What Studies Show About Weight Regain After Stopping a GLP-1

• written by Perjan Duro

There is a question people tend to ask quietly, often after a cost change, a supply gap, a side effect they could not live with, or a conversation about whether they still need to be on treatment:

What happens if I stop?

It deserves a straight answer rather than a reassuring one. The research on this is unusually consistent, and reading it honestly is more useful than hoping it says something else.

Quick answer

Across randomised trial extensions and pooled analyses, most people regain a substantial share of the weight they lost after stopping a GLP-1 medication. In the STEP 1 extension, participants regained about two-thirds of their prior loss within a year.

Researchers describe this as the expected biology of a chronic condition reasserting itself — not as a lapse in effort. The authors of the largest pooled analysis of discontinuation studies say regain occurred regardless of lifestyle interventions.

What the evidence does not do is tell you how long you personally should be on treatment. That is a clinical decision, and it belongs to you and your prescriber.

What "stopping" looks like in practice

Before the trial numbers, some context that reframes the whole topic: stopping is common, and it is frequently not a choice.

Samuels and colleagues followed 2,306 patients in an academic obesity clinic and reported that roughly half had discontinued by 12 months. Other real-world work consistently finds side effects and cost among the leading reasons people stop.

That matters for how this article should be read. If you stopped, are stopping, or are weighing it up, you are in an extremely large group, and for many people in that group the decision was made by a pharmacy price, a supply shortage, an insurer, or a body that could not tolerate the medication. None of that is a character finding.

The STEP 1 extension: the clearest single dataset

STEP 1 is the trial behind the widely quoted semaglutide numbers. Wilding and colleagues randomised 1,961 adults to 68 weeks of once-weekly semaglutide 2.4 mg or placebo, both as an adjunct to a structured lifestyle intervention, and reported a mean change of −14.9% body weight versus −2.4% with placebo.

The more revealing study is what happened next.

At week 68, treatment was withdrawn — and so was the lifestyle intervention. A representative subset of 327 completers was then followed for another full year, off everything. In that subset:

  • mean loss from week 0 to week 68 was 17.3% with semaglutide and 2.0% with placebo;
  • by week 120, the semaglutide group had regained 11.6 percentage points of lost weight, and the placebo group 1.9 points;
  • that left net changes from baseline of −5.6% and −0.1% respectively;
  • most of the cardiometabolic improvements seen at week 68 — the blood-pressure and lipid changes — also drifted back toward baseline.

The authors summarised it as regaining roughly two-thirds of prior weight loss in one year, and wrote that the findings confirm the chronicity of obesity.

Two caveats worth holding onto. The extension analyses were exploratory, run on a subset rather than the full randomised population, so they carry less inferential weight than the main trial result. And STEP 1 and its extension were funded by Novo Nordisk, the manufacturer — normal for pivotal drug programmes, and a fact to read alongside the data rather than a reason to dismiss it.

The pooled picture across drugs

The STEP 1 extension is one trial. Berg and colleagues went wider, with a systematic review and meta-analysis of discontinuation across 8 randomised controlled trials and 2,372 participants, all with a BMI of 27 or above, searching the literature through February 2024.

Their pooled findings:

  • after stopping liraglutide, participants regained a mean of 2.20 kg (95% CI 1.69 to 2.70);
  • after stopping semaglutide or tirzepatide, participants regained a mean of 9.69 kg (95% CI 5.78 to 13.60);
  • across studies, weight regain was proportional to the original weight loss.

That last point is the one most worth sitting with. The medications associated with the largest losses were also associated with the largest regains. Under this reading, a big regain is not evidence that something went wrong — it is closer to the arithmetic consequence of a big loss, with the same biology running in reverse.

The Berg authors' own stated conclusion is that regain follows discontinuation "regardless of lifestyle interventions," and that GLP-1 therapy should therefore be considered a chronic treatment. That is their interpretation of their data, and it reflects where much of the obesity-medicine field has moved. It is not a recommendation this article can make for you, and it is not the only consideration in a real prescribing decision — tolerability, cost, comorbidities, pregnancy plans, and your own preferences all sit in that conversation too.

What the evidence does not settle

Being precise about the gaps is part of reading this fairly.

Whether continued support changes the trajectory. In the STEP 1 extension, the lifestyle programme was withdrawn along with the drug, so that study cannot isolate what ongoing structured support alone would have done. Berg and colleagues concluded that regain occurred regardless of lifestyle interventions across their included trials — a broader claim, drawn from pooled data with varying designs. The honest position is that the evidence points strongly toward biological regain being difficult to prevent, without a clean trial that answers "what if only the support continued?"

What happens beyond a year or two. Most discontinuation follow-up is short. Trajectories after that are not well characterised.

Whether tapering, intermittent use, or dose reduction behaves differently from a full stop. These are active research questions, not settled ones, and they are exactly the kind of thing to raise with a prescriber rather than to experiment with alone.

How much any of it applies to you. Trial means describe groups. Individual responses in these studies varied widely — the standard deviations around every figure above are large.

Why this is physiology, not failure

The mechanism researchers describe is not mysterious. Appetite regulation, satiety signalling, and energy expenditure adapt to weight loss and push back toward a defended level. A medication that acts on those signals changes them while it is present. When it is withdrawn, the underlying regulation returns — and so, for most people, does weight.

This is why the field increasingly talks about obesity the way it talks about hypertension or asthma: a chronic condition where stopping an effective treatment tends to be followed by the return of what the treatment was managing. Nobody frames a rising blood pressure after stopping an antihypertensive as a personal failing.

If you have regained weight after stopping, the studies above describe what happened to most participants in most trials. That is worth knowing not as a reason for fatalism, but as a correction to a story you may have been telling yourself.

Bringing this into a real conversation

Duration of treatment, tapering, restarting, pausing, switching — every one of those is a decision for you and a qualified clinician who knows your history. Nothing here is a reason to change what you are doing.

What you can do is make that conversation concrete rather than approximate. Appointments are short, and "I think it started going back up in the spring, maybe?" uses several of those minutes without answering much.

Questions people find useful to bring, with dates attached:

  • what my weight actually did over the months before and after any change in treatment;
  • which side effects occurred, when, and how they tracked with dose changes;
  • when there were gaps — supply, cost, travel, illness — and what followed them;
  • how appetite and food noise changed, and when;
  • what my own goals and constraints are, including financial ones, said out loud.

That last one matters. If cost is the reason a treatment decision is looming, saying so plainly is more productive than presenting it as a motivation problem.

The honest summary

Weight regain after stopping a GLP-1 is one of the most consistently reported findings in this literature. It has been observed in a placebo-controlled trial extension, and pooled across eight randomised trials, with regain scaling to the size of the original loss.

It is described by the researchers who measured it as the expected behaviour of a chronic condition. Reading it that way does not make any decision for you — but it does replace a private sense of failure with a documented, ordinary physiological finding, which is a much better starting point for a conversation with your clinician.

Velto implementation

The one thing that makes a duration conversation concrete is a continuous record: weight over months rather than a remembered number, symptoms next to the dose that was actually taken, and the treatment timeline including the gaps.

That is what Velto keeps in one place — a private log of treatments, symptoms, appetite, and weight over time, with Premium CSV export if you want to bring the whole history to an appointment.

Velto is a tracking companion. It does not treat, diagnose, or make dosing decisions, and no app can tell you how long to stay on a medication.

References

Related reading

Medical Disclaimer: Educational content only. Not medical advice. For treatment decisions, consult a licensed clinician.
Back to Blog

Articles you might like