Weight regain after stopping Ozempic: what the trials actually show

What did @lukepat2 actually say?

The creator's core claim is simple: people lose significant weight on semaglutide 2.4mg, but once they stop, most of it comes back. They cite the STEP 1 trial, saying participants lost 17.3% of body weight versus 2% on placebo. After stopping, they claim "around 5% regained in the first 12 weeks" and "11.6% of that weight was back" by two years. They close with the argument that obesity is a chronic condition requiring long-term management, not a one-and-done fix.

They also call the drug "Zenbic" at the start, which appears to be a mispronunciation of Wegovy or a garbled version of semaglutide's brand name. Minor, but worth noting for clarity.

The framing is honest and the conclusion is medically sound. This is not a hype video. It's actually one of the more grounded GLP-1 takes you'll find at 1.3K views.

Does the science back this up?

Mostly yes, and the specific numbers are largely correct. The STEP 1 extension trial, published by Wilding et al. in 2021 in the New England Journal of Medicine, followed participants for 68 weeks on semaglutide and then tracked them through a withdrawal period. The 17.3% weight loss figure is accurate for the treatment group at week 68.

The withdrawal data comes from a separate STEP 1 extension analysis. Participants who stopped semaglutide regained roughly two-thirds of their prior weight loss within one year of stopping. By week 120 of the full study period, body weight in the treatment group was about 5.6% below baseline, meaning they had regained most but not all of the loss. The "11.6% regained by 2 years" framing is a reasonable interpretation of that data, though the exact framing depends on whether you're measuring from peak loss or from baseline.

The 100kg worked example is a fair illustration, not a guarantee, and the creator presents it as an estimate, which is appropriate.

What did they get wrong (or right)?

The timeline framing deserves some scrutiny. The creator says the trial ran for "6-8 weeks," which is almost certainly wrong. The active treatment phase in STEP 1 was 68 weeks, roughly 16 months. Six to eight weeks is closer to the dose escalation period. This is a real error, not a rounding issue, and it matters because it undersells how long people were actually on the drug before the rebound data was collected.

The "5% regained in the first 12 weeks" figure is harder to verify precisely from published tables, but it is directionally consistent with the withdrawal curve shown in the Wilding 2022 extension data. It's plausible, but the creator doesn't cite a specific source for that number, which is worth flagging.

What they got right: the conclusion. Calling obesity a chronic condition that "needs long-term management" is scientifically accurate and clinically important. This framing is consistent with guidance from the American Association of Clinical Endocrinology and the Obesity Medicine Association. It's a responsible take that pushes back against the stop-and-maintain myth.

What should you actually know?

The weight regain after stopping GLP-1 drugs is real, well-documented, and not unique to semaglutide. The SURMOUNT-4 trial (Aronne et al., 2024, JAMA) showed similar patterns with tirzepatide: participants who stopped after a loss phase regained significant weight within months, while those who continued lost more. The biology is consistent across the drug class.

Why does this happen? GLP-1 receptor agonists work by suppressing appetite signals and slowing gastric emptying. When you stop, those signals return to their prior state. The drug was managing the condition, not curing it. This is not a design flaw. It's how most pharmacological treatments for chronic conditions work, including antihypertensives and statins.

The practical implication is that anyone starting a GLP-1 medication should have an honest conversation with their prescriber about long-term plans before starting, not after stopping. Decisions about duration, dose tapering, and lifestyle support should be part of the initial treatment discussion.

The bottom line

This video is more accurate than most GLP-1 content circulating on TikTok. The core data is real, the conclusion is medically defensible, and the framing is responsible. The treatment duration error ("6-8 weeks" instead of 68 weeks) is a notable slip that should be corrected, because it changes how a viewer interprets the scale of the intervention. But the overall message, that these drugs work while you take them and that stopping means regaining, is supported by the best available evidence.