Glucagon-like peptide-1 receptor agonists (GLP-1 RAs) consistently increased GI adverse events but showed only limited evidence of benefits beyond their established cardiometabolic effects, according to a large umbrella review of meta-analyses of randomized clinical trials. Although GLP-1 receptor agonists were associated with lower odds of serious infections and respiratory disease, the investigators concluded that evidence for most noncardiometabolic outcomes remains too uncertain to change clinical practice.
“Most previous discussions of GLP-1 receptor agonists have focused on their effects on glucose control, weight loss, and cardiovascular or kidney protection,” corresponding author Yongze Li, PhD, of the department of endocrinology and metabolism at The First Hospital of China Medical University, Shenyang, China, told GI & Hepatology News. “Our study took a broader view by systematically evaluating more than 100 noncardiometabolic outcomes across multiple organ systems. We found that while GI adverse events remain the most consistent and well-supported safety signals, evidence for many other reported associations is considerably less certain than often assumed. At the same time, we identified encouraging signals suggesting lower risks of serious infections and respiratory diseases, although these findings require further confirmation.”
The review, published in JAMA Network Open, combined evidence from 60 meta-analyses that included 1,751 randomized clinical trials involving more than 3.5 million participants. Investigators searched PubMed, Embase, Web of Science, Scopus, and the Cochrane Database of Systematic Reviews through Jan. 15, 2026, to examine the effects of GLP-1 RAs beyond blood sugar control, weight loss, and their established cardiovascular and kidney benefits. They extracted data at the meta-analysis and individual-study levels and reanalyzed the evidence using random-effects models to estimate odds ratios and 95% confidence intervals.
Most of the included meta-analyses involved patients with type 2 diabetes, and about one third included patients with obesity. Follow-up ranged from three months to more than five years. Overall, the review examined 116 outcomes beyond heart, kidney, and metabolic health, including GI, neurologic, respiratory, psychiatric, liver, endocrine, fracture, cancer, and other safety outcomes.
The clearest and most consistent findings involved GI adverse effects. Compared with controls, patients receiving GLP-1 RAs had about 2.5 times the odds of nausea, nearly 2.8 times the odds of vomiting, and about twice the odds of diarrhea. The evidence linking the drugs to nausea and diarrhea was rated high quality, while the evidence for vomiting was rated moderate quality. However, the investigators noted considerable differences among studies, meaning the exact size of these risks remains uncertain.
The review found high-quality evidence for an association between GLP-1 receptor agonist use and lower odds of serious infections. Patients receiving these drugs had 11% lower odds of serious infections than controls, supported by high-quality evidence. The investigators also found a possible 15% lower odds of respiratory disease, but considered that finding preliminary because it was based on a relatively small number of studies.
The investigators also found several possible benefits that did not meet their standards for high-confidence evidence. These included lower odds of fractures and all-cause dementia, along with more than threefold higher odds of resolving metabolic dysfunction-associated steatohepatitis. The authors cautioned that these findings should be viewed as exploratory because they were based on limited evidence or relatively few studies.
The review also examined several long-standing safety concerns. It found possible increases in the odds of GERD and gallbladder or biliary disease among patients receiving GLP-1 RAs. Evidence linking the drugs to pancreatitis also remained inconclusive. The authors said conflicting findings from recent observational studies and evolving regulatory safety warnings highlight the need for larger studies with longer follow-up to better define these potential risks.
Cancer outcomes remained uncertain. Colorectal cancer showed a nominal signal toward higher odds, thyroid cancer estimates were imprecise, and pancreatic cancer showed a nominal signal toward lower odds, but none met the investigators’ credibility thresholds. Analyses of rare cancer events also showed that many of these apparent associations disappeared when a single trial was removed, suggesting the findings were not robust.
Subgroup analyses showed that semaglutide and liraglutide were the most extensively studied GLP-1 RAs. However, analyses of tirzepatide, treatment duration, and dose were based on relatively few studies, so the investigators considered those findings exploratory.
The study was funded by the National Natural Science Foundation of China. The authors reported no conflicts of interest.
Expert Insight
GI & Hepatology News asked Dr. Li to put the review’s findings into clinical perspective.
Looking at the evidence as a whole, what did you find most interesting or unexpected?
Dr. Li: One finding that stood out was how few noncardiometabolic associations were supported by high-certainty evidence despite the enormous volume of published literature. We were also encouraged to see consistent signals suggesting reduced risks of serious infections and respiratory diseases. Conversely, several concerns that have received substantial attention, such as cancer-related outcomes, pancreatitis, or thyroid disorders, were generally supported by limited or low-certainty evidence after applying rigorous credibility assessments and sensitivity analyses.
How might the findings influence clinical practice?
Dr. Li: Our findings should reassure clinicians that GLP-1 RAs continue to have a favorable overall benefit-risk profile. At the same time, clinicians should remain attentive to common GI adverse effects, particularly during treatment initiation and dose escalation. Importantly, our review does not support changing prescribing practices based on many of the currently discussed noncardiometabolic safety concerns alone, as most of these signals remain exploratory and require stronger evidence before influencing routine clinical decision-making.
What are the key unanswered questions after this review?
Dr. Li: Several important gaps remain. First, most randomized trials were not originally designed to evaluate many of these noncardiometabolic outcomes, resulting in limited event numbers and relatively short follow-up. Second, more long-term randomized trials and large prospective studies are needed to clarify uncommon but clinically important outcomes, including pancreatitis, gallbladder disease, neurological disorders, and cancer. Finally, future studies should investigate whether these associations differ by specific GLP-1 RAs, dose, treatment duration, or patient population, since current evidence remains limited in these areas.
Is there anything else you’d like to say about this work?
Dr. Li: One important message from our study is that not every statistically significant finding should be interpreted as clinically established. As the use of GLP-1 RAs expands rapidly beyond diabetes into obesity and other chronic diseases, it becomes increasingly important to distinguish robust evidence from preliminary signals. We hope this review helps clinicians and researchers interpret the growing literature more critically and provides a stronger evidence base for future research.
I would also like to emphasize that our findings should not discourage the appropriate use of GLP-1 RAs. Rather, they support individualized clinical decision-making while highlighting areas where better evidence is still needed.