Weight Regain After Stopping Ozempic: What the Research Actually Shows

By week 20 of the STEP 4 trial, every participant had been on semaglutide for five months. They had lost, on average, about 11% of their starting body weight. Then half of them — chosen at random — were quietly switched to placebo injections that looked identical to the medication. The other half stayed on the drug.

A year later the two groups had diverged by nearly 20 percentage points of body weight. The placebo arm had regained most of what it had lost. The semaglutide arm had kept going.

What STEP 4 actually measured

The STEP 4 trial, led by obesity specialist Thomas Wadden at the University of Pennsylvania and published in JAMA in 2021, was designed to answer one specific question: what happens to people who stop semaglutide after they have already responded to it?

The protocol was deliberate. All 803 participants ran through a 20-week titration phase on weekly subcutaneous semaglutide, working up to the 2.4 mg maintenance dose. At week 20, those who had tolerated the drug were randomised either to continue semaglutide for another 48 weeks or to switch to a matching placebo. Both groups received the same lifestyle counselling throughout.

From week 20 onwards, the continuation group lost an additional 7.9% of body weight. The withdrawal group regained 6.9%. The net difference at week 68 was 14.8 percentage points. About two-thirds of the weight lost during the open-label phase came back in the year after switching to placebo.

The cardiometabolic markers tracked the weight. Blood pressure, lipid profiles, glycaemic indices, and physical function scores improved while patients were on semaglutide and reverted, in parallel with the regained weight, after withdrawal.

SURMOUNT-4 told the same story with tirzepatide

Three years later, Louis Aronne at Weill Cornell led a near-identical trial design with tirzepatide — the dual GIP/GLP-1 agonist marketed as Zepbound. SURMOUNT-4 ran a 36-week open-label lead-in, then randomised participants to continue tirzepatide or switch to placebo for 52 weeks. Continuation produced an additional 5.5% loss. Withdrawal produced a 14% regain. The mechanism was different. The pattern was the same.

Why the regain isn't surprising

The biology that drives weight regain after dieting — elevated ghrelin, suppressed leptin and PYY, reduced resting metabolic rate, persistent food preoccupation — does not disappear when a person loses weight on a GLP-1. The drug works by overlaying the dysregulated system with sustained appetite suppression, slowed gastric emptying, and altered reward signalling in the hypothalamus and brainstem. Stop the drug, and the underlying biology reasserts itself.

Priya Sumithran's NEJM paper from 2011 had already mapped this terrain in dieters. Hunger and satiety hormones remained shifted in the direction of regain for at least a year after a weight-loss programme ended. STEP 4 demonstrated that semaglutide, while taken, holds those forces in check; SURMOUNT-4 showed the same for tirzepatide. Neither trial showed any evidence that the medications cure the underlying dysregulation.

That framing is consistent with how obesity medicine specialists describe the condition. The biology that fights weight loss is not switched off by a year of successful treatment; it is held off. When the holding action ends, the pressure returns.

What patients describe in the months after stopping

Quality-of-life data from the STEP 4 extension and from observational follow-up cohorts paint a fairly consistent picture of what discontinuation feels like. Patients describe hunger climbing within two to six weeks of the last dose — sooner for shorter half-life agents, more gradually for semaglutide given its roughly week-long half-life and slow washout. Food preoccupation tends to return on a similar timeline.

The reward dimension is what many describe as most striking. Foods that had felt neutral or uninteresting on semaglutide begin to draw attention again. Several qualitative studies, including patient-reported outcomes published alongside the STEP programme by John Blundell's group at the University of Leeds, have documented this attenuation of food reward as a distinct feature of treatment — and its reversal after stopping is equally distinctive.

Not everyone regains at the same rate. The STEP 4 distribution had a wide spread. A minority maintained most of their loss for the duration of the trial. Several extension analyses have tried to identify predictors — pre-treatment BMI, peak weight loss, age, baseline insulin resistance — without finding a single clean answer. Individual variability is real, but on the population level, regain is the default trajectory.

What lifestyle anchors do at the margin

The STEP 4 protocol included structured lifestyle counselling for both arms. The placebo group received the same dietary and physical activity guidance as the continuation group. They still regained. This is the cleanest available evidence that lifestyle intervention alone — applied at trial-level intensity — does not prevent regain after GLP-1 withdrawal.

That does not mean lifestyle is irrelevant. The pace and magnitude of regain in observational cohorts vary with a few measurable factors. Protein intake at or above 1.4–1.6 g/kg of reference body weight protects lean mass, which protects resting metabolic rate. Resistance training two to three times per week has a similar effect. Stuart Phillips at McMaster University has spent decades documenting these levers in the muscle physiology literature. Adequate sleep modulates ghrelin and leptin signalling. None of this is sufficient to prevent regain on its own, but each adjusts the slope.

For patients who do plan to discontinue, the more reasonable framing is harm reduction rather than maintenance. A slower, more graceful regain that preserves some of the metabolic and cardiovascular benefit may be the realistic target. Strategies for discontinuing more safely are an active area of clinical interest, though the evidence base remains thin compared to the data on continuation.

Maintenance dosing as an alternative to stopping

Between "full dose indefinitely" and "stop entirely" sits a less-explored middle: lower-dose maintenance. The 2.4 mg weekly dose of semaglutide was selected for the obesity indication because it produced the largest weight loss in dose-finding studies. For maintenance — once a patient has reached their target — lower doses may suffice to hold weight without the same intensity of appetite suppression.

The evidence for sub-2.4 mg maintenance is observational rather than randomised. Endocrinologists working in obesity medicine, including W. Timothy Garvey at the University of Alabama at Birmingham, have described clinical experience with 1.0 mg or 1.7 mg maintenance regimens in patients who plateau on the higher dose. Some patients hold their weight; others gradually drift upward. The STEP 5 trial extension data, which followed patients on the 2.4 mg dose to 104 weeks, suggested the full dose continues to produce benefit at year two — but no equivalent trial has formally tested dose reduction as a planned strategy.

The framing that emerges from STEP 4, SURMOUNT-4, and the surrounding evidence base is the chronic disease model. Hypertension medications lower blood pressure while taken; when stopped, blood pressure rises. Statins lower LDL while taken; when stopped, LDL drifts back. GLP-1 receptor agonists reduce weight while taken; when stopped, weight returns. The pattern is not a failure of the drugs. It is a feature of how chronic biological dysregulation responds to interruption of treatment.

Key takeaways

• STEP 4 randomised patients who had lost weight on semaglutide to continue or switch to placebo; the placebo arm regained 6.9% while the continuation arm lost an additional 7.9% over the next 48 weeks.
• SURMOUNT-4 showed an essentially identical pattern with tirzepatide: 14% regain after withdrawal versus continued loss on the drug.
• Cardiometabolic improvements — blood pressure, lipids, glycaemic markers — track weight and revert with regain after discontinuation.
• The biology driving regain (ghrelin elevation, leptin suppression, reduced metabolic rate) is held in check by GLP-1 therapy rather than cured by it.
• Lifestyle counselling at trial intensity did not prevent regain in STEP 4, though protein adequacy, resistance training, and sleep modulate the slope.
• The chronic disease framing — ongoing treatment for an ongoing condition — is increasingly the clinical default in obesity medicine.