Independent · Evidence-led · We don't sell peptides
EU / NordicsUpdated weeklyEN

Explore this article's sources with AI

ChatGPTClaudePerplexityGeminiGrokGoogle AI

Follow PeptideMethods on Google

First published

GLP-1 Oral Health Signals: FAERS

A 2026 FAERS analysis found oral-health reporting signals for five GLP-1 medicines, but it could not estimate patient risk or prove causation.

Why we wrote this. FAERS percentages can be mistaken for patient risk. This readout separates a useful oral-health signal from incidence and causation.

In this article (5 sections)
  1. The analysis covered five medicines through 2025
  2. Dry mouth and reflux-related events produced signals
  3. Why 8.6% is not a patient risk estimate
  4. What this readout means for oral symptoms
  5. What the study cannot answer yet

A 2026 pharmacovigilance study screened reports for five GLP-1 receptor agonists in the United States Food and Drug Administration Adverse Event Reporting System, known as FAERS. Of 364,274 adverse-event reports, 31,195 included oral, craniofacial or dental manifestations[1]. The result is a reporting signal, not an estimate that 8.6% of people taking these medicines will have an oral problem.

Spontaneous reports can help identify patterns that deserve study. They cannot establish how often an event occurs because FAERS does not provide a complete denominator of everyone exposed, and reporting can be selective. A report also cannot show by itself that a medicine caused the event. The authors called for prospective studies to characterize incidence, timing and mechanisms.

The analysis covered five medicines through 2025

Researchers screened FAERS from each medicine's initial market introduction through December 2025. The included drugs were exenatide, liraglutide, semaglutide, dulaglutide and tirzepatide[2]. The paper grouped them as GLP-1 receptor agonists for the analysis, while still reporting drug-specific patterns.

The study first counted adverse-event reports and identified those involving oral, craniofacial or dental manifestations. It then used disproportionality measures, including proportional reporting ratios and reporting odds ratios, to look for events reported more prominently with a drug than expected within the database. Those measures compare reporting patterns. They do not calculate clinical risk.

Tirzepatide accounted for the highest overall adverse-event volume despite being one of the more recently approved medicines in the analysis. Semaglutide had the largest proportion of oral health-related reports. These are different observations: one concerns total report volume, while the other concerns the share of a drug's reports that matched the oral-health grouping.

The disproportionality analysis found a drug-specific dry-mouth signal for semaglutide. Vomiting, regurgitation and reflux-related events produced significant signals across all five medicines[1]. The abstract does not provide the event counts, point estimates or confidence intervals for each of those comparisons, so their magnitude and precision cannot be reconstructed from the indexed record.

Sensory events, including dysgeusia and other taste disturbances, also had significant proportional reporting ratios and reporting odds ratios for several medicines[2]. The abstract does not identify every medicine-specific estimate. It therefore supports a class-wide screening result with selected drug-specific findings, not a ranked comparison of which treatment is most likely to cause a taste change.

An association analysis found links between vomiting and taste disturbances, and between dry mouth and dysgeusia, particularly in reports involving semaglutide and dulaglutide. Co-reporting can generate hypotheses about related symptoms or reporting behavior. It does not establish that one symptom caused another, nor does it prove a biological pathway.

Why 8.6% is not a patient risk estimate

The denominator in the paper was a set of adverse-event reports, not a cohort of treated patients. People without an event generally do not appear in FAERS, and one person's experience may generate more than one report or more than one coded event. Exposure counts, treatment duration and systematic dental examinations were not supplied by the abstract. An 8.6% reporting share cannot be converted into incidence.

Reporting can change with publicity, product uptake, time on the market and who files a report. Differences in age, diabetes severity, other medicines, oral-health history or access to dental care may also affect what is noticed and submitted. Disproportionality is useful for signal detection, but those factors prevent a causal or comparative-risk conclusion from this analysis alone.

The medicine-specific signals should stay attached to their evidence level. Our semaglutide research overview and tirzepatide research overview summarize evidence from other designs. Trial outcomes, routine-care cohorts and spontaneous reports answer different questions and should not be treated as interchangeable.

What this readout means for oral symptoms

The study supports clinical awareness, not a self-diagnosis rule. Dry mouth, altered taste, vomiting or reflux can have several explanations, including other medicines and underlying illness. A dental or medical professional can review timing, severity and competing causes. This paper cannot decide whether a GLP-1 medicine caused one person's symptom.

People should not stop or alter a prescribed medicine solely because an event appears in this database analysis. Persistent or severe symptoms warrant individualized review, especially when they affect eating, drinking or oral health. Urgent symptoms require prompt medical assessment. The appropriate response depends on the person and the symptom, not on a reporting ratio alone.

What the study cannot answer yet

The abstract does not report verified diagnoses, treatment duration, dose, symptom onset, dental examination findings or resolution after treatment changes. It also does not provide adjusted comparisons that account for patient-level differences among the five medicine groups. Those gaps limit conclusions about incidence, timing, mechanisms and relative risk.

Prospective studies could define oral outcomes before follow-up, measure baseline dental health and track exposure over time. Active-comparator cohorts could test whether the signals persist after stronger confounding control. Until then, the 2026 paper is best read as a map of reports that merit investigation, not proof that semaglutide or tirzepatide causes a particular oral condition.

Medical disclaimer: This article is for educational and journalistic purposes only and does not constitute medical advice. Peptides discussed may be classified as prescription medicines or research chemicals depending on your jurisdiction. Always consult a qualified healthcare professional before using any peptide product. PeptideMethods.com does not sell, distribute, or facilitate the sale of any peptide product.

Frequently asked

Did the study find that 8.6% of patients had oral adverse events?

No. The 8.6% figure was the share of selected FAERS adverse-event reports that included oral, craniofacial or dental manifestations. FAERS does not provide a complete denominator of treated patients, so the figure is not incidence.

Which GLP-1 medicines were included?

The analysis included exenatide, liraglutide, semaglutide, dulaglutide and tirzepatide from each medicine's market introduction through December 2025.

What oral-health signals did the researchers report?

The abstract reported a semaglutide-specific dry-mouth signal, reflux-related signals across all five medicines and taste-disturbance signals for several medicines. It did not establish incidence or causation.

Should someone stop a GLP-1 medicine because of this study?

This database analysis cannot guide an individual medication change. People with persistent or severe symptoms should seek individualized medical or dental review rather than stopping or altering prescribed treatment on their own.

Sources

  1. [1]Oral Health-Related Adverse Events of GLP-1 Receptor Agonists. JDR Clinical and Translational Research. 2026. PMID 42935458.Tier 1 · primary↩
  2. [2]NCBI PubMed EFetch record and abstract for PMID 42935458Tier 1 · primary↩

No revisions yet. First published .

About the editorial team

PeptideMethods is written and edited by the PeptideMethods Editorial Team and published by Digital Compass Group Ltd. The team is not made up of medical professionals; every health, regulatory or dosage claim on the site is tied to a primary source and is not a substitute for advice from a qualified clinician.

See our editorial policy and methodology for how we research, source and verify.

Read the pillars