Coming off Ozempic, Wegovy or Mounjaro without losing the result: maintenance dosing, tapering, and the work behind both

If you've been told that most of the weight comes back when you stop a GLP-1, you've been told something the research confirms — but only part of it. The major published trials show that patients who stop Ozempic, Wegovy or Mounjaro abruptly regain around two-thirds of their loss within a year. The other part is that this figure describes a very specific scenario. Patients who stop cold, with no support and no plan. There are two approaches in current clinical practice that produce different outcomes, and neither is what those trials tested. We've covered the first part in detail in our companion article on what the science actually says about weight regain. This article picks up the second part: what those two approaches actually look like, and how to decide between them.

A lower dose, kept on.

In clinical practice, maintenance dosing is the principle that once a patient has reached the weight loss they were aiming for, the medication is reduced to the lowest dose that holds the result. For Ozempic or Wegovy (semaglutide), that often means coming down from 2.4mg or 2.0mg to 1.0mg, 0.5mg, or in some cases lower. For Mounjaro (tirzepatide), it often means coming down from 10mg or 15mg to 5mg. The dose that produced the loss is rarely the dose required to maintain it, because the underlying biology being managed in maintenance is different. The body is no longer being asked to give up fat stores. It is being kept from reclaiming them.

The strongest published evidence on this approach comes from a Danish digital weight management programme, Embla, whose data was presented at the 2024 European Congress on Obesity and published in Diabetes, Obesity and Metabolism in 2025. The programme treated 2,246 patients with semaglutide doses individualised to the lowest level that produced steady weight loss, alongside structured behavioural support. The first finding was that lower doses were as effective as higher ones for most patients. The average maximum dose used was 0.77mg, well below the standard 2.4mg target. The second finding is more relevant here. Of the patients who reached their goal and chose to come off the medication entirely, the dose was tapered gradually over an average of nine weeks while they continued to receive coaching on diet, exercise and habits. At 26 weeks after the last dose, weight remained stable, with an average additional loss of 1.5%. This is the closest thing the current literature has to a structured taper protocol that actually works.

Where the trial data ends.

The major published trials of GLP-1 discontinuation, STEP 1 extension and SURMOUNT-4, were designed as straightforward comparisons. Patients either stayed on the full dose or were switched abruptly to placebo. There was no gradual taper. In the STEP 1 extension, there was no continued behavioural support either. This is the cleanest way to isolate the medication's effect, but it is not how most patients actually come off these medications in practice. The result is that the trial regain data describes a worst case rather than a typical one. Real-world data from clinics like Embla, along with the Cleveland Clinic cohort, points in a different direction. What happens after the medication depends heavily on how it is stopped and what support is in place when it is. Trial evidence on tapered maintenance is still being built. Clinical practice is, for now, ahead of it.

The South African reality.

For South African patients, the financial reality is harsher than it is in most other markets. Medical aids in South Africa do not currently cover GLP-1 medications for weight management. Obesity is not recognised as a Prescribed Minimum Benefit, which means that even members on comprehensive plans are paying the full retail price out of pocket or through their day-to-day medical savings, roughly R2,700 to R3,700 per month at standard doses, depending on the medication and the pharmacy. A few schemes will cover the medication for people with co-existing type 2 diabetes, but for someone using Ozempic, Wegovy or Mounjaro primarily for weight loss, it is a private expense.

This is the real reason maintenance dosing matters here, not just as a clinical refinement but as a practical question of whether long-term treatment is financially sustainable at all. Someone holding their result on 5mg of Mounjaro is paying meaningfully less than someone holding the same result on 10mg. Stepping down to 0.5mg of Ozempic with structured behavioural support sits in a different financial position from paying for 2.4mg of Wegovy indefinitely. None of this changes the underlying biology of the medication. It changes what is actually affordable. Any honest conversation about long-term GLP-1 treatment in this country has to begin with that fact, not finish with it.

The work the medication doesn't do.

In our companion article on weight regain, we described a 2026 Cleveland Clinic study of nearly 8,000 patients who stopped Ozempic, Wegovy or Mounjaro. The real-world numbers tell a different story from the trial ones. In that study, the average regain at one year was just 0.5%. The authors are clear about why. Real-world patients are not doing what trial patients are forced to do. They are not stopping abruptly with no support and no plan. They are restarting medication, switching to other obesity treatments, or building structured behavioural support around the transition. The medication does pharmacological work. The support around it does structural work, building the patterns, protecting the muscle, holding the result in place when the dose comes down. Neither replaces the other.

Every major published guideline of the last two years agrees on this. The 2025 American Society for Nutrition advisory, the 2025 Frontiers review on exercise and GLP-1s, and the AACE guidelines all describe medication and lifestyle support together as the standard of care for obesity. Patients who receive a GLP-1 alongside structured nutrition guidance lose more weight, stick with treatment longer, and are more likely to hold their results after stopping the medication.

Three things, specifically:

There are three things that structured behavioural support does during GLP-1 treatment that the medication does not do on its own.

The first is protecting muscle. Ozempic, Wegovy and Mounjaro produce significant total weight loss, but a substantial portion of that loss is lean mass. In STEP 1, patients on semaglutide lost around 9.7% in lean mass alongside 19.3% in fat mass. In SURMOUNT-1, roughly a quarter of the total weight lost on tirzepatide was lean mass. Some loss of muscle is normal when anyone loses weight. But these medications also reduce appetite so effectively that patients often eat too little protein without noticing, which makes the muscle loss worse than it needs to be. The Optimizing GLP-1 Therapies review published in late 2025 recommends a protein intake of at least 1.2g per kilogram of body weight per day, evenly distributed across meals, combined with resistance training. Without protein and resistance training, more of the weight lost is muscle rather than fat. The result is a smaller version of the same body, not a meaningfully different one.

The second is what happens to appetite and food noise when the medication is reduced or stopped. On a GLP-1, the constant thinking about food quiets down. Fullness arrives sooner, cravings ease off, and eating becomes less emotionally loaded. But the medication only addresses one side of the equation. Everything else is still there: the stress that drives appetite up, the environments where food is constantly available, and the routines built around eating. The medication does not change those things. It just makes it easier to eat less in spite of them. When the dose drops or the medication stops, what remains is what was there before. This is what structured behavioural support during treatment is for. Not to reinvent the patient, but to give them tools and a plan for living their actual life after the medication is no longer doing the heavy lifting.

The third is the daily habits that hold weight off long-term, which have been studied for nearly thirty years through the National Weight Control Registry. The Registry tracks more than 10,000 American adults who have lost at least 13.6kg and kept it off for a year or more. What the Registry has found is consistent across thousands of people. The ones who hold the loss tend to weigh themselves weekly and eat breakfast. They get about an hour of moderate movement most days, and they eat similarly on weekdays and weekends rather than letting the weekend drift. None of this is dramatic. These are small habits, sustained over years. The Registry data is now long enough that ten-year follow-ups have been published, and the same picture holds: people who maintain these habits maintain their weight loss, and people who lose these habits gradually regain.

How the two fit together.

The honest reading is that the medication and the behavioural support are doing different jobs, and the patient needs both. The medication overrides the body's defended baseline, which makes it possible to eat less without the punishing hunger that derails diet alone. The behavioural support builds the eating, movement, and self-monitoring habits that hold the new weight in place, while protecting the muscle the medication on its own does not.

When the medication is being reduced to a maintenance dose, the habits carry more of the load. When the medication is being tapered toward zero, they have to carry most of it. The Cleveland Clinic patients who held their weight loss after stopping the medication were doing exactly this. They had behavioural support in place. The medication had brought them to the result, but the support was what kept them there.

This is the part of the conversation that gets lost when GLP-1s are discussed as a miracle drug or as a scam. They are powerful medications that produce real, measurable changes in the body's appetite-regulation system while the patient takes them. What happens after that depends on the work done while they were taking them, and the work that continues after.

Starting with the ending in mind.

Most patients arrive with the goal of losing the weight, stopping the medication, and hoping the result holds. The data is unambiguous that this is the approach least likely to work. The body's appetite-regulation system does not reset to a new baseline once weight has been lost. The medication does not retrain the body, and the conditions of the patient's life are still there.

What does work, based on the evidence available now, is one of two approaches.

The first is to treat the medication as a long-term intervention, with the dose stepped down over time to the lowest level that maintains the result. The 2025 AACE guidelines describe this directly, and the real-world data is starting to support it. For South African patients, this option has to account for the financial reality of paying out of pocket. At a maintenance dose, the monthly cost is meaningfully lower than at a weight-loss dose.

The second is to plan, from the start, for a structured taper at the end of treatment, with strong behavioural support built underneath it. The Embla data described earlier in this article suggests that a nine-week gradual taper, combined with continued behavioural support, can hold weight stable in the months after the medication stops. This second option is not yet supported by randomised trial evidence at the same level as the first, but the early observational data is encouraging.

The third option, which is the one most people imagine when they first consider GLP-1 treatment, is to lose the weight on the medication and then stop. The trials are clear that this produces substantial regain within a year. It is the version of treatment the regain statistics describe, and the one current clinical practice has moved away from.

The choice between the first two options is not one to be made alone, and certainly not from an internet article. It is a clinical decision, made with a team that knows the patient's medical history, financial situation, and the support that will be in place during and after the medication. Made well, at the beginning, it is the difference between starting a GLP-1 to lose weight and starting one to change the path.