What happens after stopping a GLP-1? What two 2026 reviews found

What the source reports

Two peer-reviewed systematic reviews published in 2026 examined what happens to body weight after GLP-1-based treatment ends. Both found the same broad pattern: average weight moved upward again after treatment stopped, although the amount and timing varied substantially between studies.

A GLP-1-specific review in EClinicalMedicine estimated that participants had regained 60% of the weight they had lost one year after stopping. A broader BMJ review of weight-management medicines estimated average regain of 0.4 kg per month across all included medications.

These are group-level estimates assembled from earlier studies—not a forecast for an individual. The reviews do not show that everyone regains all lost weight, and much of the apparent longer-term trajectory is based on statistical projection rather than years of direct observation.

The GLP-1-specific analysis

The EClinicalMedicine researchers identified 48 studies reporting weight after GLP-1 receptor agonist cessation. Their main nonlinear model used six randomised trials involving 3,236 participants with follow-up data extending to a maximum of 52 weeks after treatment stopped.

The model estimated a faster initial regain that gradually slowed. At week 52, average regain represented 60% of the weight originally lost during treatment.

The included evidence covered different medicines and populations. It included liraglutide and semaglutide, as well as tirzepatide, which acts at both GIP and GLP-1 receptors rather than GLP-1 alone.

The 75% figure is a projection

The same model projected that average regain might eventually plateau at 75.3% of the original weight loss. Its 95% confidence interval ranged from 68.9% to 81.6%.

That projection is sometimes simplified into a claim that people will regain three quarters of their lost weight. The important qualification is that the underlying trial observations stopped at 52 weeks; the proposed plateau beyond that point was extrapolated.

The authors explicitly said that no large-scale trials had reported longer post-cessation trajectories. The plateau could be confirmed, revised or disproved when longer follow-up becomes available.

What the broader BMJ review found

The BMJ review included 37 studies, 63 treatment arms and 9,341 participants who had stopped a weight-management medicine. Average treatment duration was 39 weeks and average follow-up after stopping was 32 weeks.

Across all medicines, average regain was estimated at 0.4 kg per month. The authors projected a return to starting weight after 1.7 years, although that estimate combines different drug classes and relies on extending the observed trend.

For a smaller subgroup involving newer incretin medicines such as semaglutide and tirzepatide, the estimated rate was faster at 0.8 kg per month, with a projected return to baseline at 1.5 years. That subgroup contained only eight studies and no post-treatment follow-up beyond 12 months.

Health markers may move back too

The BMJ analysis also examined HbA1c, fasting glucose, cholesterol, triglycerides and blood pressure. Across weight-management medicines generally, improvements diminished after treatment ended and were projected to approach baseline within about 1.4 years.

There were not enough data to analyse those cardiometabolic changes specifically for the newer, more effective incretin medicines. It would therefore be too strong to attach the 1.4-year estimate directly to semaglutide or tirzepatide users as a certainty.

The direction of the evidence is still relevant: some health improvements track with weight and active treatment, so stopping can mean losing part of the measured benefit as well as regaining weight.

Why regain does not mean personal failure

Neither review establishes that weight regain reflects a failure of motivation or character. GLP-1-based medicines change biological signals related to appetite, fullness and glucose regulation while they are being used.

When treatment is removed, those treatment effects no longer operate in the same way. The reviews measured what happened to weight; they were not designed to allocate blame or prove one biological mechanism for each person's trajectory.

People also stop for very different reasons, including adverse effects, cost, supply, prescribing limits, pregnancy planning or a jointly agreed clinical decision. Those circumstances may affect what happens next but were not consistently captured in the evidence.

Not everyone followed the average

Study results varied, and the GLP-1-specific model described an average curve rather than a rule. Some participants regained more, some less and some studies reported much smaller changes.

The researchers could not test how comorbidities, other medicines, duration of GLP-1 treatment or weight-maintenance support altered the regain trajectory because reporting was not detailed or consistent enough.

The model also excluded studies in which average on-treatment weight loss was less than 3 kg. People with a more modest initial response may follow a different path from the one modelled.

Does tapering prevent regain?

The reviews do not establish a best discontinuation method. Gradual dose reduction, lower maintenance dosing, behavioural support, dietary approaches and digital monitoring are all being discussed or studied, but the available evidence did not allow reliable comparisons between strategies.

A small amount of included evidence suggested that particular support or longer treatment might alter regain, but it was not strong enough to prove a general solution. The BMJ researchers found no established evidence that more behavioural support during treatment reliably slowed regain after medication ended.

Tapering should therefore not be presented as a proven way to prevent rebound. Well-designed trials comparing post-treatment strategies are still needed.

What remains uncertain

Most included studies were not originally designed around the period after stopping. Follow-up was often short, outcome reporting varied and the GLP-1-specific review judged most studies to have a moderate risk of bias.

The six-trial trajectory model was too small for a reliable assessment of publication bias. It also combined different drugs, study designs and patient populations, including people with and without type 2 diabetes.

We still need multi-year randomised evidence, individual-level predictors and direct comparisons of stopping, tapering and maintenance approaches. Until then, precise claims about how much any one person will regain are not supported.

Funding and independence

The EClinicalMedicine review reported no funding. The BMJ project was funded in part by the publicly funded National Institute for Health and Care Research Oxford Biomedical Research Centre, with additional public or charitable support for several researchers.

The BMJ paper stated that its funders had no role in the study design, analysis, manuscript or publication decision. Some authors disclosed involvement in publicly funded trials where commercial organisations had donated weight-loss interventions.

Both papers were systematic reviews rather than manufacturer trials of a new product. Their conclusions are still limited by the quality, duration and reporting of the studies they were able to include.

What this means for the individual

The evidence makes weight regain after stopping an important possibility to plan for, not an inevitable personal outcome. It does not provide a formula for predicting a specific number of kilograms or a precise timetable.

It also does not provide a reason to stop, continue or alter treatment without an individual clinical assessment. Decisions can depend on why treatment is ending, the approved indication, adverse effects, other health conditions, access and the available follow-up plan.

Recording weight trends, appetite changes, symptoms and relevant health measurements can help make a subsequent clinical discussion more specific, regardless of whether treatment continues or ends.

Bottom line

The best current synthesis suggests that substantial average weight regain is common after GLP-1-based treatment stops. One 2026 model estimated that 60% of lost weight was regained at one year; another broader review estimated rapid regain after weight-management medicines and projected that much of the associated metabolic benefit would diminish.

Those findings support treating obesity and weight maintenance as long-term questions. They do not prove that everyone regains everything, validate blame, or tell us the safest and most effective way to discontinue treatment.