McCall · Expert opinion on drug safety 2026 · retrospective pharmacovigilance disproportionality study · n=81,078 reports

Safety analysis of compounded GLP-1 receptor agonists: a pharmacovigilance study using the FDA adverse event reporting system.

Cited 20 times in the scientific literature.

Level 4 - case-series / case-control

Retrospective pharmacovigilance disproportionality analysis of spontaneous adverse event reports (FAERS)

PubMed 40285721 · doi:10.1080/14740338.2025.2499670 · record verified 2026-08-26

What was done

Researchers conducted a retrospective pharmacovigilance analysis of the FDA Adverse Event Reporting System (FAERS) database from 2018 to 2024. They compared adverse event reports, medication errors, and product quality issues for compounded versus non-compounded formulations of liraglutide, semaglutide, and tirzepatide. Reporting odds ratios (RORs) with 95% confidence intervals (CIs) were calculated using adjusted logistic regression.

What was found

Of 81,078 GLP-1 receptor agonist reports identified, 707 involved compounded products. Compounded formulations demonstrated significantly higher reporting odds for abdominal pain (ROR 2.84 [95% CI 2.29, 3.49]), diarrhea (1.59 [1.25, 1.99]), nausea (1.27 [1.05, 1.52]), suicidality (6.34 [4.32, 8.99]), cholecystitis (3.39 [1.61, 6.31]), and hospitalization (2.35 [1.94, 2.83]). Compounded products also showed higher reporting odds for preparation errors (48.92 [12.63, 189.6]), prescribing errors (4.46 [2.49, 7.98]), contamination (19.00 [4.24, 85.03]), and compounding/manufacturing issues (8.51 [5.17, 14.0]), but lower reporting odds of administration errors (0.29 [0.16, 0.53]) and dosing errors (0.24 [0.17, 0.32]).

Why it matters

As demand and shortages drive the use of compounded alternatives to brand-name GLP-1 receptor agonists, this study provides pharmacovigilance signals linking compounded products to higher odds of adverse events, contamination, preparation errors, and hospitalizations.

Limits

The study is based on FAERS, a passive surveillance database subject to voluntary reporting, underreporting, reporting bias, and inconsistent data quality. Total patient exposure is unknown, precluding incidence calculation, and disproportionality metrics cannot establish direct causality.

Cited by