- •Maintenance is not the same as stopping — it means finding the lowest effective dose that holds your weight, which for most people is not zero.
- •SURMOUNT-4 (JAMA 2024): stopping tirzepatide led to 14% weight regain in 52 weeks; continuing led to 5.5% further loss.
- •STEP 4 (JAMA 2021): people who stopped semaglutide regained 6.9% while those who continued lost another 7.9%.
- •Protein and resistance training are non-negotiable in maintenance — the goal shifts from losing fat to protecting muscle.
- •Some people successfully lower their dose or extend their dosing interval; this should always be a doctor-supervised experiment, not a solo one.
What is a GLP-1 maintenance dose?
A maintenance dose is the lowest dose of a GLP-1 medication that keeps your weight stable — not the dose that gets you to your goal, and not zero.
This distinction matters more than almost anything else in GLP-1 treatment, because it reflects what these medications actually do. GLP-1 receptor agonists (and dual GIP/GLP-1 agonists like tirzepatide) work by slowing gastric emptying and acting on appetite centers in the brain. They change the signal, not the setpoint. Remove the drug and the old signal — hunger, food noise, a body defending its previous weight — comes back.
That's why maintenance is a dose, not a graduation.
In practice, maintenance usually looks like one of three things:
1. Staying at your treatment dose. The most common and best-evidenced path. If 10mg tirzepatide got you to goal, 10mg tirzepatide often keeps you there. 2. Stepping down one level. Some people find that a lower dose holds their weight with fewer side effects. This works for some and not others, and it needs monitoring. 3. Extending the interval. Dosing every 10 days instead of every 7, for example. This is done in practice but is not what the trials studied, so the evidence is thin and it should only be attempted with a prescriber.
None of these are failure. All of them are treatment.
What do the trials say about stopping vs. continuing?
This is the clearest evidence base in the whole GLP-1 conversation, and it says one thing consistently: stopping leads to regain.
SURMOUNT-4 (JAMA, 2024) is the sharpest test. Participants took tirzepatide for 36 weeks — losing about 20.9% of body weight — and were then randomized either to continue or to switch to placebo. Over the next 52 weeks, the placebo group regained 14% of their body weight. The continuation group lost an additional 5.5%.
STEP 4 (JAMA, 2021) ran the same experiment with semaglutide. After a 20-week run-in, people who continued lost a further 7.9%; people switched to placebo regained 6.9%.
STEP 1 extension found that a year after stopping semaglutide, participants had regained roughly two-thirds of the weight they'd lost, and cardiometabolic improvements largely reversed with it.
The honest reading of this data is not 'you're trapped on a drug forever.' It's that obesity behaves like every other chronic condition we treat. Nobody is surprised when blood pressure rises after stopping a blood pressure medication. This is the same biology.
What the trials *don't* tell us: whether a reduced maintenance dose works as well as a full one, or whether people who build strong nutrition and training habits during the loss phase regain less. Those are real, open questions — and they're the ones most worth asking your prescriber about. Our guide to [stopping a GLP-1 and tapering safely](/blog/stopping-glp1-weight-regain-and-how-to-taper-safely) covers the exit path in more detail if that's your plan.
How do you know when you've reached maintenance?
You've reached maintenance when your weight has been stable for 4-8 weeks without you actively trying to lose — and you and your doctor agree you're where you want to be.
The tricky part is that a plateau and maintenance look identical on the scale. The difference is intent. A plateau is an unwanted stall in the middle of a loss phase; maintenance is a deliberate stop. If you're not sure which one you're in, that's a conversation to have — our breakdown of [why GLP-1 weight loss plateaus happen](/blog/glp1-weight-loss-plateau-why-stalls-happen-and-how-to-break-them) can help you tell them apart.
Signs you're genuinely at maintenance:
- •Weight stable within a 3-5 lb band for at least a month, with normal day-to-day fluctuation.
- •You feel good. Energy is steady, you're not fighting constant fatigue or hunger.
- •Body composition is holding. If you have access to a DEXA or even a decent smart scale, lean mass isn't dropping.
- •Your goal is met — by health markers (blood pressure, A1c, lipids), not just a number you picked in 2011.
A note worth saying out loud: the number you chose as a goal years ago may not be the right target now. Especially in midlife, chasing an aggressive low weight often costs muscle and bone you can't easily get back. A weight that's 5-10 lb higher but comes with more lean mass, better labs, and no white-knuckling is a better outcome, not a compromise.
Can you lower your dose in maintenance?
Sometimes — and it's a legitimate conversation, not a fantasy. But it needs to be run like an experiment, not a hope.
The case for lowering. Lower doses generally mean fewer side effects and lower cost. Some people find that once they've reached goal, a step-down dose is enough to hold appetite in check, particularly if they've built strong eating and training habits.
The case for not lowering. The trial evidence is all built on maintaining the full dose. When you drop the dose, you're going off-map. Regain, when it happens, is usually gradual and easy to rationalize away until it's substantial.
How to do it safely, if you and your doctor decide to try:
- •Change one thing at a time. Drop one dose level, not two, and don't simultaneously change your diet or training.
- •Give it 8-12 weeks. Regain from a dose reduction is slow. A month isn't long enough to know.
- •Track weekly, not daily. Look at a 7-day rolling average. Daily weight is mostly water and noise.
- •Set a tripwire in advance. Decide the number that means 'go back up' *before* you start — for example, 5 lb above your maintenance band, sustained for two weeks. Deciding in the moment is how people drift.
- •Watch food noise, not just weight. If the mental chatter about food comes roaring back, that's an early warning that usually precedes the scale by weeks.
And the most important rule: never adjust or stop a prescription on your own. If cost is the driver, say so directly to your prescriber — there are often better options than silently rationing doses. Our [2026 savings guide](/blog/how-to-get-glp1-cheaper-savings-cards-telehealth-2026) covers those.
What should you eat and do in maintenance?
The goal changes. In the loss phase you were protecting muscle *while* losing fat. In maintenance you are simply protecting muscle — and that requires more food, not less.
Protein is the anchor. Aim for roughly 1.2-1.6 g per kg of body weight per day (about 0.5-0.7 g per pound), spread across meals rather than dumped into dinner. GLP-1 medications reduce appetite, which makes it easy to under-eat protein without noticing — and a substantial share of GLP-1 weight loss can come from lean mass if protein is inadequate. Our [protein and muscle guide](/blog/protein-needs-on-glp1-during-menopause-prevent-muscle-loss) has the practical version.
Resistance training, 2-3x per week. This is the single highest-leverage thing in maintenance. Lifting signals to your body that muscle is worth keeping. Without that signal, in a calorie-restricted state, muscle is expensive tissue your body is happy to discard. Full-body sessions, compound movements, progressive overload — nothing exotic required.
Eat enough. This sounds absurd after a year of eating less, but it's the most common maintenance error. If you keep eating at a loss-phase deficit, you'll keep losing — and what you lose next is disproportionately muscle and bone.
Keep hydration and micronutrients on the radar. Reduced food volume means reduced micronutrient intake. [Deficiencies in B12, iron, and vitamin D](/blog/micronutrient-deficiencies-on-glp1-what-to-watch-for) are worth checking annually.
Re-check labs. Maintenance is the right time to re-baseline: A1c, lipids, blood pressure, thyroid, and — if you're over 50 or postmenopausal — a bone density scan. The point of all this was health, and health is measurable.
What if you regain some weight?
Some fluctuation is normal and some regain is common. Neither is a moral event.
A few pounds is noise. Body weight swings 3-5 lb on water, sodium, hormones, and glycogen alone. A single high number after a holiday weekend is not regain.
A sustained upward trend is signal. If your 7-day average has climbed steadily for 4+ weeks, something has shifted — dose, appetite, activity, stress, sleep, or life. This is information, not a verdict.
Check the usual suspects, in order: 1. Is protein still where it was? Appetite recovering often means carbohydrate and fat creep back in first. 2. Has training dropped off? Muscle loss lowers resting metabolic rate, which quietly widens the gap. 3. Has anything changed with the medication? A missed dose, a pharmacy switch, a compounded formulation change, or a dose reduction that hasn't held. 4. Sleep and stress. Both drive appetite hormones directly. If you're in perimenopause or menopause, [sleep disruption is a known compounding factor](/blog/sleep-on-glp1-during-menopause-fixing-restless-nights).
Then act early. The single biggest predictor of how much weight comes back is how long you wait to respond. Ten pounds is a conversation. Forty pounds is a restart.
And the thing worth internalizing: regain after stopping is pharmacology, not failure. SURMOUNT-4 and STEP 4 didn't measure willpower — they measured what happens when you remove a drug that was doing something real. If you regained, the medication was working. That's the whole finding.
Frequently asked questions
- Continued Treatment With Tirzepatide for Maintenance of Weight Reduction (SURMOUNT-4) (2024)
- Effect of Continued Weekly Semaglutide vs Placebo on Weight Loss Maintenance (STEP 4) (2021)
- Tirzepatide Once Weekly for the Treatment of Obesity (SURMOUNT-1) (2022)
- Two-year effects of semaglutide in adults with overweight or obesity (STEP 5) (2022)
Lea is an AI health companion trained on landmark clinical studies covering GLP-1 medications and menopause. Our content is evidence-based and regularly updated to reflect the latest research.
This article is for informational purposes only and is not medical advice. Always consult your healthcare provider.
Learn more about LeaHave questions about this?
Ask Lea — she'll apply this directly to your medication, your symptoms, your week.
Talk to Lea