After You Stop a GLP-1: Weight Regain and What Lifting Can Realistically Do

Collage of a man in his fifties doing a one-arm dumbbell row on a flat bench, with the headline "Stopping GLP-1?"

On average, people regain weight after stopping a GLP-1 drug, and by an indirect comparison they regain it faster than people who finish a diet program. A January 2026 BMJ review found an average regain of about 0.8 kg (1.8 lb) a month after semaglutide or tirzepatide, with a projected return to starting weight in roughly 1.5 years. Lifting has not been shown to stop that regain, because I could not find a published trial that tested it. What it can do is protect your muscle and strength while the scale moves, and exercise added to a GLP-1 (liraglutide) helped people keep more weight off after the drug ended in one small Danish trial. Treat lifting as one layer of a plan that also covers food, protein, daily movement and your prescriber's advice.

This article is general education, not medical advice. Whether to stop, taper, switch or restart a GLP-1 medicine is a decision for you and the clinician who prescribes it. Do not stop a prescribed medicine because of anything you read here.

What the BMJ review found

The paper is West and colleagues, published in The BMJ in January 2026 by a University of Oxford group. It pooled 37 studies, 63 treatment arms and 9,341 adults with overweight or obesity; 35 of the studies were randomized trials. The main numbers:

  • During treatment: an average loss of 8.3 kg (about 18 lb) in the medication groups versus 3.2 kg in comparison groups, after an average of 39 weeks.
  • After stopping any weight-management drug: regain of about 0.4 kg a month, with a projected return to baseline weight at about 1.7 years.
  • After stopping semaglutide or tirzepatide: about 0.8 kg a month, return to baseline projected at about 1.5 years, and 9.9 kg (about 22 lb) back in the first year on average.
  • Health markers: blood pressure, glucose and cholesterol drifted back toward baseline, all projected to get there within about 1.4 years. That analysis covered all drug types together; the authors said the data were too thin to analyze these markers for semaglutide and tirzepatide on their own.

How it compares with diet programs

The authors also compared their results with data from an earlier review of behavioral weight-management programs, which offer diet and physical-activity support. People in those programs lost 5.1 kg on average and regained about 0.1 kg a month, with a projected return to baseline in about 3.9 years. Regain after medication was faster by roughly 0.3 kg a month, whatever the amount lost. The authors describe the comparison as indirect, because the people in the two sets of studies may differ. "Baseline" means where you started, not heavier.

What the review cannot tell you

Read these numbers as averages from mixed evidence, not a forecast for your own body.

  • The newer drugs have the thinnest data. Only eight studies covered semaglutide and tirzepatide, with no more than about 12 months of follow-up after stopping. The 1.5-year figure is a straight-line projection beyond that. Marie Spreckley of the University of Cambridge made this point in comments collected by the Science Media Centre.
  • Study quality varies. The designs were mixed, many were rated at risk of bias, and the comparison with diet programs was indirect.
  • It measures weight only. As far as I could confirm, the review did not analyze lean mass, fat mass or bone, so it cannot say what the regained weight is made of or whether lifting changes that.
  • It does not test exercise. In most included studies the support after stopping was identical across groups. Only two compared metformin or behavioral support with none, and the review found no evidence that either changed the rate of regain. Spreckley noted that it does not show behavioral support reliably prevents regain. Tricia Tan of Imperial College London said the review does not discuss structured exercise as a strategy, and argued that obesity needs long-term treatment, as high blood pressure does.

Real-world data looks less dramatic

Gasoyan and colleagues (Diabetes, Obesity and Metabolism, 2026) followed 7,938 patients at one US health system who stopped semaglutide or tirzepatide after 3 to 12 months. A year later, those treated for obesity had an average change of about +0.5% of body weight, 55% of those with weights on record had gained weight, and about 14% had restarted the original drug (19.6% across the whole cohort). Another records study (Murugadoss et al., Biology Methods and Protocols, 2026; 4,182 patients) found about two-thirds had held their weight or kept losing in the six months after their last prescription. Both are observational, short courses leave less weight to regain, and a last prescription is not always the day someone stopped. The trial curves are not destiny, but regain is not rare either.

What randomized withdrawal trials show

In the STEP 1 extension (Wilding et al., Diabetes, Obesity and Metabolism, 2022), 327 adults were followed for 52 weeks after treatment stopped, and the trial's lifestyle program stopped with it. Those who had taken semaglutide had lost an average of 17.3% of body weight by week 68. They regained about two-thirds of that, 11.6 percentage points of body weight. They still finished 5.6% below baseline, against 0.1% for placebo, and about 48% held a loss of 5% or more.

In SURMOUNT-4 (Aronne et al., JAMA, published online December 2023), 670 adults lost weight on tirzepatide for 36 weeks and were then randomized to continue or switch to placebo. Over the next 52 weeks the placebo group gained 14.0% while the continuing group lost a further 5.5%. Everyone was counseled throughout on a 500 kcal-a-day deficit diet and at least 150 minutes of physical activity a week. That advice did not prevent the regain, though the trial was not designed to test the advice itself.

What lifting can and cannot do

I could not find a published trial that randomized people who stopped semaglutide or tirzepatide to a lifting program or nothing and then tracked their weight. Trials of resistance training during treatment are underway, but they ask a different question, which is keeping muscle on the drug (see lifting on Ozempic, Wegovy or Zepbound). After stopping, you are working from indirect evidence:

  • Exercise after a GLP-1 ended. In the S-LiTE trial in Copenhagen (Lundgren et al., NEJM, 2021, with a follow-up by Jensen et al., eClinicalMedicine, 2024), 195 adults first lost an average of 13.1 kg on an 8-week low-calorie diet. They then got a year of supervised exercise, liraglutide, both or placebo. In the follow-up, one year after treatment ended, the combination group's body weight change since randomization was 5.1 kg better than the liraglutide-only group's (95% CI 0.2 to 10.0 kg), and regain over that year was about 6.0 kg larger after stopping liraglutide than after stopping exercise. The limits: only 109 people attended the follow-up, liraglutide is an older drug, the start was a diet-induced loss, and the supervised moderate-to-vigorous program is not described in the abstracts as lifting.
  • Activity and long-term maintenance. The American College of Sports Medicine position stand (Donnelly et al., Medicine and Science in Sports and Exercise, 2009) found that weight maintenance after loss was better with more than 250 minutes of physical activity a week, but that evidence came from cross-sectional and prospective studies, and no well-designed randomized trials existed for preventing regain. It added that resistance training does not enhance weight loss but may increase fat-free mass. In the National Weight Control Registry (Thomas et al., American Journal of Preventive Medicine, 2014), an estimated 87% or more of 2,886 people who had lost at least 30 pounds and kept it off for a year still held a 10% loss at 5 and 10 years, and regain was linked with decreases in leisure-time physical activity and in how often people weighed themselves. It is observational and not specific to drug users.
  • Lifting while dieting. A 2025 network meta-analysis of 62 trials (Xie et al., Frontiers in Nutrition) found that pairing calorie restriction with moderate- or low-intensity resistance or aerobic exercise gave the best mix of fat loss and lean-mass retention. The trials were in healthy adults, not people on GLP-1 drugs, and the authors noted a paucity of long-term follow-up data.

Put together, lifting is a reasonable bet for protecting muscle, strength and function, and for giving your week a structure that does not depend on appetite. It will not replace the drug's effect on hunger, so judge success by whether you keep the strength you built.

A maintenance plan that does not rely on willpower

The goal is to make the default week do the work, because your appetite may return to its pre-drug level. Agree the eating side with your prescriber or a dietitian; this is the movement and tracking side.

Lift two or three times a week

Do two or three full-body sessions of 45 to 60 minutes with a rest day between them. A template:

  1. Goblet squat, 3 sets of 8 to 12
  2. Romanian deadlift with dumbbells, 3 sets of 8 to 10
  3. Dumbbell bench press or push-ups, 3 sets of 8 to 12
  4. Seated cable row, 3 sets of 8 to 12
  5. Overhead dumbbell press, 2 sets of 8 to 12

Stop each set with about two reps left in the tank, and write that number beside the set in whatever log you use. Progress by adding a rep each session until you hit the top of the range, then add the smallest weight jump available. For example, if you managed 16 kg (35 lb) for 10, 10 and 10 on the goblet squat, try 11, 10 and 10 next time, and move to 18 kg (40 lb) once you reach 12, 12 and 12. If two days is all you can sustain, two sessions a week still count. New to lifting? Start lighter than you think and consider a few coached sessions.

Set a protein floor

In a meta-analysis of 49 studies and 1,863 healthy adults (Morton et al., British Journal of Sports Medicine, 2018), the extra gain in fat-free mass from protein supplementation levelled off at a total intake of about 1.6 g per kg of body weight per day in people doing resistance training, and the benefit shrank with age. For an 80 kg (176 lb) person that is about 128 g a day, or roughly 32 g across four meals. Those data come from healthy adults, not people coming off a GLP-1, so treat the number as a starting point and plan meals rather than relying on hunger cues. If you have kidney disease or were told to limit protein, check with your clinician first. The details are in how much protein you need, and the protein calculator gives a range for your weight.

Keep daily movement from falling

The ACSM's threshold of more than 250 minutes a week works out to about 36 minutes a day. A practical version is to hold whatever weekly step average you reached on the drug. Walking and lifting do different jobs, which is the subject of 10k steps versus weight training.

Weigh on a schedule and agree a tripwire

Less frequent weighing was linked with regain in the registry data, so pick one or two fixed mornings a week and use a four-week average. Then write down a trigger before you need it. For example: if your four-week average is 2 kg (4 to 5 lb) above the weight you and your clinician chose, you book an appointment. That number is an illustration, not a validated threshold, so set yours with your care team. Also ask about follow-up checks of blood pressure, glucose and cholesterol.

Do not answer regain with a crash diet

In the network meta-analysis above, calorie restriction alone was linked with lean-mass loss compared with a control group, while no exercise arm paired with it showed a significant loss. A hard cut with low protein is therefore a risk to muscle. Holding maintenance calories while you rebuild strength is a reasonable default to discuss with a dietitian; the trade-offs are in cutting versus bulking.

Track strength, not just the scale

When body weight is noisy, your lifts are a steadier signal. Easy Reps is a free iPhone, iPad and Apple Watch lifting log with no account. Every set goes into workout history with charts and a calendar, and personal records and estimated 1RM update automatically, so you can see whether strength is holding while weight changes. It does not log body weight or nutrition, so use your scale and a separate notebook for those.

Common mistakes

  • Dropping training the week the prescription ends. The exercise in the Danish trial was supervised and structured, and starting a routine before the last dose is likely easier than starting from scratch afterwards.
  • Waiting until regain is obvious. About 0.8 kg a month is easy to miss without a regular weigh-in.
  • Treating protein as optional. It is a low-cost way to support the muscle you are trying to keep during any weight change.
  • Chasing max lifts too soon. Rebuild a few weeks of consistent training before you test heavy singles.
  • Treating regain as personal failure. It is the average pattern in the trial data, which is why a plan matters more than willpower.

Who should be careful, and what a clinician should decide

  • Stopping is not your call alone. Whether to stop, taper, switch (including to a newer oral option such as the one in our orforglipron pill post) or restart belongs to your prescriber. Do not bridge a gap with unregulated products; call the prescriber instead.
  • Blood pressure, diabetes and heart disease. These markers drifted back toward baseline after stopping in the trial data, so your clinician needs to plan monitoring and any medication changes. If you take insulin or other glucose-lowering medicine, ask how weight change and a new exercise routine affect your doses and your risk of low glucose.
  • Eating-disorder history. Weigh-ins and tripwires can backfire, so work with your clinician on a different way to monitor.
  • Older adults, osteoporosis, joint problems or frailty. Start with coached sessions and lighter loads.
  • Red flags. Stop exercising and seek medical care for chest pain or pressure, fainting, severe breathlessness or new palpitations. If you have diabetes, contact your clinician about extreme thirst, frequent urination or blurred vision.

Bottom line

After a GLP-1 ends, weight tends to come back on average, faster than after diet programs by an indirect comparison, though real-world patterns vary and the newer-drug data stop at about a year. I found no study of lifting after semaglutide or tirzepatide ends. A sensible response is a long game: lift two or three times a week, hold a protein target near 1.6 g per kg, keep daily movement up, weigh on a schedule, and agree a plan with your prescriber before the last dose, not after the scale has moved.

Sources