Pharmacovigilance is the system for watching a medication's safety after it is approved. Clinical trials enroll a few thousand carefully selected people for a limited time, so rare, delayed, or population-specific side effects only surface once millions of people are taking a drug. Regulators (the FDA, the European Medicines Agency), manufacturers, and academic researchers collect adverse-event reports, compare them against expected rates, and decide whether a label change, warning, or further study is needed.
Many GLP-1 safety questions (suicidal ideation, gastroparesis, hair loss, a rare eye condition called NAION) have been raised or investigated this way. One common tool is the "reporting odds ratio," which compares how often a side effect is reported with one drug versus all other drugs in the database. A high ratio flags a signal worth investigating, but it does not by itself show the drug caused the effect, because reporting is uneven. In the FDA's database (called FAERS until March 2026, when it became part of the FDA Adverse Event Monitoring System, or AEMS), reporting is voluntary for patients and clinicians, while manufacturers are required to forward the reports they receive (FDA AEMS dashboard FAQ (external link); FDA, March 2026 (external link)). Follow-up usually means large cohort studies or trial re-analyses. That is how the European Medicines Agency's safety committee concluded in April 2024 that the available evidence did not support a causal link between GLP-1s and suicidal thoughts (EMA PRAC, April 2024 (external link)). See mood changes and anhedonia on a GLP-1.