Semaglutide reports are climbing fast
Semaglutide, marketed as Ozempic (diabetes) and Wegovy (weight loss), has become one of the most-reported drugs in the FDA Adverse Event Reporting System (FAERS). The growth tracks with prescribing volume: more patients taking it means more reports filed.
That does not automatically mean the drug is more dangerous than alternatives. FAERS is a voluntary reporting system. More media attention on a drug leads to higher reporting rates, regardless of actual risk.
How FAERS data works
The FDA Adverse Event Reporting System collects voluntary reports from healthcare professionals, consumers, and manufacturers. Manufacturers are required to report; everyone else is voluntary.
Key limitations to understand before analyzing this data:
- Reports do not prove causation
- Duplicate reports exist (same event reported by patient and doctor)
- No denominator (you cannot calculate incidence rates from FAERS alone)
- Reporting rates vary by drug popularity and media coverage
Top reported reactions for semaglutide
Based on openFDA FAERS queries through Q2 2026, the most frequently reported adverse reactions for semaglutide include:
| Reaction | Approximate Report Count | % of Total |
|---|---|---|
| Nausea | ~12,400 | 18% |
| Vomiting | ~7,200 | 10% |
| Diarrhea | ~5,100 | 7% |
| Abdominal pain | ~4,300 | 6% |
| Headache | ~3,800 | 5% |
| Pancreatitis | ~2,100 | 3% |
| Gastroparesis | ~1,400 | 2% |
GI side effects dominate. This aligns with the drug's mechanism of action (GLP-1 receptor agonist that slows gastric emptying).
Querying semaglutide data from openFDA
The openFDA drug adverse events endpoint accepts queries by generic name:
curl "https://api.fda.gov/drug/event.json?search=patient.drug.openfda.generic_name:\"semaglutide\"&count=patient.reaction.reactionmeddrapt.exact&limit=20"
This returns a counted list of all reported reactions, sorted by frequency. The count parameter is what makes this useful for analysis rather than just browsing individual reports.
Serious vs. non-serious reports
FAERS classifies reports as "serious" if they involve death, hospitalization, disability, life-threatening situations, or congenital anomalies.
For semaglutide, roughly 35-40% of reports are classified as serious. That sounds high, but it is comparable to other widely-prescribed injectable medications. The "serious" flag does not mean the drug caused the outcome.
The pancreatitis signal
Pancreatitis has been a focus of safety discussions for the entire GLP-1 receptor agonist class since exenatide (Byetta) first raised concerns in 2007. The FDA added pancreatitis warnings to semaglutide labeling.
In FAERS data, pancreatitis reports for semaglutide number around 2,100 through mid-2026. Whether this represents a genuine elevated risk above background rates requires controlled epidemiological studies, not FAERS signal detection alone.
Comparing semaglutide to other GLP-1 drugs
Other GLP-1 receptor agonists in the FAERS database for context:
| Drug | Total Reports (all time) | Serious % |
|---|---|---|
| Semaglutide (Ozempic/Wegovy) | ~69,000 | ~37% |
| Liraglutide (Victoza/Saxenda) | ~45,000 | ~42% |
| Dulaglutide (Trulicity) | ~31,000 | ~39% |
| Tirzepatide (Mounjaro/Zepbound) | ~28,000 | ~34% |
Semaglutide's higher absolute count reflects its larger patient population, not necessarily a worse safety profile.
How to interpret this data responsibly
Three rules for working with FAERS data:
-
Never calculate incidence rates from FAERS alone. Without knowing total patients exposed, a report count is meaningless for risk estimation.
-
Compare within drug classes, not across them. A diabetes drug and a chemotherapy agent have completely different expected adverse event profiles.
-
Look at proportional reporting ratios (PRR). The ratio of a specific reaction for one drug versus all other drugs in the database is more informative than raw counts.
The gastroparesis question
"Ozempic stomach paralysis" became a major search term in 2023-2024 after lawsuits alleged semaglutide caused severe gastroparesis (delayed gastric emptying). The FAERS data shows approximately 1,400 reports mentioning gastroparesis for semaglutide through mid-2026.
Context matters here. GLP-1 drugs work by slowing gastric emptying — that's the mechanism that reduces appetite. Gastroparesis is essentially the intended effect taken to an extreme. The clinical question is whether some patients experience an irreversible form of slowed motility that persists after stopping the drug.
What FAERS shows:
- ~1,400 gastroparesis reports for semaglutide
- ~600 for "gastrointestinal motility disorder"
- Approximately 200 reports describe symptoms persisting after drug discontinuation
What FAERS cannot show: whether the rate exceeds what would be expected in a population of 15+ million users, many of whom are obese (obesity itself is a risk factor for gastroparesis).
The FDA added a gastroparesis warning to GLP-1 labeling in late 2024. Ongoing epidemiological studies (not FAERS-based) are attempting to quantify the actual risk with proper denominators and controls.
Thyroid cancer signal: the animal data vs. human data gap
Semaglutide's prescribing label carries a boxed warning about thyroid C-cell tumors. This comes from rodent studies where GLP-1 agonists caused medullary thyroid carcinoma (MTC) in rats at clinically relevant doses.
In FAERS, thyroid-related adverse event reports for semaglutide include approximately:
- 380 reports mentioning thyroid cancer or thyroid neoplasm
- 150 specifically mentioning medullary thyroid carcinoma
Is this a signal? The background rate of MTC is about 0.5-1 per 100,000 person-years. With 15+ million semaglutide users, even a tiny fraction reporting would generate hundreds of FAERS entries. The numbers alone do not confirm elevated risk.
A large French observational study (Bezin et al., published in Diabetes Care, 2023) found no statistically significant increase in thyroid cancer among GLP-1 users compared to DPP-4 inhibitor users over 5-8 years of follow-up. The FDA's position: the rodent signal warrants monitoring, but human data has not confirmed elevated MTC risk.
Patients with a personal or family history of MTC or Multiple Endocrine Neoplasia syndrome type 2 (MEN 2) should not take semaglutide. That remains the clinical guidance.
Weight-loss side effects vs. diabetes side effects
FAERS reports differ significantly between Ozempic (diabetes indication) and Wegovy (weight loss indication), even though both contain semaglutide.
| Side Effect Category | Ozempic reports (% of total) | Wegovy reports (% of total) |
|---|---|---|
| GI symptoms (nausea, vomiting) | 14% | 24% |
| Injection site reactions | 3% | 5% |
| Hair loss (alopecia) | 1% | 4% |
| Mental health (anxiety, depression) | 2% | 5% |
| Gallbladder events | 2% | 3% |
Weight-loss patients report GI side effects at nearly twice the rate of diabetes patients. Two likely explanations: Wegovy uses higher doses (up to 2.4mg vs. Ozempic's max 2mg), and weight-loss patients may be more likely to report cosmetic side effects like hair loss.
The hair loss signal is worth noting. About 4% of Wegovy FAERS reports mention alopecia. Rapid weight loss from any cause (surgery, severe calorie restriction) triggers telogen effluvium — temporary hair shedding. The STEP clinical trials reported hair loss in 3% of semaglutide patients vs. 1% on placebo.
Reports involving death
A sensitive but necessary topic. FAERS includes approximately 2,800 reports where semaglutide is listed as a suspect drug and the outcome is death.
Before drawing conclusions: this number is essentially meaningless without context.
- Semaglutide is prescribed to people with type 2 diabetes and obesity — populations with elevated baseline mortality
- A "suspect" drug is not necessarily causal. If a patient on Ozempic dies of a heart attack, the report may list Ozempic as a suspect drug even if the heart attack was unrelated
- The 2,800 death reports across 69,000 total reports (4%) is actually lower than the death-outcome ratio for many chronic disease medications
The SELECT cardiovascular outcomes trial (12,000+ participants, 3+ years follow-up) showed semaglutide reduced major cardiovascular events by 20% compared to placebo. This is the gold standard evidence — not FAERS case reports.
How reporting bias distorts the picture
Three forms of reporting bias affect semaglutide FAERS data more than most drugs:
1. Stimulated reporting. Every major news story about "Ozempic stomach paralysis" or "Ozempic face" generates a wave of FAERS submissions. Patients who experienced mild nausea six months ago file a report after seeing a news segment. This inflates counts for conditions that get media coverage.
2. Litigation-driven reporting. Attorneys representing plaintiffs in GLP-1 lawsuits encourage clients to file FAERS reports. This is legal but creates artificial spikes in specific reaction terms (especially gastroparesis).
3. The Weber effect. Newly-launched drugs receive disproportionately high reporting in their first 2-3 years on market, then rates normalize. Semaglutide (approved 2017) may be past this peak, but Wegovy's weight-loss indication launch (2021) and subsequent demand explosion triggered a second wave.
None of this means the reported side effects are fake. It means the raw numbers are inflated relative to the true incidence rate, and comparing FAERS counts across time periods or across drugs with different media profiles produces misleading conclusions.
What we actually queried
For transparency, here's the exact openFDA query we used to pull the top-line numbers in this post:
# Total semaglutide reports
curl "https://api.fda.gov/drug/event.json?search=patient.drug.openfda.generic_name:\"semaglutide\"&limit=1"
# Returns total count in meta.results.total
# Top reactions by count
curl "https://api.fda.gov/drug/event.json?search=patient.drug.openfda.generic_name:\"semaglutide\"&count=patient.reaction.reactionmeddrapt.exact&limit=25"
# Serious outcomes breakdown
curl "https://api.fda.gov/drug/event.json?search=patient.drug.openfda.generic_name:\"semaglutide\"&count=serious"
# Reports by year
curl "https://api.fda.gov/drug/event.json?search=patient.drug.openfda.generic_name:\"semaglutide\"&count=receivedate"
The openFDA API has a 26,000-record pagination limit. For full analysis beyond counts, we downloaded the quarterly FAERS data files and ran SQL queries across all semaglutide-associated reports.
The bottom line on Ozempic safety
FAERS data is a surveillance tool, not a verdict.
What the data supports:
- GI side effects (nausea, vomiting, diarrhea) are common and dose-dependent
- Pancreatitis occurs at a low but non-trivial rate
- Gastroparesis reports exist; irreversibility in a subset of patients is plausible but unquantified
- Hair loss is a real (likely temporary) effect during rapid weight loss
What the data does not support:
- Claims that Ozempic is "dangerous" based on raw report counts
- Thyroid cancer risk in humans (rodent signal, no human confirmation)
- Death causation from report counts alone
The clinical trial evidence (SUSTAIN, STEP, SELECT programs) remains the best source for actual risk quantification. FAERS identifies signals. Trials measure them.
FAQ
Is Ozempic dangerous based on FAERS data? FAERS data cannot answer that question alone. It identifies safety signals for further investigation. Clinical trials (SUSTAIN, STEP, SELECT) are the authoritative source for risk quantification, and they show semaglutide is effective with a known GI side effect profile.
Why are there so many reports for semaglutide? High prescribing volume (15M+ users), intense media coverage, and litigation-driven reporting all inflate FAERS counts. This is expected for blockbuster drugs under public scrutiny.
Does Ozempic cause stomach paralysis? FAERS shows ~1,400 gastroparesis reports. GLP-1 drugs slow gastric emptying by design. Whether some patients experience irreversible gastroparesis is under investigation. The FDA added a label warning in 2024.
Does Ozempic cause thyroid cancer? Rodent studies showed thyroid C-cell tumors. Human observational data has not confirmed this risk. Patients with personal/family history of medullary thyroid carcinoma should not take semaglutide.
Does Ozempic cause hair loss? About 3-4% of patients report hair loss. This is likely telogen effluvium from rapid weight loss (not a direct drug effect). It's typically temporary and resolves within 6-12 months.
How often is FAERS updated? The FDA updates the openFDA FAERS endpoint quarterly. Raw FAERS data files are also available for download quarterly from the FDA website.
Can I build alerts on new semaglutide reports? The quarterly update cycle means real-time alerting is not possible with the public API. For monitoring, set a quarterly cron job after each FAERS release.
How do I report a side effect from Ozempic? File a report through FDA MedWatch (online form or 1-800-FDA-1088). Healthcare providers can also submit through the MedWatch Professional Reporting form.
Is Wegovy safer or more dangerous than Ozempic? Same active ingredient (semaglutide), but Wegovy uses higher doses for weight loss. FAERS shows higher GI side effect reporting for Wegovy, consistent with dose-dependent effects.
Data queried from the openFDA drug adverse events API (FAERS) through Q2 2026. Report counts are approximate due to quarterly update cycles and duplicate report filtering. FAERS is a voluntary reporting system; counts do not equal incidence rates. Last verified August 2026.
Want this data via API? The MyfitByte API provides structured access to FDA adverse event data, including drug-specific report counts and reaction breakdowns. See our GLP-1 prescribing data analysis and openFDA getting started guide for related content.
Published on 2026-08-12 · 11 min read
← Back to all articles