500 Deaths, 22,000 Adverse Events: What the FDA's Own Data Says About Semaglutide

The FDA Adverse Event Reporting System (FAERS) is the largest post-market drug safety database in the world. It's also publicly available, frequently underutilized by plaintiff firms, and full of litigation-ready intelligence.

Here's what FAERS reveals about GLP-1 drugs.

The Raw Numbers

LEWS continuously ingests and analyzes FAERS data. For semaglutide-based drugs (Ozempic, Wegovy, Rybelsus), the database contains:

• Nearly 500 deaths reported in the United States
• 22,287 total adverse reaction records
• Reports spanning gastroparesis, pancreatitis, bowel obstruction, gallbladder disease, and vision loss

For compounded versions of semaglutide and tirzepatide — the knockoff formulations sold by compounding pharmacies — the FDA has documented:

• 900+ adverse event cases as of December 31, 2024
• 17 deaths
• 520+ reports for compounded semaglutide specifically
• 480+ reports for compounded tirzepatide

A Critical Distinction: Tirzepatide vs. Semaglutide

Not all GLP-1 drugs carry the same risk profile. A recent analysis of FAERS data published in Medscape found that tirzepatide has zero reported gastroparesis cases in the FDA database. This distinguishes it sharply from semaglutide and liraglutide, which have significant gastroparesis signal.

This matters for litigation strategy. Mounjaro/Zepbound (tirzepatide) cases may need to be argued differently than Ozempic/Wegovy (semaglutide) cases. Firms that understand the drug-specific risk profiles will build stronger claims.

What FAERS Can Tell Plaintiff Firms

FAERS data is powerful for case development because it provides:

1. Signal volume — How many reports exist for a specific drug-injury combination
2. Seriousness markers — Hospitalization, disability, life-threatening, death
3. Patient demographics — Age, sex, weight distributions
4. Concurrent medications — What else patients were taking (relevant for causation defense)
5. Reporter type — Physician-reported vs. consumer-reported (credibility weighting)
6. Temporal patterns — When adverse events peak relative to drug launch or label changes

How LEWS Uses FAERS

Our FAERS ingestion agent runs weekly, backfilling data windows going back to 2018. We don't filter by drug — we ingest everything, then let our clustering algorithm identify emerging injury patterns.

This approach has caught signals that drug-specific monitoring would miss. When a new injury type starts appearing across multiple drugs in the same class, our system detects the pattern before anyone is specifically looking for it.

For GLP-1 drugs specifically, LEWS has identified 547 FAERS signals that our system classified as high-relevance, cross-referencing them with:

• 14,700+ PubMed research papers on GLP-1 safety
• 2,133 Reddit consumer complaints
• 619 court filings tracked through CourtListener
• 551 news articles on GLP-1 litigation

The Regulatory Enforcement Angle

In September 2025, the FDA issued 55+ warning letters to online sellers of compounded GLP-1 medications. Separately, Novo Nordisk received a warning letter for failing to properly investigate and report adverse events.

These enforcement actions are gold for plaintiff attorneys — they demonstrate that both the manufacturers and the distribution chain failed in their safety obligations.

Why This Data Matters Now

Bellwether trials in MDL 3094 are expected in late 2026. The firms that have already mapped the FAERS landscape — understanding exactly which injuries, which patient populations, and which reporting patterns support their cases — will have a significant advantage in discovery and trial preparation.