Evidence library

What are the side effects of GLP-1s?

Safety is where honesty matters most. Here the evidence is sorted into three buckets: the common and usually manageable, the rare and serious, and the scary-sounding risks that remain unproven. Saying which is which plainly is the accurate move.

How to read these grades

Evidence strength

  • Strong evidence (does / reduces / improves)
  • Moderate evidence (probably / likely)
  • Early / weak evidence (may / might)

Tested in GLP-1 users?

  • Tested in GLP-1 users
  • Not tested in GLP-1 users
  • Trial underway

Digestive side effects are common, usually mild, and tend to fade

Strong evidence Tested in GLP-1 users

The most common side effects are gastrointestinal: nausea, vomiting, diarrhea, and constipation. They affect a large share of users but are typically mild to moderate, cluster during dose increases, and ease over time. The drug-caused excess is smaller than the raw rate, because placebo groups also report them.

What it rests on

  • Klein 2024 1

    A pooled analysis of four trials measured gastrointestinal side effects when starting a GLP-1. Rates ran about 7 to 40% depending on drug and dose, highest for semaglutide 2.4 mg; placebo groups ran 2 to 30%, so the true drug-caused excess is smaller than the raw rate.

    Pooled RCTs (n=16,996)

  • Filippatos 2014 2

    An older review of the drug class. Nausea ran up to about 50% and was dose-dependent; the pancreatitis and pancreatic-cancer signals were contested and not confirmed by regulators, and the thyroid-tumor signal came only from rodents. It predates semaglutide and tirzepatide.

    Class review, first-generation drugs

  • Mozaffarian 2025 advisory 3

    A joint advisory from four medical societies summarizing current best practice for supporting people on GLP-1s. It is an expert consensus document, not a trial, and is the standard reference for what the field recommends and where evidence is still missing.

    Four-society expert consensus

Stomach effects are the top reason people quit, but slow dosing keeps most on track

Moderate evidence Tested in GLP-1 users

Gastrointestinal intolerance is the leading reason people stop, but slow, individualized dose increases and simple eating changes let most people continue. The management advice is expert experience rather than trial-proven, and one study found flexible dosing did not clearly improve tolerability.

What it rests on

  • Wharton 2022 4

    An expert commentary offering practical ways to manage stomach side effects: patient education, slow dose escalation, and simple diet changes. The authors are clear these come from clinical experience, not trials, and the paper is drug-maker funded.

    Expert commentary (drug-maker funded)

  • Klein 2024 1

    A pooled analysis of four trials measured gastrointestinal side effects when starting a GLP-1. Rates ran about 7 to 40% depending on drug and dose, highest for semaglutide 2.4 mg; placebo groups ran 2 to 30%, so the true drug-caused excess is smaller than the raw rate.

    Pooled RCTs (n=16,996)

The serious risks are worth knowing, and mostly rare or unproven

Early / weak evidence Not tested in GLP-1 users

Serious concerns exist but are mostly uncommon or unconfirmed. Pancreatitis and pancreatic cancer have not been shown to be caused by these drugs, the thyroid-tumor signal comes from rodents rather than people, and kidney injury is rare and tied to dehydration. The main evidence here is an older review of first-generation drugs.

What it rests on

  • Filippatos 2014 2

    An older review of the drug class. Nausea ran up to about 50% and was dose-dependent; the pancreatitis and pancreatic-cancer signals were contested and not confirmed by regulators, and the thyroid-tumor signal came only from rodents. It predates semaglutide and tirzepatide.

    Class review, first-generation drugs

  • Mozaffarian 2025 advisory 3

    A joint advisory from four medical societies summarizing current best practice for supporting people on GLP-1s. It is an expert consensus document, not a trial, and is the standard reference for what the field recommends and where evidence is still missing.

    Four-society expert consensus

The suicidality question was reviewed and resolved in favour of reassurance

Moderate evidence Tested in GLP-1 users

In the weight-management trials, semaglutide did not raise depression or suicidal thoughts versus placebo. The FDA then reviewed the question directly and did not find that these medicines increase suicidal thinking or behaviour, and the Suicidal Behavior and Ideation warning was removed in February 2026 from the three labels that carried it (Wegovy, Zepbound and Saxenda). The other GLP-1s never carried it. One honest limit remains: the trials excluded people at higher psychiatric risk, so they cannot speak to that group, and any new or worsening low mood is still worth raising with your prescriber.

What it rests on

  • Wadden 2024 5

    A pooled analysis of the STEP weight-management trials checked for mental-health effects. Depression scores and reports of suicidal thoughts stayed at or below 1% and no worse than placebo, but the trials had excluded people at higher psychiatric risk.

    Pooled STEP trials (higher-risk patients excluded)

  • Mozaffarian 2025 advisory 3

    A joint advisory from four medical societies summarizing current best practice for supporting people on GLP-1s. It is an expert consensus document, not a trial, and is the standard reference for what the field recommends and where evidence is still missing.

    Four-society expert consensus

The side effect most coverage ignores: muscle and bone loss

Moderate evidence Tested in GLP-1 users

Rapid weight loss takes off lean muscle and bone along with fat, a real and under-discussed downside. It is not unavoidable, which is exactly why the protein and training evidence matters. The specific figures live in the muscle and bone sections.

What it rests on

  • Mozaffarian 2025 advisory 3

    A joint advisory from four medical societies summarizing current best practice for supporting people on GLP-1s. It is an expert consensus document, not a trial, and is the standard reference for what the field recommends and where evidence is still missing.

    Four-society expert consensus

Sources

  1. 1. Klein KR, Clemmensen KKB, Fong E, Olsen S, Abrahamsen T, Lingvay I. Occurrence of Gastrointestinal Adverse Events Upon GLP-1 Receptor Agonist Initiation With Concomitant Metformin Use: A Post Hoc Analysis of LEADER, STEP 2, SUSTAIN-6, and PIONEER 6. Diabetes Care. 2024;47(2):280-284. doi:10.2337/dc23-1791
  2. 2. Filippatos TD, Panagiotopoulou TV, Elisaf MS. Adverse Effects of GLP-1 Receptor Agonists. Rev Diabet Stud. 2014;11(3-4):202-230. doi:10.1900/rds.2014.11.202
  3. 3. Mozaffarian D, Agarwal M, Aggarwal M, et al. Nutritional priorities to support GLP-1 therapy for obesity: a joint Advisory from the American College of Lifestyle Medicine, the American Society for Nutrition, the Obesity Medicine Association, and The Obesity Society. The American Journal of Clinical Nutrition. 2025;122(1):344-367. doi:10.1016/j.ajcnut.2025.04.023
  4. 4. Wharton S, Davies M, Dicker D, et al. Managing the gastrointestinal side effects of GLP-1 receptor agonists in obesity: recommendations for clinical practice. Postgraduate Medicine. 2022;134(1):14-19. doi:10.1080/00325481.2021.2002616
  5. 5. Wadden TA, Brown GK, Egebjerg C, et al. Psychiatric Safety of Semaglutide for Weight Management in People Without Known Major Psychopathology. JAMA Intern Med. 2024;184(11):1290. doi:10.1001/jamainternmed.2024.4346