BACKGROUND
Since its public release on 30 November 2022, ChatGPT has shown a promising potential in diverse healthcare applications, despite challenges related to ethics, privacy issues, and possible biases. Bibliometric analysis is a useful tool to assess the themes and recommendations of the most cited publications in the field of ChatGPT utility in healthcare, a newly emerging research topic.
OBJECTIVE
To identify the top ten healthcare related ChatGPT publications in different scientific databases and to assess the future trajectory of needed research based on the recommendations of these influential papers.
METHODS
The study employed an advanced search on three databases: Scopus, Web of Science, and Google Scholar to identify ChatGPT-related records in healthcare education, research, and practice between 27 November and 30 November 2023. Ranking was based on the retrieved citation count in each database. Other metrics evaluated included Semantic Scholar highly influential citations, PlumX metrics, and Altmetric Attention Scores (AASs).
RESULTS
A total of 22 unique records published in 17 different scientific journals from 14 different publishers were identified in the three databases. Only two publications were found in the top 10 list across the three databases. Variable publication types were identified with the most common being editorial/commentary (n=8/22, 36.4%). Nine out of the 22 records had corresponding authors affiliated to institutions in the United States (40.9%). The range of citation count varied per database with the highest in Google Scholar (1019–121), Scopus (242–88), and Web of Science (171–23). Significant positive correlations were detected between the Google Scholar citations and Semantic Scholar highly influential citations (Spearman’s correlation coefficient (ρ)=.840, P<.001), PlumX captures (ρ=.831, P<.001), PlumX mentions (ρ=.609, P=.004), and AASs (ρ=.542, P=.009). The highly cited papers recommended exploring ChatGPT use in healthcare education, streamlining clinical workflow, investigation of its ethical use, usefulness in medical writing, and mitigating misinformation risks.
CONCLUSIONS
Despite the several acknowledged limitations, bibliometric analysis in this study showed the evolving landscape of ChatGPT utility in healthcare and highlighted the future trajectories of research in this swiftly evolving research subject. This included the need for more rigorous studies to evaluate the performance of ChatGPT in medical education, as an aid in medical research, and its ability to enhance the clinical workflow. There is an urgent need for collaborative initiatives by all stakeholders involved to establish guidelines for ethical, transparent, and responsible use of ChatGPT in healthcare.