GLP-1 medicines and adverse-event reporting patterns

Research question

Are particular events disproportionately reported with the selected medicines in the frozen FAERS snapshot?

Examine disproportionate reporting for two selected GLP-1 medicines and two exact adverse-event terms in one complete public FDA quarterly release. This is a reporting-pattern analysis, not an estimate of clinical incidence, causality, or comparative drug safety.

Data and comparison design

A reporting odds ratio compares the balance of event and other-event reports for a medicine with the corresponding comparator reports. Case eligibility and matching rules define that comparison. FAERS provides reports, rather than the total number of treated or untreated people.

Input or source	Role in this case	Source context
FDA quarterly extract listing	Identifies the 2026 Q2 ASCII release	Public FDA source, inspected 2026-10-08
FDA 2026 Q2 ASCII ZIP	Original quarterly reporting tables and documentation	Retained in the Session; cite FDA and respect source-specific reuse terms
FDA AEMS limitations	Interpretation of spontaneous-reporting data	Reports do not establish incidence or causation

Recorded findings

Drug	Pancreatitis reporting odds ratio	Cholelithiasis reporting odds ratio
Semaglutide	4.20	4.60
Liraglutide	3.58	6.71
Eligible FAERS cases	422,458 in 2026 Q2	Frozen primary matching rules
Measure boundary	Reporting disproportionality	Not incidence or a head-to-head clinical safety ranking

The analysis retained 422,458 eligible unique cases from FDA FAERS 2026 Q2. All four planned exact-term comparisons were estimable. Primary reporting odds ratios were 4.20 for semaglutide–Pancreatitis, 4.60 for semaglutide–Cholelithiasis, 3.58 for liraglutide–Pancreatitis, and 6.71 for liraglutide–Cholelithiasis. The report supplies raw cell counts, intervals, Holm-adjusted tests, suspect-role sensitivity estimates, and limitations. Liraglutide co-reporting counts are sparse; combination products and an ambiguous brand-containing name remain under the frozen matching rule and are disclosed.

What the comparison shows

The ratios flag reporting patterns for investigation. Differences between medicines can reflect reporting and use patterns, so these values cannot rank their absolute clinical risks.

Assumptions and limits

Reporting odds ratios are not incidence, causality or head-to-head safety rankings. Reporting bias, absent exposure denominators and sparse co-reports limit interpretation. The corrected reproduction bundle fixes metadata without changing scientific results.

Sources and research materials

The unchanged research package comes from upstream Case 26. It includes the archived session, analysis materials, results and figures. The poster uses an AI-generated illustrative photograph. This introduction summarizes archived findings; the analysis was not rerun.