An analysis of over 22,000 FDA adverse event reports found semaglutide has the strongest signal for metabolic side effects among GLP-1 drugs, with dehydration being the most common serious problem across all agents.
Dehydration: 25–33% of serious metabolic AEsAcross semaglutide, tirzepatide, liraglutide, and dulaglutide, dehydration was the single most common serious metabolic side effect reported to the FDA
What the researchers found
Mining the FDA's adverse event database (FAERS) revealed that semaglutide, liraglutide, and exenatide have the highest rates of metabolic and nutritional adverse events among GLP-1 drugs. Semaglutide had the strongest signal (ROR 3.34), followed by liraglutide (ROR 2.78) and exenatide (ROR 2.15).
Dehydration emerged as the most common serious metabolic side effect across multiple GLP-1 drugs: it accounted for 25.1% of serious metabolic adverse events for semaglutide, 32.9% for tirzepatide, 23.9% for liraglutide, and 20.9% for dulaglutide. Dulaglutide had the most total adverse event signal types (22), followed by semaglutide (20) and liraglutide (16).
Why it matters
With tens of millions of people now taking GLP-1 drugs, understanding their real-world side effect patterns is critical. This study highlights dehydration as a particularly important — and potentially underrecognized — risk across all major GLP-1 drugs. Dehydration can lead to kidney injury, electrolyte imbalances, and hospitalization, especially in elderly patients or those on diuretics.
The numbers in context
22,404+ total reports analyzed · semaglutide ROR 3.34 · liraglutide ROR 2.78 · exenatide ROR 2.15 · dehydration: 25-33% of serious metabolic AEs across drugs · 7 GLP-1 RAs compared
How the study worked
Pharmacovigilance analysis using the FDA Adverse Event Reporting System (FAERS) database. Researchers extracted all metabolic and nutritional adverse event reports for seven GLP-1 receptor agonists from their respective launch dates through Q2 2023. Four statistical methods were used to identify safety signals: reported odds ratio (ROR), proportional reporting ratio (PRR), Empirical Bayesian Geometric Mean, and Bayesian Confidence Propagation Neural Network. Time-to-onset analysis was also performed.
Who was studied
All metabolic/nutritional adverse event reports for 7 GLP-1 receptor agonists in the FDA FAERS database through Q2 2023
What this study cannot tell us
FAERS is a voluntary reporting system with inherent biases — popular drugs get more reports, and reporting rates vary over time. The data can't establish causation, only associations. Media attention on semaglutide and tirzepatide may inflate their report counts. No adjustment for patient demographics, comorbidities, or concomitant medications. Under-reporting is a known limitation of all pharmacovigilance databases.
How to read the evidence
Pharmacovigilance studies using FAERS provide important real-world safety signals but cannot prove causation. The voluntary reporting system has known biases including under-reporting and reporting influenced by media attention. The use of four statistical methods strengthens signal detection, but the evidence is observational.
When this study was published
Published in 2024 with data through Q2 2023. This is relatively current but predates the massive expansion of tirzepatide prescriptions and new agents entering the market. Updated analyses will likely show different patterns as prescribing volumes change.
The bigger picture
As GLP-1 drugs expand from diabetes into obesity, MASH, cardiovascular disease, and addiction, their safety profile becomes even more important. This is the largest pharmacovigilance comparison of metabolic side effects across GLP-1 drugs to date. The dehydration finding is particularly relevant — patients on these drugs often eat and drink less, and the combination of reduced fluid intake with the drugs' gastrointestinal effects could create a dangerous dehydration risk that needs proactive management.
Questions still open
- Is the higher metabolic adverse event signal for semaglutide a true pharmacological difference or driven by its greater market share and media attention?
- Should patients on GLP-1 drugs be routinely counseled about hydration and monitored for kidney function?
- Do the metabolic side effect profiles differ between diabetes-dose and obesity-dose GLP-1 prescriptions?
Common questions
Does this mean semaglutide is less safe than other GLP-1 drugs?
Why is dehydration such a common problem with GLP-1 drugs?
Read the original research
Pharmacovigilance study of GLP-1 receptor agonists for metabolic and nutritional adverse events.
Frontiers in pharmacology, 15, 1416985
Citation
He, Long; Li, Qiuyu; Yang, Yongfeng; Li, Jiahao; Luo, Wei; Huang, Yilan; Zhong, Xiaoyan. (2024). Pharmacovigilance study of GLP-1 receptor agonists for metabolic and nutritional adverse events.. Frontiers in pharmacology, 15, 1416985. https://doi.org/10.3389/fphar.2024.1416985