A new umbrella review has shed light on how Artificial Intelligence (AI) tools, particularly Chat Generative Pre-Trained Transformer (ChatGPT), are transforming healthcare across the globe. The findings highlight both the promise and challenges of integrating large language models (LLMs) into clinical settings.
Key Findings and Study Quality
Conducted in line with PRISMA guidelines, the review examined research drawn from major databases including PubMed, Scopus, and the Cochrane Library. Covering studies from inception until February 2024, the researchers analysed 17 reviews, 15 systematic and 2 meta-analyses, focusing on ChatGPT’s role in medicine. The reviewers evaluated each study’s methodological quality using the AMSTAR-2 checklist, uncovering that five reviews were rated as moderate quality while twelve were considered low. Many of the lower-rated studies lacked justification of study design and transparency over funding sources.
ChatGPT’s Expanding Role in Medicine
The review found that most research, 82.4%, examined ChatGPT’s general applications in healthcare, while 17.6% investigated specialised areas such as its use in medical examinations and systematic reviews. Of these studies, 52.9% addressed general healthcare contexts, and 41.2% focused on particular fields including radiology, neurosurgery, gastroenterology, public health dentistry, and ophthalmology.
The review underscored that while the technology shows great potential for improving diagnostic accuracy, enhancing decision-making, and streamlining administrative workloads, it also raises serious ethical and legal questions.
Ethical Concerns and the Path Forward
Researchers stressed the need for careful implementation, improved reliability testing, and the establishment of robust ethical frameworks before widespread clinical adoption. Concerns around data bias, misinformation, and accountability remain at the forefront of policy discussions.
The review concluded that ChatGPT’s role in healthcare could be transformative if deployed with caution and proper regulatory oversight. Policymakers and medical educators are urged to play a central role in guiding AI’s integration into patient care, ensuring that innovation is matched by safety, transparency, and public trust.
Reference
Iqbal U et al. Impact of large language model (ChatGPT) in healthcare: an umbrella review and evidence synthesis. J Biomed Sci. 2025;32(1):45.
