After the Loss: What Actually Happens in Maintenance

By Ada

TL;DR: New research says stopping a GLP-1 leads to regain almost four times faster than ending a lifestyle-only weight-loss effort. Real-world data tells a second half of the story: people who keep some structure in place mostly hold their loss. You didn't fail — your body did exactly what the study said it would. Structure is the variable you still control.

Key Takeaways

• A University of Oxford meta-analysis (37 studies, 9,341 people) found regain after stopping semaglutide or tirzepatide runs about 1.8 lb a month — almost four times faster than after ending a lifestyle-only program with no medication involved.
• Real-world data tells a different story: in a 7,938-patient Cleveland Clinic study, people who stopped a GLP-1 for obesity regained just 0.5% of body weight on average after a year, and 45% kept losing or held steady.
• The gap between the two numbers isn't the chemistry — it's what people did next: switch medications, restart, or add lifestyle support with a clinician.
• Two-year persistence on GLP-1s sits around 15% overall (Wegovy 24%, Ozempic 22%) — the second year, not the first, is where most plans break.
• Stopping rarely means done: over 40% of people who discontinue a GLP-1 restart within a year — an interrupted plan more often than a canceled one.

The scale ticks up two pounds. Your first thought isn't a number — it's a verdict: I did this wrong. You didn't. If you're in the stretch after a GLP-1 — tapering, holding a maintenance dose, or newly off the medication altogether — a small regain isn't a discipline story. It's the single most predictable event in this entire category. The useful question isn't "why did I slip." It's "what happens next, and who's still standing a year from now."

The number everyone is citing. A University of Oxford meta-analysis published in The BMJ in January 2026 pooled 37 studies and 9,341 people who had stopped a weight-management medication (West et al., BMJ, 2026). For semaglutide and tirzepatide specifically, regain averaged about 0.8 kg — roughly 1.8 lb — a month, putting most people back near their starting weight within about a year and a half. Regain after stopping a GLP-1 ran almost four times faster than regain after ending a lifestyle-only program with no medication involved. Blood pressure, cholesterol, and blood sugar drifted back with the weight. The researchers' read: obesity behaves like a chronic, relapsing condition, and the medication alone doesn't rewrite that.

The real-world number tells a different story. A Cleveland Clinic study published in Diabetes, Obesity and Metabolism in March 2026 followed 7,938 patients who had stopped semaglutide or tirzepatide (Cleveland Clinic, 2026). One year out, the people originally treated for obesity had regained just 0.5% of body weight on average — 45% were still losing or holding steady. Yes, 55% gained some weight back — this isn't zero regain. But the average trajectory barely resembles the RCT curve. What patients actually did next: 27% switched to a different medication, 20% restarted the original one, 14% added structured lifestyle support with a clinician. Under 1% had surgery. Almost nobody just stopped and hoped.

Sit with both numbers, because the gap between them is the whole story. The trial-conditions number describes a body left alone after the medication stops. The real-world number describes a person who kept some kind of structure in place — a follow-up plan, a new routine, a habit that didn't depend on the injection to exist. That gap is the size of the behavior layer, and it holds whether your prescription is a weekly dose or a daily pill: whatever comes after cessation is a behavioral question as much as a pharmacological one.

There's a reason the maintenance stretch feels like the hardest one. Real-world claims data from Prime Therapeutics — a 2024 analysis still cited as the category benchmark — puts two-year persistence on a GLP-1 around 15% overall, with Wegovy holding highest at 24% and Ozempic at 22% (Prime Therapeutics). The first year isn't where most plans break. The second year is. And "stopped" rarely means "done": a study presented at ENDO 2026 followed over 60,000 people with type 2 diabetes and found 41.5% restarted within a year, nearly 58% within two (Endocrine Society, ENDO 2026). Most of this isn't abandonment. It's an interrupted plan, not a canceled one.

None of this is a script for what to do with your own dose — that conversation belongs to you and your clinician, not an article. What the research does tell you is where to put your attention. The walk you took anyway. The protein at breakfast. The check-in that catches a stall in week three instead of month three. None of that shows up on a scale by Tuesday. All of it is what "45% held the line" is actually made of. If you're counting wins this week, count the ones that aren't the number on the scale. That's not a consolation prize. That's the mechanism.

This is the exact seam a companion is built to sit inside — not to replace your clinician, and not to promise the medication does more than the data says it does, but to keep the structure part visible on the days the motivation that got you here goes quiet. The window between "I stopped" and "I regained it all" isn't fixed. Some of it is chemistry. A meaningful part of it is what you build around the chemistry — the part that stays yours whether the dose changes, the drug changes, or nothing changes at all.

Keep reading:

• Closer to a stall than a stop? What to do about a GLP-1 weight-loss plateau covers what to check before assuming failure.
• Want the fuller research picture on regain? Weight regain after stopping GLP-1s goes deeper than this piece did.
• Worried about muscle specifically? How to prevent muscle loss on GLP-1 medication covers protein and resistance training.

Not sure the structure is already there? The habit readiness assessment takes two minutes and tells you.

We track studies like these before the headlines simplify them — subscribe to the newsletter for the real numbers.

Gila is in its free pilot right now. Sign up for the free pilot and bring the structure with you into maintenance.

This article covers behavioral science and published research — not medical advice. Gila doesn't diagnose, prescribe, or recommend changes to your medication. Any question about tapering, switching, restarting, or stopping treatment belongs with your clinician.