Restarting GLP-1 medication after stopping and regaining is common, and not a failure of willpower: roughly a fifth to a half of people who stop start again within a year. When the STEP 1 extension stopped the medication overnight, about two-thirds of the lost weight returned within a year (Wilding 2022); individual results vary. The GetLean GLP-1 Taper Plan steps the dose down instead.
Every withdrawal trial took the medication away, by stopping it or by switching people to placebo, and in the STEP 1 extension the trial's lifestyle support ended the same day (Wilding 2022). That is what happened to the trial groups as a whole, not what separated some participants from others: the medication's effect on appetite fades once it is stopped, and the habits in place by then are what is left.
In Singapore, tirzepatide is registered as Mounjaro, and semaglutide injection as Wegovy for weight management and as Ozempic for type 2 diabetes. The figures in this article are given by molecule, as the trials and product labels report them.
Weight cycling: repeated loss and regain of body weight. In the GLP-1 context it usually means stopping, regaining and starting again — a pattern the trials were not designed to study.
How common is stopping and restarting?
Common enough to be the normal pattern rather than the exception.
In a cohort of 125,474 US adults newly starting liraglutide, semaglutide or tirzepatide, one-year discontinuation was 64.8% among those without type 2 diabetes and 46.5% among those with it. Of those who stopped and had a weight measurement at discontinuation, 36.3% without type 2 diabetes and 47.3% with it had restarted within the following year. Regaining weight was what pushed people back: every 1% of body weight regained after stopping was associated with a 2.8% higher hazard of restarting in those without type 2 diabetes, and 2.3% in those with it (Rodriguez 2025).
A second US cohort followed 7,938 adults who discontinued within three to twelve months. Over the following year 19.6% restarted the same medication, and a further 35.2% took up another obesity treatment — most often a different medication (Gasoyan 2026).
Both studies report the diabetes and non-diabetes figures separately because they differ substantially, and both are observational electronic-health-record studies set in a US insurance and access context where cost and coverage drive stopping in ways that do not transfer to Singapore.
Does a second course work as well as the first?
Nobody has measured it. That absence is the central fact of this topic.
The searches behind that statement covered PubMed for reinitiation, restart and retreatment crossed with semaglutide, tirzepatide and GLP-1, plus trial-extension papers and real-world cohorts. Reinitiation rates come back repeatedly; the weight outcome of a restart comes back nowhere. The discontinuation cohort above does not report weight change among those who reinitiated — its abstract does not contain it and its full text is paywalled, so it should not be assumed to (Gasoyan 2026).
The nearest human data is not a restart at all. In an Italian real-world cohort of 211 people with obesity and without type 2 diabetes, the 24% who came to semaglutide having previously been on a GLP-1 medication lost weight more slowly at first than those new to the class (−2.7 kg against −6.0 kg at the first follow-up, −6.5 kg against −10.5 kg at the second), and by twelve months the two groups were comparable (Milani 2026). The prior exposure there was liraglutide and the new drug semaglutide — a switch between molecules rather than a stop, regain and restart of the same one — in an uncontrolled single-country study with 55 people left at twelve months. It does not answer the question.
A direct experiment on cycling has been run in animals. In male mice with diet-induced obesity, semaglutide was given in two-week cycles alternating with two-week withdrawal periods, and the same biological system exists in humans. Restarting the drug did produce weight loss, but the cycled animals "failed to reach the prior weight nadir achieved during cycle 1", and they finished with more fat and a lower percentage of lean tissue than continuously treated controls — absolute lean mass was unchanged; the lean percentage fell because fat rose. The authors state that "validation in females and in human populations will be essential to determine translational relevance" (Son 2026). The cohorts were six to seven animals per arm, all male, and no equivalent study exists in people.
So a second course has not been shown to work less well in humans. It has not been shown to work as well either. Anyone stating a figure for how much a restart achieves is stating something that has not been measured.
Why rebound follows an abrupt stop
Because the withdrawal trials took the medication away, by stopping it or by switching people to placebo. The appetite effect fades over the weeks after the last dose, and the habits in place by then are what is left. That is biology, not a lack of willpower.
In the STEP 1 extension, semaglutide and the trial's lifestyle support both stopped at week 68. Participants had reached a mean 17.3% weight loss, and over the following year they regained 11.6 percentage points of it — about two-thirds — finishing 5.6% below their original starting weight (Wilding 2022).
That figure describes a trial group after an overnight stop, not any particular reader: it carries a wide spread, with a standard deviation of 7.7 percentage points. It is also the scenario a planned exit is built around. At GetLean the dose is stepped down at Dr Quek's monthly reviews rather than stopped overnight, and the habits are built with daily WhatsApp coaching while the medication is still working. Why two-thirds of the weight comes back goes through those numbers in detail.
What is the regained weight made of?
Nobody has measured that either, and it is worth knowing before anyone tells you otherwise.
Not one withdrawal trial or meta-analysis reports the fat-versus-lean split of the weight that returns. The largest pooled analysis covered body weight, waist circumference and BMI, and contains no fat-mass or lean-mass data at all (Berg 2025). The claim that it comes back as fat is an extrapolation from a different literature: twelve men refed after 24 weeks of severe semi-starvation, whose fat and fat-free mass recovered on separate trajectories (Dulloo 1997), and the related work on preferential catch-up fat in extreme catabolic states (Dulloo 2015). Those are far more extreme physiological conditions than medically supervised weight loss.
The case for protecting muscle does not need it. What has been measured is the composition of the weight that comes off: roughly 75% fat and 25% lean mass in the SURMOUNT-1 body-composition sub-study, in the placebo arm as much as in the drug arm (Look 2025). What regained weight is made of works through the whole question.
Is weight cycling itself harmful?
Genuinely contested — and the literature changed direction recently, which is the part usually left out.
The case for harm is observational and large. A systematic review and meta-analysis pooling 23 prospective cohort studies across 441,199 people found body-weight fluctuation associated with all-cause mortality (relative risk 1.41, 95% CI 1.27–1.57), cardiovascular mortality (1.36, 1.22–1.52), cardiovascular disease (1.49, 1.26–1.76) and hypertension (1.35, 1.14–1.61), with no association for cancer mortality (Zou 2019). These are cohort associations, not experiments. The abstract does not state how fluctuation was defined across the pooled studies, or whether the weight loss involved was intentional — and unintentional weight loss is precisely the confounder at issue.
The case against is a 2026 reappraisal. A Personal view in Lancet Diabetes & Endocrinology went back over the same body of work and concluded that "the current evidence does not support a causal link between weight cycling per se and clinical harm in people with obesity", attributing the reported associations to ageing, unintentional weight loss, reverse causality and longer cumulative exposure to obesity, and judging the benefits of intermittent weight reduction to outweigh the risks of fluctuation (Magkos 2026). A Personal view is an argued appraisal by two authors rather than a systematic review or a guideline, and both authors declare consultancy or research funding from pharmaceutical companies, including manufacturers of these medicines.
On GLP-1 cycling specifically, a 2026 narrative review in Nature Reviews Endocrinology states that repeated cycles of initiation, interruption and re-initiation "might induce body weight and HbA1c fluctuations — two risk factors for cardiovascular and microvascular events", while conceding in the same abstract that "few data are available regarding the incidence of hard outcomes in people discontinuing such drugs" (Ceriello 2026). Every clause of the concern is hedged, and the review says outright that the outcome data do not exist.
Three positions, then: a large observational literature pointing at harm, a 2026 argument that the associations are confounded, and a hedged review of the GLP-1-specific case conceding the outcome data are missing. Nobody can tell a reader that stopping and restarting will harm them. Nobody can tell them it certainly will not.
What restarting actually involves
A fresh clinical assessment, not a resumption — and the labels address something narrower than people assume.
Semaglutide's product information does contemplate restarting at a reduced dose. The Singapore and EU wording is identical: if more doses are missed, reducing the starting dose for re-initiation should be considered — unquantified, and framed as a consideration. The US label instead names a specific trigger of two or more consecutive missed doses. Tirzepatide's product information contains no re-initiation guidance at all, in the US, the EU or Singapore (product information).
Two limits on reading anything into that. All of it addresses missed doses rather than a deliberate break of months, and none of it says whether a second course is as effective. And the US wording is not the Singapore wording, which is why advice found online often does not describe the rule that applies here.
What dose someone restarts at, whether they restart at all, and what has changed in their health since they stopped are questions for a prescribing doctor who can see them.
A former GetLean patient who wants to restart has a review consultation with Dr Quek first. The previous dose can usually be continued where it is clinically appropriate; that is his decision at the review, not an automatic resumption.
At GetLean, the programme is designed to end: our philosophy is that GLP-1 medication should act as a catalyst — not something to depend on indefinitely. The plan for coming off is the GetLean GLP-1 Taper Plan: after hitting your goal weight with the right nutrition and exercise habits built, the dose is stepped down at Dr Quek's monthly reviews towards your lowest effective dose and, where clinically appropriate, off the medication. Someone who has stopped, regained and is considering starting again is not starting from failure: rebound is the pressure the habits and the taper are there for, and the way back in is a review with Dr Quek rather than a dose taken on one's own. How long should you stay on GLP-1 covers how the length of treatment is set.
Individual results vary, and clinical-trial figures describe the populations studied. Never restart, change or stop a prescribed medication on your own; that is a conversation with a doctor.
Common questions
Does GLP-1 medication work as well if you restart it?
Nobody has published the answer. Reinitiation rates are well counted, but no human study reports the weight outcome of a second course after stopping and regaining. In a direct experiment in male mice, restarting semaglutide did produce weight loss but the animals failed to reach the weight they had reached on the first course (Son 2026) — and its authors state that validation in humans is still needed.
Is it common to stop and restart GLP-1 medication?
Yes. Among 125,474 US adults, 64.8% of those without type 2 diabetes stopped within a year, and 36.3% of those who stopped had restarted within the following year (Rodriguez 2025). A separate US cohort found 19.6% restarted within a year and a further 35.2% took up another obesity treatment (Gasoyan 2026). Both are observational US records in an access context that does not carry over to Singapore.
Does regaining weight after stopping mean the treatment failed?
No. In the withdrawal trials the medication was taken away, by stopping it or by switching people to placebo, and regain was what happened to the groups as a whole, not what separated some participants from others. In the STEP 1 extension, where the medication and the lifestyle support stopped together at week 68, about two-thirds of the lost weight returned within a year (Wilding 2022). That is the medication's effect fading, which is biology, not a lack of willpower. At GetLean, if the weight starts coming back, a review consultation with Dr Quek comes first, and whether to restart is his decision.
Is weight cycling bad for you?
The evidence disagrees with itself. A meta-analysis of 23 prospective cohorts covering 441,199 people found body-weight fluctuation associated with 35–50% higher relative risks of all-cause mortality, cardiovascular mortality, cardiovascular disease and hypertension (Zou 2019). A 2026 Lancet Diabetes & Endocrinology Personal view reappraised that literature and concluded the evidence does not support a causal link, attributing the associations to confounding (Magkos 2026). One is observational, the other is an argued appraisal, and neither settles it.
What does the label say about restarting after a break?
Semaglutide's Singapore and EU product information says that if more doses are missed, reducing the starting dose for re-initiation should be considered; the US label instead names a two-consecutive-dose trigger. Tirzepatide's product information contains no re-initiation guidance in any of the three jurisdictions (product information). All of that addresses missed doses rather than a deliberate break, and none of it says whether a second course works as well.