Meta-analysis quantifies weight regain after stopping GLP-1 drugs

A systematic review and meta-analysis published under the World Obesity Federation banner has put numbers on something many patients already suspect: weight tends to come back after GLP-1 medications are stopped. Pooling eight randomized controlled trials with 2,372 participants, the authors found an average regain of 9.69 kg (roughly 21 pounds) among people who had taken semaglutide or tirzepatide, and 2.20 kg (about 5 pounds) among those who had taken liraglutide ^[1].

What the analysis looked at

The researchers searched multiple databases through EBSCOhost — including Academic Search Premier, CINAHL Ultimate, the Cochrane Central Register of Controlled Trials, MEDLINE with full text, and the Cochrane Database of Systematic Reviews — plus a separate PubMed search, covering studies from database inception to February 1, 2024 ^[1]. The initial search returned 497 articles; after screening, 8 randomized controlled trials were kept ^[1].

Every participant in the included trials had a body mass index of 27 kg/m² or higher ^[1]. The team pulled data on body weight change from baseline both on and off treatment, along with changes in waist circumference and BMI on and off treatment ^[1]. The pooled mean differences were calculated in RevMan version 5.4 using a DerSimonian-Laird random effects model ^[1].

The numbers

For liraglutide, the pooled regain after stopping was 2.20 kg (95% confidence interval 1.69 to 2.70, P < 0.00001) ^[1]. For semaglutide or tirzepatide — the newer, more potent drugs sold as Ozempic and Wegovy (semaglutide) and Mounjaro and Zepbound (tirzepatide) — the pooled regain was 9.69 kg (95% CI 5.78 to 13.60, P < 0.00001) ^[1].

The wide confidence interval on that second figure matters: the true average could plausibly sit anywhere from under 6 kg to nearly 14 kg, which suggests the underlying trials varied quite a bit ^[1].

The authors' central observation is that regain tracked with loss. "After discontinuing GLP-1RA therapy, weight regain was proportional to the original weight loss," they wrote ^[1]. In other words, the bigger regain seen with semaglutide and tirzepatide appears tied to the fact that those drugs produced more weight loss in the first place, not necessarily to some unique rebound effect ^[1].

The authors also reported that regain happened "regardless of lifestyle interventions," and concluded that these medications "should therefore be considered a chronic therapy to prevent weight regain and associated undesirable outcomes related to obesity" ^[1]. They add that the prospect of regain "should be discussed when stopping therapy" ^[1].

Why it matters for patients

Most public conversation about GLP-1 drugs centers on how much weight people lose while taking them. This analysis is about what happens after — a question that becomes concrete when insurance coverage lapses, a prescription runs out, a supply problem hits, or side effects make continuing hard.

The framing of these drugs as chronic rather than short-course treatment has practical consequences for cost planning, coverage decisions and expectations. If regain is proportional to loss, someone who lost a large amount on a high-potency drug may face a correspondingly larger rebound than someone who lost less on an older agent ^[1]. That is an average across trial populations, though, not a prediction for any one person.

It is also worth noting what the published abstract does not spell out. It does not state how long after discontinuation the weight was measured, which specific trials were included, how much weight participants had lost before stopping, or the pooled results for waist circumference and BMI — even though those outcomes were extracted ^[1]. It does not report on cardiometabolic markers such as blood pressure or blood sugar after stopping. Those details are not yet available from the source at hand.

What happens next

The search cutoff was February 1, 2024, so trials and follow-up data published after that date are not reflected here ^[1]. The analysis is indexed in PubMed as both a meta-analysis and a systematic review ^[1]. Whether newer discontinuation studies — including those involving other agents in the class — would shift the pooled estimates is not addressed in this paper.

Anyone weighing questions about starting, continuing or stopping one of these medications should take them up with a clinician who knows their history.