Do GLP-1 drugs cause throat symptoms?

A retrospective cohort study published in The Laryngoscope on 12 August 2026 reports that adults with obesity who started a GLP-1 receptor agonist recorded more laryngeal symptoms over the following six months than matched adults who did not^[1]. The symptoms are ordinary ones: cough, a lump-in-the-throat feeling, and changes to the voice. If you take semaglutide or tirzepatide and your throat has felt different lately, this is the paper behind the headlines. It is worth reading before you conclude anything about your own case.

What the study measured

Shah and colleagues used TriNetX, a federated network of United States electronic health records, and compared 617,296 adults with obesity who received GLP-1 receptor agonist therapy against 3,503,102 controls, across records running from 1 January 2016 to 3 December 2025^[1]. Laryngeal symptoms were assessed at one, three and six months after treatment began, with the two groups propensity score matched on age, sex, race and ethnicity^[1].

At six months the odds of any laryngeal manifestation were 19 percent higher in the treated group (odds ratio 1.19, 95% confidence interval 1.16 to 1.21, p<0.0001)^[1]. Split by symptom, cough came in at 1.46, foreign body sensation, the clinical term for a lump in the throat, at 1.41, and voice and resonance disorders at 1.19^[1]. The authors conclude that use of these drugs is linked to higher rates of laryngeal symptoms, most notably cough and voice and resonance disorders^[1].

Where tirzepatide sits in the data

The class did not behave as one block. Exenatide, liraglutide, semaglutide, dulaglutide and lixisenatide each showed higher risk, with odds ratios between 1.16 and 1.91^[1]. Albiglutide and tirzepatide showed no significant association^[1]. A null result in a subgroup is not evidence of absence. It can mean the effect is not there, or that the subgroup was too small, or that its patients had not been treated long enough for a symptom to reach the record. Our tirzepatide safety notes and semaglutide safety notes cover the adverse events the trials did establish.

Why reflux is the leading explanation

The plausible route from an abdominal injection to a hoarse voice runs through the stomach. The United States prescribing information for Ozempic states that semaglutide causes a delay of early postprandial gastric emptying^[3]. Food leaves the stomach more slowly, and the same label lists gastroesophageal reflux disease among gastrointestinal reactions occurring in fewer than 5 percent of patients, at 1.9 percent on 0.5 mg and 1.5 percent on 1 mg against 0 percent on placebo, alongside dyspepsia and eructation, which is the medical word for belching^[3].

There is a name for what happens when refluxed stomach contents travel above the oesophagus. A 2024 review in the World Journal of Gastroenterology describes laryngopharyngeal reflux disease as inflammation of the laryngopharynx and upper aerodigestive tract mucosa caused by reflux beyond the oesophagus, presenting as hoarseness, cough, sore throat, a feeling of throat obstruction and excessive throat mucus^[4]. That list overlaps almost exactly with what the Laryngoscope team counted.

A separate 2026 analysis points the same way. Gallagher and colleagues, writing in JAMA Otolaryngology Head and Neck Surgery, compared 427,555 people prescribed a GLP-1 receptor agonist with 1,614,495 prescribed other diabetes medicines across 70 United States health systems, and reported an adjusted hazard ratio for chronic cough of 1.12 (95% confidence interval 1.08 to 1.16)^[2]. Against sulfonylureas the figure was 1.32, and against SGLT2 inhibitors it was 1.03, which did not reach significance^[2]. When patients with a prior reflux diagnosis were excluded, the differences grew rather than shrank^[2].

What an association cannot tell you

Neither paper can show that the drug caused the symptom. Both are retrospective analyses of records written for clinical care, not for research. Matching on age, sex, race and ethnicity leaves out the variables most likely to matter here: starting body mass index, existing reflux, smoking, asthma, inhaled steroid use, and how often a person sees a clinician. Someone who starts a weekly injection attends more appointments, and a symptom only enters a database when somebody writes it down.

Obesity also raises the risk of reflux, and reflux raises the risk of laryngeal symptoms, so a study of adults with obesity is looking at a population already primed for both. Odds ratios describe relative change, not absolute risk. The abstract reports how far the odds moved, not how many people in a hundred developed a hoarse voice. A 19 percent relative increase on a small baseline stays a small number, and we are not printing an absolute figure because the abstract does not contain one.

What we do not yet know

Whether the association is causal is open, and whether the symptom settles after someone stops the drug has not been tested. The abstract gives no dose relationship and no follow-up past six months, so it says nothing about year two or year three. The upstream picture is thin as well. A 2025 paper in Clinical Gastroenterology and Hepatology notes that little is known about the direct effects of these drugs on oesophageal function beyond the reported increase in reflux disease, and records that roughly 1 in 8 adults in the United States have ever taken one^[5]. That work reports an association with oesophageal motility abnormalities and is early. Our open questions on semaglutide track the same gaps.

When a throat symptom warrants a clinician

The threshold does not depend on which drug you take. The American Academy of Otolaryngology guideline on dysphonia advises clinicians to perform laryngoscopy, or refer to someone who can, when a voice problem fails to resolve or improve within four weeks, or sooner if a serious underlying cause is suspected^[6]. It defines dysphonia as altered vocal quality, pitch, loudness or vocal effort that impairs communication or quality of life, and names a neck mass, respiratory distress or stridor, a tobacco history and professional voice use among the reasons to be seen sooner^[6].

Nothing in this paper is a reason to stop a prescribed medicine on your own. Treatment for diabetes or weight management carries its own risks on the discontinuation side, which the semaglutide overview sets out.

Where this fits on the site

We follow the gastrointestinal side of this class closely, because the throat findings sit downstream of it. If you arrived here from a symptom rather than a headline, read the semaglutide page or the tirzepatide page for the established adverse-event profile, check what the tirzepatide trials never answered, and take the symptom itself to a clinician who can look at your larynx.

This article is educational and is not medical advice. It does not recommend starting, stopping or adjusting any medicine. Those decisions belong with a healthcare provider who knows your history.