Long-term safety

Ozempic long-term side effects: what the trials show, and what they don't

Most “long-term side effect” lists don't say how long the evidence actually runs. Here we separate what the multi-year trials found from the genuine unknowns — with dated sources for both.

Written byGLP1Zoom Editorial, Medical content team · Independent editorial

Medically reviewed byGLP1Zoom Editorial Team, Evidence-based editorial review Reviewed Jun 10, 2026

Verified Jul 2026 · re-checked quarterlyBy the GLP1Zoom editorial teamWe compare, we don’t sell

The short answer

The longest controlled studies — STEP 5 (2 years) and SELECT (~3.3 years) — found no new long-term safety signals. The real long-term considerations are lean/muscle-mass loss, a higher rate of gallbladder problems (from rapid weight loss), and ongoing GI effects. There is no confirmed link to pancreatic cancer. The honest caveat: controlled data don't yet extend much past ~4 years, so effects over 5+ years aren't established.

Key takeaways

Longest RCTs: STEP 5 (2 years, Nature Medicine 2022) and SELECT (mean ~3.3 years, NEJM 2023) — both found no new safety signals beyond the known profile.
Lean/muscle-mass loss is the key long-term consideration: STEP 1 DXA substudy showed lean mass ~-9.7% (fat fell far more). Protein + resistance training are the standard mitigations.
Gallbladder disease is more common (~1.37x, JAMA Internal Medicine 2022) — driven by rapid weight loss changing bile chemistry.
No confirmed link to pancreatic cancer (meta-analyses non-significant). Pancreatitis is a distinct, rare, labeled warning.
Honest gaps: little controlled data beyond ~4 years; not established as safe in pregnancy (stop ~2 months before conception); post-discontinuation effects (weight regain) not fully characterized.

How long is “long-term,” really?

This is the part most articles skip. The strongest long-term evidence for semaglutide comes from two trials:

STEP 5 (Nature Medicine, 2022) — a 2-year randomized trial of semaglutide 2.4 mg for weight management. It reported sustained weight loss (about −15% vs −2.6% on placebo) and, importantly, no new safety signals over two years; side effects were mostly the familiar mild-to-moderate GI ones.
SELECT (NEJM, 2023) — over 17,000 adults followed for a mean of about 3.3 years, the longest safety exposure for semaglutide 2.4 mg. More GI-related discontinuations and more gallbladder disorders, but no excess of serious adverse events — alongside a 20% reduction in cardiovascular events.

So “long-term” today means up to roughly three to four yearsof good controlled data. That's genuinely reassuring within that window — and it's also the honest boundary of what's known.

The real long-term considerations

1. Lean and muscle mass

In the STEP 1 68-week DXA substudy, total lean mass fell about 9.7%while fat fell much more, so the proportion of lean mass to body weight actually rose. DXA “lean mass” also counts water, organs and bone, so it overstates true muscle loss — but preserving muscle over years matters. Adequate protein and 2–3 weekly resistance sessions are the evidence-based response. Full detail in our GLP-1 and muscle-loss guide.

2. Gallbladder disease

Gallbladder problems are more frequent with GLP-1 medications — a 2022 meta-analysis (JAMA Internal Medicine) estimated roughly 1.37× the risk versus placebo, and STEP obesity trials showed gallbladder disorders a couple of percent higher than placebo. The main driver is rapid weight loss itself, which changes bile composition and promotes gallstones. Watch for upper-right abdominal pain, fever, or yellowing skin.

3. Gastrointestinal effects over time

The common GI effects (nausea, constipation, diarrhea) usually ease after the early titration period, but some people have them intermittently long-term. They're rarely dangerous but can affect quality of life and hydration — persistent vomiting or diarrhea is the usual path to the rare kidney-injury warning, so staying hydrated matters.

4. The thyroid boxed warning (context, not a long-term finding)

Any long-term discussion should name the label's boxed warning: in rodents, semaglutide caused thyroid C-cell tumors, including medullary thyroid carcinoma (MTC). Importantly, whether this happens in humans is unknown — the label states the human relevance has not been determined, and the multi-year trials have not shown a thyroid-cancer signal in people. It is contraindicated for anyone with a personal or family history of MTC or MEN 2. This is a precaution based on animal data, not an observed long-term human effect — see our full safety overview. A separate, rare eye condition (NAION) that the EMA classes “very rare” is covered in our GLP-1 and eyes guide.

The pancreatic-cancer question

No confirmed causal link. Meta-analyses of the trial and observational data have not found a statistically significant association between semaglutide and pancreatic cancer. A separate, distinct issue is acute pancreatitis(inflammation, not cancer): it is a labeled warning, and while a small signal appears in some analyses, it is uncertain and weakens on closer stratification. Recognize severe, persistent abdominal pain and seek care — but the “Ozempic causes pancreatic cancer” claim is not supported by current evidence.

What is honestly still unknown

Effects beyond ~4 years

Little controlled safety data exists past about four years of continuous use, so 5-10+ year effects are not established. Ongoing trials will extend the record.

Pregnancy

Not established as safe; animal studies show reproductive harm. The label advises discontinuing about two months before a planned pregnancy because of the long half-life.

After you stop

Weight commonly returns after discontinuation, and the long-term health consequences of that regain cycle are not fully characterized.

This page is educational and reflects trials and the FDA label as of 2026-07-17. It is not medical advice — long-term decisions belong with your prescriber.

Long-term treatment needs ongoing care

Common questions about Ozempic long-term side effects

What are the long-term side effects of Ozempic?

The longest controlled studies of semaglutide are STEP 5 (2 years) and SELECT (mean about 3.3 years). They found no new long-term safety signals beyond what is already known: mostly gastrointestinal effects (which usually ease over time), a higher rate of gallbladder problems tied to rapid weight loss, and loss of some lean/muscle mass along with fat. Serious events remained rare. Importantly, controlled data do not yet extend much beyond about four years, so effects over 5+ years are genuinely not yet established.

Does long-term Ozempic use cause muscle loss?

Some lean-mass loss accompanies large weight loss on any method. In the STEP 1 68-week DXA substudy, total lean mass fell about 9.7% while fat fell much more, so the proportion of lean mass to body weight actually rose. DXA "lean mass" also includes water, organs and bone, so it overstates true muscle loss — but the concern is real over time. Adequate protein and regular resistance training are the standard, evidence-supported ways to protect muscle. See our GLP-1 and muscle-loss guide for the trial-by-trial detail.

Does Ozempic cause pancreatic cancer?

There is no confirmed causal link between semaglutide and pancreatic cancer. Meta-analyses of the available trials and studies have not found a statistically significant association. A separate issue is acute pancreatitis (inflammation, not cancer): it is a labeled warning, and while some analyses show a small signal, it is uncertain and weakens on closer analysis. The honest summary: pancreatic-cancer risk is not established; pancreatitis is a rare, labeled risk to watch for with severe, persistent abdominal pain.

What about gallbladder problems?

Gallbladder disease is more common with GLP-1 medications — a meta-analysis (JAMA Internal Medicine, 2022) estimated roughly 1.37 times the risk versus placebo, and in the STEP obesity trials gallbladder disorders ran a couple of percent higher than placebo. The main driver is rapid weight loss itself, which changes bile composition and raises gallstone risk. Upper-right abdominal pain, fever, or yellowing of the skin warrant prompt evaluation.

Is it safe to stay on Ozempic for years?

For chronic conditions like type 2 diabetes and obesity, it is generally intended as a long-term treatment, and the multi-year trials are reassuring within their time frames (up to about 3.3 years in SELECT). But controlled data do not yet extend far beyond that, so truly long-horizon effects are not fully known. Also, weight tends to return after stopping, which is why the decision to continue is an ongoing conversation with your prescriber rather than a fixed endpoint.

What long-term effects are still unknown?

Honestly, several. There is little controlled safety data beyond about four years of continuous use, so effects over 5-10+ years are not established. Long-term outcomes after stopping (including the health impact of weight regain) are not fully characterized. And it is not established as safe in pregnancy — animal studies show harm, and the label advises stopping about two months before a planned pregnancy. Ongoing trials will fill some of these gaps over the coming years.

Sources

Trial results from the primary Nature Medicine and NEJM publications; gallbladder estimate from JAMA Internal Medicine; label warnings from the FDA-approved Ozempic prescribing information via DailyMed. Reviewed 2026-07-17. Educational only — not medical advice.

References(5)

Our sourcing standards →

Two-year effect of semaglutide 2.4 mg on body weight (STEP 5, Garvey et al.) — Nature Medicine (2022)
Semaglutide and Cardiovascular Outcomes in Obesity without Diabetes (SELECT, ~3.3-year follow-up, Lincoff et al.)NEJM — New England Journal of Medicine (2023)
Once-Weekly Semaglutide in Adults with Overweight or Obesity (STEP 1, 68-week DXA substudy for body composition)NEJM — New England Journal of Medicine (Wilding et al.) (2021)
GLP-1 receptor agonists and risk of gallbladder and biliary diseases — meta-analysisJAMA — JAMA Internal Medicine (2022)

How long is “long-term,” really?

This is the part most articles skip. The strongest long-term evidence for semaglutide comes from two trials:

STEP 5 (Nature Medicine, 2022) — a 2-year randomized trial of semaglutide 2.4 mg for weight management. It reported sustained weight loss (about −15% vs −2.6% on placebo) and, importantly, no new safety signals over two years; side effects were mostly the familiar mild-to-moderate GI ones.

SELECT (NEJM, 2023) — over 17,000 adults followed for a mean of about 3.3 years, the longest safety exposure for semaglutide 2.4 mg. More GI-related discontinuations and more gallbladder disorders, but no excess of serious adverse events — alongside a 20% reduction in cardiovascular events.

So “long-term” today means up to roughly three to four yearsof good controlled data. That's genuinely reassuring within that window — and it's also the honest boundary of what's known.

The real long-term considerations

1. Lean and muscle mass

2. Gallbladder disease

3. Gastrointestinal effects over time

4. The thyroid boxed warning (context, not a long-term finding)

The pancreatic-cancer question

What is honestly still unknown

Effects beyond ~4 years

Little controlled safety data exists past about four years of continuous use, so 5-10+ year effects are not established. Ongoing trials will extend the record.

Pregnancy

Not established as safe; animal studies show reproductive harm. The label advises discontinuing about two months before a planned pregnancy because of the long half-life.

After you stop

Weight commonly returns after discontinuation, and the long-term health consequences of that regain cycle are not fully characterized.

This page is educational and reflects trials and the FDA label as of 2026-07-17. It is not medical advice — long-term decisions belong with your prescriber.