When the drug works too well

There is a particular kind of quiet distress that shows up once the weight is off. The drug worked. It worked better than anything the person had tried in twenty years of trying. And now the thought of stopping - because of cost, or supply, or a doctor’s raised eyebrow, or just the sense that surely one shouldn’t need a medication forever - arrives with a knot of something like shame. Why can’t I just keep this off on my own?

Here is the first and most important thing this piece has to say: that question is aimed at the wrong target. The difficulty of coming off a GLP-1 is not a referendum on your discipline. It is the predictable, well-documented behavior of a body that is very good at defending its own fat mass - and the better the drug worked, the more there is to defend. The efficacy and the difficulty are the same fact seen from two sides.

Bodyweight is not a passive readout of calories in and out. It is a regulated variable, held near a value the brain treats as correct - the so-called defended weight, or set point. Lose a meaningful amount of it and the hypothalamus does not shrug; it mounts a coordinated, multi-system defense to get the fat back. Three arms of that defense do most of the work:

• Hunger turns up. Ghrelin, the stomach's hunger signal, rises. Appetite climbs above where it sat before you ever lost the weight - the drive to eat is not just restored, it is amplified.
• Fullness turns down. Leptin, the hormone fat tissue uses to report how much fuel is in the tank, falls with fat mass - and satiety signals like PYY drop with it. Meals stop registering as filling as quickly.
• The engine idles lower. Energy expenditure falls by more than the smaller body should require - adaptive thermogenesis. You burn fewer calories at rest than a never-dieted person of the same size.

The cruel part is how long this lasts. In a landmark study, people who lost weight on a strict diet still had elevated hunger hormones and suppressed satiety hormones a full year later - the appetite changes had not reverted at all (Sumithran et al., 2011, NEJM). Follow-up of “The Biggest Loser” contestants found the metabolic slowdown persisted six years on (Fothergill et al., 2016). The defense does not know the weight came off on purpose. It just knows the tank looks low, and it pushes to refill it - for years.

Once you see the defense clearly, the GLP-1 stops looking like a weight- loss trick and starts looking like what it is: a counterweight. The drug does not delete the hunger signal or reset the set point. It sits on the other side of the scale, pharmacologically opposing the defense - turning appetite down, slowing gastric emptying, and making the brain read “enough” sooner. For as long as it is on board, the body’s push to regain is matched by an equal push the other way.

Which tells you exactly what stopping does. It does not return you to some neutral baseline. It removes the counterweight from a defense that never left. The hunger, the blunted fullness, the lowered burn - all of it is still there, quietly, the whole time you were succeeding. Take the drug away and that biology is simply unopposed again. Weight regain after stopping is not relapse in the moral sense. It is untreated obesity resuming, on schedule.

This is not a hunch or a metaphor stretched too far - it is one of the most consistent findings in the whole metabolic literature, because the trials were deliberately built to test it. Take people who have already lost weight on the drug, split them into keep-going and stop groups, and watch. The picture is the same every time.

Read together, these say something both sobering and clarifying: the loss holds while the drug is on board and reverses when it comes off, across different molecules and different trials. A 2026 meta-analysis pooling cessation data landed on the same trajectory (eClinicalMedicine, 2026). The authors of the first big withdrawal study put it plainly: the results “confirm the chronicity of obesity.”

That word - chronic - is the one that dissolves the shame. We do not ask why a blood-pressure pill stops working once you stop swallowing it. Nobody feels like a failure for needing thyroid hormone every morning, or for the fact that a statin’s benefit vanishes weeks after the last dose. We understand those as management of ongoing conditions, not as dependencies to be embarrassed about. Obesity, the withdrawal data says, behaves the same way.

None of which means anyone is sentenced to injections forever. It means the honest starting frame is maintenance, not cure - and that time off, when it is the right call, is a decision made with open eyes rather than a test of character you were always going to fail.

Plenty of the time, there are real and good reasons to pause, taper, or stop - and wanting to is not irrational. What matters is separating the reasons that come with a plan from the ones driven by that misplaced shame. Legitimate reasons a clinician will actually work with:

• Pregnancy or trying to conceive. Labels advise stopping well before a planned pregnancy - two months ahead, for semaglutide - because safety in pregnancy is not established. This is a clear, non-negotiable off-ramp.
• Surgery and procedures. Because these drugs slow gastric emptying, anesthesia guidance now advises holding them before procedures to lower aspiration risk. A planned, temporary pause with a restart date.
• Cost, supply, or side effects. The most common real-world reasons. They are worth naming honestly rather than dressing up as willpower - and they are exactly the constraints the next generation of drugs is built to ease.
• A considered trial off. Some people have lost a modest amount, built durable habits, and want to see whether they can hold it on a lower dose or none. A reasonable experiment - if it is run as an experiment, with a plan.

For many people with established obesity, the plain reality is that indefinite maintenance is the medically sound answer, the same way it is for hypertension. That is not a defeat. But time off is a legitimate choice, and the difference between it going well and going badly is almost entirely in how it is done.

If you and a clinician decide to step down, the biology above points to a handful of principles. None of this is a protocol - doses and timing belong to you and your prescriber - but the shape of a good off-ramp is well understood.

• Prefer a maintenance dose to a cliff. The emerging strategy is treat-to-maintain: find the lowest dose that holds the loss rather than stopping outright. The counterweight can often be smaller than the dose it took to lose the weight - it just cannot usually go to zero.
• Taper, do not drop. Stepping the dose down gradually gives appetite time to rise slowly instead of rebounding all at once. The evidence for tapering specifically is still thin, but it follows directly from how the defense behaves.
• Build the scaffolding while you are on it. The quiet-appetite window the drug creates is the time to install what has to hold afterward: protein at every meal and resistance training to protect muscle (see the muscle-preservation piece), plus the routines that outlast the prescription.
• Set a restart trigger in advance. Decide, before you taper, the amount of regain at which you resume - and treat hitting it as the plan working, not as failure. An off-ramp with a return ticket. Stopping is not a one-way test you pass or flunk.
• Expect appetite to come back. It will, and it is not the drug 'leaving your system' punishing you - it is your own physiology resuming. Naming that in advance takes most of the sting out of it.

Here is the genuinely hopeful part, and it is where this site likes to live. The hard scientific problem - can we reliably strip large amounts of fat from the human body? - is essentially solved. The frontier has moved to a friendlier question: can we make keeping it off so low- burden that staying on treatment stops feeling like a burden at all? A lot of very good work is pointed straight at that.

• Maintenance that barely registers. Oral GLP-1s like orforglipron and longer-interval dosing are turning maintenance from a weekly injection into something closer to a daily pill or an occasional one - the kind of regimen people actually stay on for years.
• Higher-quality weight to defend. Pairing a GLP-1 with myostatin-pathway drugs preserves, even builds, muscle during loss. Keep more of the metabolically active tissue and the weight you hold is sturdier and easier to maintain.
• Induction, then maintenance. The field is starting to think in two phases - a stronger agent to get the weight off, a gentler one to keep it there - borrowing a playbook that works across chronic medicine.
• Does the defense ever ease?. The open, optimistic question: whether deep, sustained loss can gradually lower the defended weight itself, so maintenance gets easier over time rather than staying a lifelong standoff. Unproven - but exactly the right thing to be studying.

So if you are the person at the top of this piece - the one it worked for, quietly worried about the day you stop - the science is on your side twice over. Once, in telling you the struggle was never a character flaw. And again, in the direction of travel: toward maintenance so easy that the question stops being how do I get off this? and becomes simply how do I keep feeling this good?