (board-style) certification questions based on evidence from multiple studies. MATERIALS AND
methodsWe searched PubMed/MEDLINE and EMBASE for studies published up to December 10, 2024, using predefined search terms or keywords. A random-effects model estimated ChatGPT's overall performance on dermatology specialty certification (board-style) questions, with 95% confidence intervals and subgroup analyses by ChatGPT version (GPT-3.5 or GPT-4.0). Statistical analysis used STATA 17, with heterogeneity assessed by the I ² statistic, and significance set at P < 0.05.
resultsA total of 13 studies were included in this meta-analysis. The overall pooled accuracy of ChatGPT, across both GPT-3.5 and GPT-4.0 versions, was 69.97% (95% CI: 64.64-75.05%, I ² = 93.28%, P value = 0.00). ChatGPT-4.0 showed the highest overall accuracy of 79.46% (95% CI: 75.36-83.30%, I ² = 72.79%, P value = 0.00) compared to ChatGPT-3.5, which showed an overall accuracy of 61.07% (95% CI: 57.36-64.72%, I ² = 72.97%, P value = 0.00). LIMITATIONS: This study includes only English-language publications and focuses solely on ChatGPT's performance in dermatology exams, limiting its generalizability for clinical or patient education use. Significant heterogeneity warrants further research into potential causes such as question difficulty, dermatology subdomains, and question format and layout.
conclusionNotwithstanding its limitations, this meta-analysis highlights ChatGPT's potential to serve as an additional educational resource for candidates preparing to sit dermatology specialty (board-level) certification examinations and for supporting their overall educational development.
Indexed as
Identifiers
What OpenQuestion holds
Registered trials
Read under generation 80e0d062 · epoch 390. Bibliography from PubMed, PubMed Central and OpenAlex; grants from NIH RePORTER; trial links from ClinicalTrials.gov; estimates, votes and beliefs from the OpenQuestion graph.