LLM analyzed 150k Reddit posts to identify side effects of unapproved new GLP-1 (www.engineering.upenn.edu)

Researchers at the University of Pennsylvania utilized AI to analyze nearly 150,000 Reddit posts discussing retatrutide, an unapproved GLP-1 weight-loss drug, revealing potential side effects reported by users. This innovative approach, led by doctoral student Neil Sehgal, allows for the identification of self-reported symptoms in a population often overlooked in traditional drug-safety monitoring. Among the findings, users frequently reported increased appetite—notably different from the predominant gastrointestinal symptoms observed in clinical trials. This discrepancy raises questions about the drug's different effects as it activates three hormone receptors, unlike existing GLP-1 medications. The study highlights the significance of AI in systematically uncovering safety signals in real-time discussions on social media, providing a valuable resource for public health researchers to explore potential concerns further. The researchers emphasize that these findings should be approached cautiously; they do not confirm a direct cause-and-effect relationship. Instead, they serve as a foundation for future studies aimed at investigating these unexpected symptoms, potentially informing regulatory oversight and patient safety as discussions around unconventional drug sources become more prevalent.