Weight & long-term care · 3 min read

Why does weight return after stopping a GLP-1 medicine?

Weight regain after semaglutide withdrawal is common, but the amount varies. A follow-up study explains why a maintenance plan matters.

Prepared by PoshanSense · AI-assisted educational writing
Our sources and editorial approach · Not clinically reviewed

An unbranded capped injection pen, a weighing scale and a winding path in a calm green room. — Original editorial illustration; a visual introduction, not a depiction of the study participants or a medical result.

Research explained · Published 2022

The part after the before-and-after photo

A dramatic weight-loss photograph says little about the next year. The more useful question is what support continues when a treatment changes. Regain should prompt a discussion about long-term care, rather than a judgement about willpower.

Study at a glance

STEP 1 extension: one year after withdrawal

327 adults in the extension analysis across Canada, Germany, Japan, the UK and the US. 68 weeks of treatment, followed by 52 weeks off treatment.

Setup, comparison and main limit

Design: Exploratory follow-up of a subset from a randomized trial
Sample: 327 adults in the extension analysis across Canada, Germany, Japan, the UK and the US.
Comparison: Previous semaglutide versus previous placebo; neither group continued study treatment or structured lifestyle support.
Duration: 68 weeks of treatment, followed by 52 weeks off treatment.
Measured: Body weight and cardiometabolic measurements; 290 participants had on-site measurements at the final visit.
Main limit: The extension did not randomize participants to continue versus stop treatment, and lifestyle support also ended.

Read the full paper →

What actually returned?

In the extension group previously assigned semaglutide, average weight loss during treatment was 17.3% of starting weight. Over the next year, 11.6 percentage points returned. Participants were still, on average, below their starting weight. “Regained two-thirds” therefore does not mean everyone regained everything, or that two-thirds of participants regained weight.

Why this does not measure willpower

The study is consistent with weight management being an ongoing process. It cannot tell us how much regain came from stopping medication, how much came from losing structured support, or which maintenance approach would have worked best. The extension was selected from the larger trial, and final measurements were missing for some participants. Novo Nordisk funded the research; the paper describes multiple industry relationships.

Does this apply to every weight-loss injection?

No single semaglutide follow-up supplies an exact forecast for tirzepatide, every dose, or every person. It also did not test an online tapering schedule. The useful takeaway is to discuss the maintenance phase before changing treatment: ongoing follow-up, nutrition, activity, affordability and what happens if appetite returns.

A better question for the next appointment

Ask “What is our plan if this treatment ends?” rather than “How do I guarantee I never regain?” A food journal can help describe changing routines, but it cannot replace clinical care or decide whether a medicine should continue. Do not start, stop or adjust prescribed treatment from a blog.

Original explanatory infographic

Read the regain figure correctly

During treatment

Average loss in the extension semaglutide group: 17.3% of starting weight.

After withdrawal

11.6 percentage points returned over the following year.

The missing comparison

The extension did not test continuing treatment against stopping it.

An explanatory guide to the evidence, not a personal treatment plan.

Sources & further reading

Study years differ from this article’s publication date. We reviewed the linked research in full, distinguishing original experiments, observational studies and reviews. Access limitations are not used to support efficacy claims. Original illustrations and diagrams; no publisher figures reproduced. Educational writing, not clinical review.

Published 16 October 2026 · Sources checked 16 October 2026. Suggest a correction.