Evidence section
Side effects and safety studies
Two pharmacists registered and practising in British Columbia read this page against the study pages and papers it lists and signed it. Their names are withheld from the site and held on file; their registration is confirmed on request.
Safety questions about GLP-1 medicines come up constantly in news and online forums. This hub collects the studies that examined them: meta-analyses of trial data, large health-record studies, and guidance for people having surgery or endoscopy. Each page explains what kind of study it is, because an observational study of records answers a different question than a randomized trial. Where a study found no increased risk, the page also says what it could not rule out.
What these studies show together
The answer depends on the question. Gallbladder and bile duct problems are the most consistent signal: a pooled analysis of 76 randomized trials found more of them with GLP-1 medicines, especially in weight-loss trials, at higher doses and with longer use. For thyroid cancer, pancreatitis, diabetic eye disease and suicidal thoughts, the studies here either found no clear increase or disagree with each other. Around surgery and endoscopy, the evidence points to individual planning with the care team rather than a single rule.
The studies side by side
| Question | Study | Design and who was studied | Result, as reported |
|---|---|---|---|
| Gallbladder and bile ducts | Meta-analysis of 76 trials | Randomized trials, 103371 adults with diabetes, obesity or other conditions | RR 1.37 (1.23 to 1.52) overall; 2.29 (1.64 to 3.18) in weight-loss trials |
| Gallbladder and bile ducts | US claims cohorts | Observational, adults with type 2 diabetes, compared with SGLT2 inhibitors | Biliary events HR 1.15 (1.05 to 1.26), described as fewer than one extra event per 1000 person-years |
| Pancreatitis | US claims cohorts | Same study | Hospital admission for acute pancreatitis HR 1.01 (0.90 to 1.13), no clear difference |
| Thyroid cancer | French case-control study | Observational, adults with type 2 diabetes, compared with non-use | 1 to 3 years of use: adjusted HR 1.58 (1.27 to 1.95) |
| Thyroid cancer | Scandinavian cohort | Observational, new users compared with DPP-4 inhibitor users | HR 0.93 (0.66 to 1.31), no clear difference |
| Diabetic eye disease | Meta-analysis of 6 heart outcome trials | Randomized trials, adults with type 2 diabetes and heart risk | OR 1.10 (0.93 to 1.30) overall; trials with larger HbA1c drops reported more eye events |
| Suicidal thoughts | US health records cohort | Observational, semaglutide compared with other weight-management medicines | New suicidal thoughts HR 0.27 (0.20 to 0.36) over six months |
| Endoscopy | US endoscopy cohort | Observational, adults with type 2 diabetes, compared with SGLT2 inhibitors | Aspiration RR 0.98 (0.73 to 1.31); procedure stopped RR 1.99 (1.56 to 2.53) |
| Surgery | US surgery cohort | Observational, adults having one of 14 common operations | Aspiration pneumonia adjusted OR 0.78 (0.57 to 1.06), no clear difference |
| Before a procedure | Multisociety guidance | Consensus guidance for clinicians | Decisions shared between the prescriber, the procedure team and the anaesthesia team, based on the person's own situation |
| Other medicines | Drug interaction review | Systematic review, mostly healthy volunteers | Peak levels of oral medicines, including contraceptive pills, lower or later; total exposure not changed in a way the authors considered clinically significant |
A ratio of 1 means no difference. When the range in brackets includes 1, the study did not find a clear increase or decrease.
What is consistent
Gallbladder and bile duct events appear in both a pooled analysis of randomized trials and a large insurance claims study. The claims study describes the increase as small in absolute terms, and it found a similar increase with DPP-4 inhibitors, another diabetes medicine class. Rapid weight loss is itself linked to gallstones, and neither study can fully separate the effect of the medicine from the effect of losing weight.
Around procedures, none of the three pages supports a single rule for everyone. The endoscopy study found no clear difference in aspiration but more procedures stopped, the surgery study found no clear increase in aspiration pneumonia, and the guidance asks the teams involved to decide together for each person.
Where the studies disagree
The two thyroid cancer studies reach different results. The French study, which compared users with non-users, found a higher recorded rate after 1 to 3 years of use. The Scandinavian study, which compared new users with people starting a different diabetes medicine, found no clear increase, though its range still allows for up to a 31% relative increase. The difference in comparison groups, and the possibility that people on GLP-1 medicines have more tests and so more cancers found, mean neither study settles the question on its own.
What these studies cannot tell you
Most of these studies were done in people with type 2 diabetes, and most grouped GLP-1 medicines together, so they do not describe people using a single product at weight-management doses without diabetes. The drug interaction review did not include tirzepatide. Nearly all are observational, so they show associations rather than cause and effect, and several report only relative risks, which do not say how many people out of 1000 were affected. Follow-up was short for rare outcomes: an average of 3.9 years in the Scandinavian thyroid study, a median of 3.4 years for eye disease and six months for suicidal thoughts. The suicidal thoughts study does not show that semaglutide protects anyone.
Whether to pause a medicine before a procedure, how to manage contraception, and what a personal or family history means for a particular medicine are questions for the prescriber or pharmacist, who can see the full picture. For each medicine's Canadian product information, see semaglutide in Canada and tirzepatide in Canada.
This summary draws only on the study pages listed below and adds no figures that are not on them. Written 2026-09-23; reviewed by two pharmacists registered in British Columbia, 2026-09-23.
Studies in this section (10)
GLP-1 medicines and diabetic eye disease: what six large trials show
Six GLP-1 heart-outcome trials pooled: no clear link to diabetic retinopathy, but bigger HbA1c drops were tied to more eye events. Canadian summary.
Do GLP-1 medicines raise the risk of gallbladder problems?
A meta-analysis of 76 randomized trials linked GLP-1 medicines to more gallbladder and biliary disease, most in weight-loss trials. Canadian summary.
GLP-1 medicines and thyroid cancer: a French insurance-data study
A French study of people with type 2 diabetes linked 1 to 3 years of GLP-1 use with thyroid cancer. What the finding means for Canadian readers.
Do GLP-1 injections change how oral medicines are absorbed?
A systematic review found injectable GLP-1 medicines can delay the peak of oral drugs such as the pill without changing total exposure much. Canada review.
GLP-1 medicines and thyroid cancer: the Scandinavian cohort study
A Scandinavian study of 145410 GLP-1 users found no clear rise in thyroid cancer versus DPP-4 inhibitors over 3.9 years. A summary for Canadian readers.
GLP-1 medicines before an endoscopy: aspiration and stopped procedures
US study of 43365 adults with diabetes: GLP-1 use before endoscopy did not raise aspiration risk but was tied to more stopped procedures. Canadian summary.
Semaglutide and suicidal thoughts: a large health-record study
A US health-record study of 240618 people found no higher rate of recorded suicidal thoughts with semaglutide than with other medicines. Canadian summary.
GLP-1 medicines and aspiration pneumonia after common surgery
A US study of 366476 surgical patients found no clear rise in aspiration pneumonia in people prescribed a GLP-1 medicine before surgery. Canadian summary.
GLP-1 medicines before surgery: what the multisociety guidance says
Multisociety consensus guidance on GLP-1 medicines around surgery calls for shared decisions across prescriber, procedure and anaesthesia teams. Canada.
GLP-1 medicines, pancreatitis and gallbladder events in US claims data
A US claims study of people with diabetes found no clear rise in pancreatitis with GLP-1 medicines and a small rise in biliary events. Canadian summary.
Other evidence sections
How to read these pages
Every page in this section explains one published study. The numbers are reported as the researchers published them, with a link to the paper on PubMed. Pages say what a study cannot tell you and never rank one medicine against another. They are education, not medical advice. For how a medicine is used and paid for in Canada, see medicines, public drug coverage by province and compare services.
Reviewed and signed by two pharmacists registered in British Columbia, 2026-09-23.