Is Ozempic Safe Long-Term? The Honest Answer From the Current Evidence Base

Trust signals

> Reviewed by FormBlends Medical Team · Last updated May 2026 · 12 sources cited · Topic: long-term safety profile of semaglutide

Key Takeaways

• Within the 2-4 year randomized trial window, semaglutide has a favorable safety profile in eligible patients
• SELECT (3.3-year follow-up, 17,604 patients) showed net cardiovascular benefit and no major unexpected safety signals
• SUSTAIN-6 (2.1-year follow-up in diabetic patients with CV risk) showed 26% MACE reduction
• Known long-term concerns include GI tolerability, gallstones during weight loss, lean-mass loss, and ongoing surveillance for rare events including NAION and thyroid cancer
• The drug has been in widespread U.S. use since 2017-2018; multi-decade safety remains under active characterization

Direct answer

Within the 2-4 year evidence window that constitutes "long-term" for this medication, semaglutide has a favorable safety profile in eligible patients. The major cardiovascular outcome trials (SELECT for obesity, SUSTAIN-6 for diabetes) showed net benefit with adverse-event patterns dominated by GI tolerability and gallbladder events rather than catastrophic outcomes. The known concerns are real but mostly manageable. The unknown questions (multi-decade cancer profile, very long-latency rare events) require longer cohort follow-up that is accumulating now. The current evidence does not support avoiding the medication on long-term safety grounds for patients who meet indications and tolerate it well.

Table of contents

1. How long is "long-term" for Ozempic?
2. What the major trials showed about safety
3. The established long-term concerns
4. The under-surveillance concerns
5. Mortality outcomes in long-term trials
6. How Ozempic compares to other long-term diabetes therapies
7. What happens after stopping
8. Decision framework for long-term safety questions
9. The contrary view: limits of current safety data
10. FAQ
11. Sources

How long is "long-term" for Ozempic?

The phrase "long-term" gets used loosely. For Ozempic specifically:

• FDA approval was December 2017
• Longest randomized trial follow-up: SELECT, 3.3 years (mean)
• Longest weight-management trial: STEP 5, 2 years
• Diabetes cardiovascular outcomes: SUSTAIN-6, 2.1 years
• Renal outcomes: FLOW, median 3.4 years
• Real-world early adopters: now reaching 8-9 years of continuous exposure

This is short by chronic-disease epidemiology standards but reasonable for a medication that received its first U.S. approval less than a decade ago. The horizon will extend as cohorts mature.

For most clinical questions, the 2-4 year evidence supports answers. For multi-decade questions, the answers are still being formed.

What the major trials showed about safety

SELECT (NEJM 2023). 17,604 adults with obesity and established cardiovascular disease, randomized to semaglutide 2.4 mg or placebo, mean follow-up 3.3 years. Primary outcome reduced by 20%. Adverse events more common in semaglutide arm: GI tolerability (more nausea, vomiting, diarrhea, constipation; more discontinuation for these reasons), gallbladder events (more cholelithiasis). Adverse events not differentially increased: thyroid cancers (rare in both arms, not statistically different), pancreatic cancer (rare in both arms), pancreatitis (uncommon, similar rates).

SUSTAIN-6 (NEJM 2016). 3,297 patients with type 2 diabetes and high cardiovascular risk, randomized to semaglutide or placebo, median 2.1 years. MACE reduced 26%. Adverse events: GI tolerability and small increase in diabetic retinopathy complications in patients with pre-existing retinopathy (an early-worsening phenomenon, not unique to semaglutide).

STEP 5 (Nature Medicine 2022). 304 patients with obesity randomized to semaglutide 2.4 mg or placebo for 104 weeks. Sustained weight loss of about 15% on treatment. Adverse events consistent with the broader semaglutide profile.

Across these and other trials, the long-term safety pattern is consistent: GI tolerability dominates the adverse-event picture, gallbladder events appear during rapid weight loss, rare events occur at low frequencies, and the overall risk-benefit favors continued therapy in eligible patients.

The established long-term concerns

The most consistent long-term issues:

• GI tolerability persists in a subset of patients; many improve over months but some discontinue
• Gallstones develop in a meaningful minority of patients during rapid weight loss; surgical intervention is occasionally required
• Acute pancreatitis is uncommon but documented; patients with prior pancreatitis are at elevated risk
• Lean-mass loss (~25-40% of total weight loss) affects body composition; resistance training mitigates
• Cosmetic changes (facial volume loss with significant weight reduction) are universal in proportion to weight loss
• Hair shedding during rapid weight loss is typically transient
• Hypoglycemia in diabetic patients on insulin or sulfonylureas requires dose adjustment of those medications

None of these is catastrophic. All are manageable with appropriate clinical attention.

The under-surveillance concerns

Several rare events are under active pharmacovigilance:

• NAION: the 2024 JAMA Ophthalmology paper from Mass Eye and Ear reported roughly 4x increased hazard in semaglutide users vs comparators. Absolute risk remains small. Regulatory review is ongoing.
• Thyroid C-cell tumors: rodent finding underlies the boxed warning. Human population studies have not confirmed a measurable increase at clinical doses. Surveillance continues.
• Pancreatic cancer: early signal raised in 2013 has not held up under repeated rigorous study.
• Severe gastroparesis: some case reports suggest more severe and prolonged delayed gastric emptying in a subset of patients. Mechanism may be exaggerated normal action.
• Suicidal ideation: EMA reviewed early signals and concluded no causal association supported by current data. Continued monitoring under post-marketing surveillance.

Patients should understand "under surveillance" as a legitimate ongoing category, not as either "confirmed risk" or "ruled out." The honest framing is that these questions remain open and conclusions will sharpen with more data.

Mortality outcomes in long-term trials

SELECT and SUSTAIN-6 both showed reduced cardiovascular mortality on semaglutide compared to placebo. All-cause mortality has trended favorably in pooled analyses across the semaglutide development program.

Specific findings:

• SELECT: cardiovascular death reduced (HR ~0.85 in secondary analysis); all-cause death also reduced
• SUSTAIN-6: numerical reductions in cardiovascular and all-cause mortality, though not all reached individual statistical significance
• Pooled semaglutide trial analyses: consistent favorable trends for mortality outcomes

The strongest single statement supported by current evidence is that semaglutide does not increase mortality in eligible patients and likely reduces it. This is among the most important elements of the long-term safety profile because mortality is the ultimate hard endpoint.

How Ozempic compares to other long-term diabetes therapies

Comparative long-term safety:

• vs older sulfonylureas: semaglutide produces better cardiovascular outcomes with lower hypoglycemia risk
• vs insulin: semaglutide produces weight loss rather than weight gain and has comparable or better glycemic effects
• vs DPP-4 inhibitors: semaglutide produces stronger glycemic effect, weight loss, and cardiovascular benefit
• vs SGLT2 inhibitors: both classes are guideline-preferred for high-cardiovascular-risk diabetes; head-to-head trials are limited; combination is increasingly used
• vs older GLP-1 agents (liraglutide, exenatide): semaglutide is generally more potent for glycemic and weight effects

The comparative long-term safety story has generally favored semaglutide on cardiovascular outcomes, weight outcomes, and hypoglycemia profile. The cancer and rare-event profiles compared head-to-head are less clearly characterized at the multi-year horizon.

What happens after stopping

Long-term safety includes what happens after a patient stops the medication. The pattern:

• Glycemic control deteriorates in diabetic patients within weeks to months
• Weight regain begins gradually and accelerates; STEP 1 extension data showed regain of about two-thirds of lost weight within 1 year
• Blood pressure, lipid, and inflammatory marker improvements partly reverse
• Cardiovascular event risk likely returns toward pre-treatment levels over time
• Pharmacologic effects of semaglutide on the body resolve over weeks to months; no chronic residual exposure

This pattern reflects the chronic nature of the underlying conditions (obesity, type 2 diabetes). Discontinuation is appropriate in some situations (intolerable side effects, pregnancy, contraindication) but typically results in regression of the gained benefit.

Decision framework for long-term safety questions

Continuing therapy is reasonable if:

• You meet FDA indications and have no contraindications
• You tolerate the medication adequately
• You are achieving meaningful benefit (glycemic control, weight management, cardiovascular risk reduction)
• Routine monitoring (HbA1c, blood pressure, lipid, renal function, eye exams as risk-stratified) is in place
• You and your prescriber review the risk-benefit annually

Stopping or switching warrants discussion if:

• Persistent severe side effects despite dose adjustment
• Development of a contraindication (new MTC diagnosis, etc.)
• Pregnancy or pregnancy planning
• Apparent loss of effect that does not respond to titration
• Significant adverse event under surveillance applies to your situation (NAION diagnosis, etc.)

Returning to the question annually:

• The evidence base is still maturing; statements that were uncertain in 2026 will be clearer in 2030
• Personal risk-benefit can change as you age and as health conditions evolve
• New therapies may emerge that change the relative positioning of semaglutide

The contrary view: limits of current safety data

The strongest arguments for greater caution than current data alone suggest:

The horizon is still short. Three to five years of randomized follow-up cannot detect events with longer latency. Multi-decade cancer risk, slow bone density changes, and very rare events may emerge only with more time.

Trial populations differ from real-world populations. SELECT and SUSTAIN-6 enrolled patients with specific clinical profiles. Patients outside those profiles (very elderly, very high BMI, multiple severe comorbidities, ongoing substance use, eating disorders) may have different safety experiences.

Adherence in trials exceeds real-world adherence. Side-effect profiles look milder in trials because patients with severe intolerance often discontinue and are not included in long-term safety analyses. The "average tolerator" picture is partly a selection effect.

Pharmacovigilance signals continue to emerge. NAION was not on the radar before 2024. The next decade will likely produce additional signals that are not currently visible. This is a feature of all medications, not specific to semaglutide.

These limitations argue for continued surveillance and individualized risk-benefit assessment, not for avoiding the medication. Within the current evidence base, the long-term safety picture is favorable for eligible patients.

FAQ

Is Ozempic safe long-term?

Within the available 2-4 year evidence base, yes, in eligible patients. Multi-decade safety remains under study.

How long has Ozempic been studied?

Approved December 2017. Longest randomized trial follow-up is SELECT at 3.3 years. Real-world cohorts now reach 8-9 years.

What does the SELECT trial say about long-term safety?

17,604 adults followed mean 3.3 years showed 20% reduction in MACE with adverse events dominated by GI tolerability and gallbladder events.

What are the main long-term safety concerns?

GI tolerability, gallstones, lean-mass loss, NAION (under surveillance), and the boxed warning about thyroid C-cell tumors.

Has long-term Ozempic use caused any deaths in trials?

Cardiovascular outcome trials showed reduced cardiovascular mortality. All-cause mortality trends favorably.

Is Ozempic safer than other diabetes medications long-term?

Comparatively favorable in major outcome trials, especially vs older sulfonylureas. Head-to-head with SGLT2 inhibitors is more nuanced.

What happens if I stop Ozempic after years of use?

Glycemic control deteriorates in diabetic patients; weight regain is common. The cardiovascular and renal benefits attenuate.

Are there long-term effects on muscle or bone?

Lean-mass loss is part of weight reduction. Long-term bone implications are still being characterized.

Should I worry about long-term cancer risk?

Current evidence does not support a clear increase. Surveillance continues for long-latency effects.

What about long-term mental health effects?

Suicidal ideation signals were reviewed by the EMA in 2024 and not confirmed as causally linked. Mental health surveillance continues.

Will I need monitoring tests during long-term use?

Routine diabetes monitoring if you have diabetes; periodic blood pressure and lipid checks; eye care appropriate to your risk profile; symptom-driven evaluation for any new concerns.

Related guides

• Long-Term Effects of Wegovy: A Honest Read of STEP 5, SELECT, and What We Don't Yet Know
• Does Zepbound Cause Cancer? Reading the Boxed Warning Against the Tirzepatide Evidence Base
• Why Zepbound Kills Alcohol Cravings: Reward Circuits, GLP-1 Signaling, and Honest Limits of the Evidence
• What to Eat on Ozempic: The Long-Term Dietary Framework
• Long-Term Effects of Ozempic: SUSTAIN-6, Real-World Diabetes Outcomes, and the Open Questions
• Is There an Ozempic Shortage in 2026? The Current Supply Picture

Sources

1. Lincoff AM et al. Semaglutide and Cardiovascular Outcomes in Obesity Without Diabetes (SELECT). New England Journal of Medicine. 2023.
2. Marso SP et al. Semaglutide and Cardiovascular Outcomes in Patients with Type 2 Diabetes (SUSTAIN-6). New England Journal of Medicine. 2016.
3. Garvey WT et al. Two-Year Effects of Semaglutide in Adults with Overweight or Obesity (STEP 5). Nature Medicine. 2022.
4. Wilding JPH et al. Once-Weekly Semaglutide in Adults with Overweight or Obesity (STEP 1). New England Journal of Medicine. 2021.
5. Perkovic V et al. Effects of Semaglutide on Chronic Kidney Disease in Patients with Type 2 Diabetes (FLOW). New England Journal of Medicine. 2024.
6. Rubino D et al. Effect of Continued Weekly Subcutaneous Semaglutide on Weight Loss Maintenance (STEP 4). JAMA. 2021.
7. Wang Y et al. Glucagon-Like Peptide-1 Receptor Agonist Use and Risk of Thyroid Cancer. JAMA. 2024.
8. Hathaway JT et al. Risk of Nonarteritic Anterior Ischemic Optic Neuropathy in Patients Prescribed Semaglutide. JAMA Ophthalmology. 2024.
9. European Medicines Agency. Review of GLP-1 Receptor Agonists and Suicidal Thoughts. 2024.
10. American Diabetes Association. Standards of Care in Diabetes 2025.
11. FDA. Ozempic and Wegovy Prescribing Information. 2024 updates.
12. Bain SC et al. Worsening of Diabetic Retinopathy with Rapid Improvement in Systemic Glucose Control: A Review. Diabetes, Obesity and Metabolism. 2019.

Footer disclaimers

Platform Disclaimer. FormBlends is a telehealth platform connecting patients with independent licensed clinicians and U.S. pharmacies. We do not deliver direct medical care. Treatment decisions belong to you and your provider.

Compounded Medication Notice. Compounded semaglutide is not FDA-approved. It is prepared by state-licensed 503A pharmacies in response to individual prescriptions and is not interchangeable with branded Ozempic or Wegovy.

Results Disclaimer. Safety statements summarize current evidence from clinical trials, observational studies, and pharmacovigilance available as of May 2026. Longer-term findings continue to accumulate. Your individual safety experience may differ from population averages.

Trademark Notice. Ozempic and Wegovy are registered trademarks of Novo Nordisk A/S. Mounjaro and Zepbound are registered trademarks of Eli Lilly and Company. FormBlends has no affiliation with, endorsement from, or sponsorship by Novo Nordisk, Eli Lilly, the FDA, the EMA, or any other entity referenced in this article.