Semaglutide has substantial evidence for weight management, but “semaglutide works” is an incomplete answer. The useful questions are which product was tested, who joined the study, what outcome changed and whether the result applies to the approved formulation being discussed. This guide brings those questions together rather than treating every semaglutide brand or trial percentage as interchangeable.
Start with the product, then the molecule
Semaglutide is a GLP-1 receptor agonist, the active ingredient rather than a single prescription product. The current EMA Wegovy record describes weight management with weekly injection and daily tablets; the tablets are intended for adults, while the injection also has an adolescent indication. Ozempic’s EMA record describes treatment of type 2 diabetes. Sharing an ingredient does not give the two brands identical authorised uses.
There are differences within a brand as well. The US Wegovy prescribing information distinguishes injection from tablets, and its injection-only liver-disease indication has a separate evidence and approval basis. FDA’s 19 March 2026 approval letter added a higher-dose Wegovy injection for adult weight reduction and maintenance. It did not change what dose an older trial tested. These are regulatory facts, not instructions to select a formulation or increase a dose.
For the practical brand distinction, read Wegovy versus Ozempic. The semaglutide tracker entry records the molecule; this article explains the evidence behind it.
A map of the pivotal evidence
The table deliberately keeps weight-loss and cardiovascular trials in different rows. Both are relevant, but they answer different questions. The trial doses identify the research intervention and are not a personal treatment recommendation.
| Study | Participants and comparison | Time and analysis | Main result |
|---|---|---|---|
| STEP 1 | 1,961 adults with obesity, or overweight plus a weight-related condition; no diabetes. 2.4 mg weekly injection versus placebo; lifestyle support in both groups. | 68 weeks; primary estimand includes discontinuation and rescue interventions. | Mean weight change: −14.9% versus −2.4%. |
| STEP 2 | 1,210 adults with type 2 diabetes and overweight or obesity; the trial included 2.4 mg, 1.0 mg and placebo groups with lifestyle support. | 68 weeks; primary intention-to-treat analysis of 2.4 mg versus placebo. | Mean weight change: −9.6% versus −3.4%. |
| SELECT | 17,604 adults aged at least 45, with established cardiovascular disease and overweight or obesity, without diabetes; 2.4 mg weekly injection versus placebo. | Mean follow-up 39.8 months; time-to-first cardiovascular-event analysis. | Cardiovascular death, nonfatal heart attack or nonfatal stroke: 6.5% versus 8.0%; hazard ratio 0.80. |
These were sponsor-funded trials. Randomisation strengthens the comparison within each trial, but it does not make a comparison between separate trials equally reliable. Different eligibility rules, baseline risks and treatment persistence remain important.
How to read a weight-loss percentage without overpromising
STEP 1 is the source of the widely repeated “around 15%” result. That was a group average for a particular injectable regimen over more than a year. It was not a minimum result, and it cannot be attached to every semaglutide formulation. The smaller average in STEP 2 illustrates why diabetes status belongs beside the number; it does not establish that diabetes alone explains the entire difference between studies.
An average also hides the range of individual responses. Ask whether a headline reports the whole randomised group, an estimate assuming continued treatment or only those who finished. The answer affects what the number represents. Our guide to reading weight-loss trials explains these statistical choices without turning them into a drug-ranking score.
Weight reduction is one outcome. A claim about reduced cardiovascular events needs an outcomes trial and the right population. SELECT supplied such evidence in people with existing cardiovascular disease; it did not test primary prevention in every adult hoping to lose weight. Its reported event proportions differ by 1.5 percentage points, while the hazard ratio describes a relative time-to-event comparison. Those are different descriptions of the study, not interchangeable estimates of a reader’s personal benefit.
What happens when treatment stops?
The STEP 1 extension followed 327 participants after the trial medication and lifestyle intervention ended. Over the following year, participants previously receiving semaglutide regained roughly two-thirds of their earlier weight loss, on average. This was an exploratory subset rather than the full randomised population.
The important limitation is easy to miss: trial lifestyle support stopped too. The extension cannot cleanly isolate the effect of stopping medication from every other change after the main study. It nevertheless makes maintenance planning a serious question. A decision to begin treatment should include a discussion of affordability, follow-up and what happens if it becomes unavailable, rather than assuming the end of weight loss is the end of care.
Safety requires more than a nausea checklist
The US label warns about thyroid C-cell tumours observed in rodents; the relevance to humans remains unknown. Wegovy is contraindicated with a personal or family history of medullary thyroid carcinoma or MEN 2. Its warnings also cover pancreatitis, gallbladder problems, severe gastrointestinal reactions, dehydration-related kidney injury and aspiration around anaesthesia. Other medicines, pregnancy plans and retinopathy history can matter. This is a summary, not the complete safety information; consult the current product label and a qualified clinician.
Trial averages cannot decide whether a particular symptom is harmless or whether a person should continue treatment. Nor does reading about one product support converting doses between tablets, injections or compounded preparations. The prescription, formulation and clinician’s monitoring plan need to remain together.
What this evidence can and cannot answer
- Well supported: weight reduction with the studied regimens in the enrolled populations; cardiovascular benefit in SELECT’s secondary-prevention population.
- Requires separate evidence: each newer formulation, higher dose, additional indication or population. An updated label is not permission to reuse an older percentage without qualification.
- Not predicted by a group mean: an individual’s final weight, tolerability, treatment persistence, coverage or out-of-pocket cost.
For a comparison with a different medicine, use semaglutide versus tirzepatide and check whether the evidence is head-to-head. Newer research, such as the REMODEL kidney report, should be interpreted by its own endpoints rather than folded into a single claim that semaglutide improves every health outcome.
Questions for a useful clinical conversation
- Which approved product, formulation and indication are we discussing?
- Which trial population most closely matches the clinical question?
- How will benefits, adverse effects and nutrition be monitored?
- What is the maintenance plan if cost, access or tolerability changes?
- Which changes would require contacting the prescriber?
These questions help organise care without selecting a treatment. The fuller before-starting-medication guide covers preparation. This publication provides educational information and does not diagnose, prescribe or replace individual medical assessment.



