Metabolic Health · Maintenance
Most of what has been written about GLP-1 based therapy describes the first year of treatment, covering the titration schedule, the management of gastrointestinal effects, and the falling number on the scale. Considerably less has been written about the phase that follows, which is longer, quieter, and where the durability of the result is actually determined.
The active reduction phase occupies roughly the first 60 to 80 weeks of treatment, while whatever follows occupies the remainder, and the questions that define the second phase differ enough from the first that patients who did well early are often unprepared for what comes next. The most useful question to answer at that point is what remains to be treated once the weight has already been lost, because the answer reframes every practical decision that follows.
What continues to be addressed during maintenance is not the excess weight itself but the set of physiological adaptations that follow any substantial weight reduction and act to restore the previous body mass. Those adaptations involve appetite signaling, satiety hormones, and energy expenditure, and the published literature indicates that they persist for years rather than resolving once a target weight is reached. Maintenance therapy is the intervention that holds them in check, which is why continuing treatment is a clinically meaningful act rather than a redundant one.
The SURMOUNT-4 trial examined this directly through a randomized withdrawal design, and its findings are the reference point for most of what follows in this article. Every figure described here is drawn from published research and is presented for education, while decisions about continuing, reducing, or stopping any medication belong with the prescribing clinician. Individual results vary.
SURMOUNT-4 enrolled adults into an open label period of tirzepatide treatment lasting 36 weeks, during which participants achieved a mean weight reduction of approximately 21 percent. Participants were then randomized either to continue the medication or to receive placebo for a further 52 weeks, with lifestyle support continuing in both groups throughout.
Participants who continued treatment lost further weight over the second period, reaching a mean total reduction of roughly 25 percent from baseline. Participants randomized to placebo regained a mean of approximately 14 percent of body weight, which returned them to a total reduction near 10 percent and gave back close to half of everything the open label period had achieved. Since both groups received the same lifestyle support, the divergence reflects what the medication was doing rather than differences in effort or adherence to diet.
Figure 1
Mean Weight Change From Baseline During the Open Label Period and the Randomized Withdrawal Period
Sources: Mean results reported from the SURMOUNT-4 randomized withdrawal trial of tirzepatide, published in JAMA in 2024. Values are rounded and intermediate points interpolated for legibility. Group means describe a study population under trial conditions and do not predict any individual outcome. Individual results vary.
Figure 2
Fate of the Weight Lost During the Open Label Period Among Participants Switched to Placebo
Sources: Calculated from the mean weight reduction figures reported in SURMOUNT-4, comparing the mean reduction at randomization with the mean reduction at week 88 in the placebo group. Proportions are approximate and describe group means rather than the experience of any individual participant. Individual results vary.
Figure 3
Share of Participants Maintaining at Least 80 Percent of the Weight Lost During the Open Label Period
Sources: Proportions reported in SURMOUNT-4 for participants maintaining at least 80 percent of the body weight lost during the open label period, rounded for illustration. These are trial population proportions observed under study conditions and are not a prediction of any individual result. Individual results vary.
Weight reduction of any meaningful size triggers a coordinated set of compensatory changes that are well characterized in the physiology literature and are not specific to medication, since they appear after dietary restriction and after intensive lifestyle programs as readily as they appear after pharmacotherapy.
Leptin. This hormone signals adequate fat stores to the hypothalamus, and it falls in proportion to the fat mass that has been lost, frequently declining further than the change in fat mass alone would predict.
Ghrelin. Concentrations of the principal hunger signal rise after weight reduction, which increases the drive to eat at precisely the point when intake needs to remain lower than it was before.
Peptide YY and cholecystokinin. Both of these satiety signals decline, so meals produce less fullness at the reduced weight than the same meals produced beforehand.
Resting energy expenditure. Daily energy requirements fall partly because a smaller body costs less to operate and partly through an additional adaptive component that exceeds what the change in body size alone would predict.
The combined result is a body that is hungrier, more metabolically efficient, and less readily satisfied at its new weight than the body of someone who has always been that weight. Studies following participants after large intentional reductions have found several of these changes still measurable years afterward, which is the mechanism behind the long standing observation that maintaining a reduction is harder than achieving one.
Figure 4
Direction of Reported Change in Appetite and Energy Regulation Following Substantial Weight Reduction
Sources: Directions of change described across the weight reduction physiology literature, including longitudinal studies reporting persistence of several of these adaptations for a year or more after intentional weight loss. Bar heights are schematic and indicate direction rather than quantified effect size. These are population level physiological patterns and do not describe any individual. Individual results vary.
This is the most common question raised at the point of transition, and it has the least tidy answer. The reasoning behind it is sound enough, since the maximum tolerated dose was required to drive the reduction and a lower dose might plausibly be sufficient to hold the result, but in practice this succeeds for some patients and not for others without a reliable way to identify which group someone belongs to in advance.
What can be stated from the trial evidence is that maintenance was studied at continued dosing rather than at reduced dosing, because the continuation group in SURMOUNT-4 remained on their maximum tolerated dose of either 10 or 15 milligrams. No randomized evidence establishes that a reduced maintenance dose preserves the result, which means that any reduction during maintenance represents a clinical judgment made without direct trial support rather than a validated protocol.
What clinicians describe observing is a wide spread of responses. Some patients hold their weight comfortably at a substantially lower dose than the one on which they lost it, while others find that appetite returns within a few weeks of any reduction and that weight follows shortly afterward. A cautious approach involves reducing by a single step and observing for several weeks before deciding anything further, with a willingness to return to the previous dose if hunger, preoccupation with food, or a sustained upward trend in weight reappears. Returning to the previous step is not a failure of the attempt but rather information about where that individual's effective maintenance threshold sits.
| Approach after reaching goal weight | What it involves | Evidence support |
|---|---|---|
| Continuation at the loss phase dose | The dose that produced the reduction is maintained without change | Direct trial evidence |
| Single step reduction with observation | The dose is lowered one level and reassessed over several weeks, reversing if appetite or weight drifts | Clinical judgment only |
| Extended interval dosing | Injections are spaced further apart than the labeled weekly schedule | Not label supported |
| Discontinuation | Treatment stops entirely with lifestyle measures continuing | Studied, with regain the common outcome |
Summary of approaches discussed in clinical literature, presented to describe where evidence exists rather than to recommend any course of action. Dosing decisions are individualized and belong with the prescribing clinician. Individual results vary.
Extended interval dosing arises frequently in this conversation and deserves a direct answer. Tirzepatide has an elimination half life of approximately five days, which is the property that supports weekly administration in the first place, so lengthening the interval produces a trough during which drug concentrations fall well below the range that the labeled schedule maintains. The practical consequence tends to be a cyclical return of appetite during the days preceding the next injection rather than a stable lower level of exposure, and the approach is neither described in the labeling nor established as a maintenance strategy.
Figure 5
Illustrative Drug Concentration Across a Weekly Schedule Compared With a Lengthened Injection Interval
Sources: Schematic illustration constructed from the reported elimination half life of tirzepatide of approximately five days and the labeled once weekly dosing schedule. Curves are drawn to convey the concept of a concentration trough and are not measured pharmacokinetic values. Dosing intervals other than those in the approved labeling are not established maintenance strategies. Individual results vary.
Maintenance is not simply the loss phase conducted at a lower intensity, because several things are genuinely different once the scale stops moving.
Energy requirements are permanently lower. A body carrying substantially less mass requires meaningfully fewer calories to maintain than it did previously, and the adaptive component of the decline in expenditure widens that gap slightly beyond what body size alone would predict, so an intake that represented maintenance at the former weight represents a surplus at the new one.
Protein and resistance training matter more rather than less. The lean tissue preserved during the reduction phase has to remain preserved, while the appetite suppression that made a calorie deficit straightforward also makes it easy to under consume protein indefinitely, so both variables need to persist into maintenance rather than being treated as tactics belonging only to the earlier phase.
Micronutrient adequacy becomes a longer horizon question. Months of reduced total intake can produce deficiencies that take considerable time to become symptomatic, with iron, vitamin B12, and vitamin D among the more commonly discussed, which is why periodic assessment is preferable to assuming that a stable weight indicates stable nutrition.
The useful measurements change. Weight served as a reasonable proxy for progress during the reduction phase, whereas during maintenance it distinguishes poorly between holding a given weight with lean tissue preserved and holding the same weight while gradually exchanging muscle for fat, so body composition, strength performance, waist circumference, and metabolic laboratory values give a more accurate picture.
The monitoring plan should thin out relative to the reduction phase without disappearing altogether, and the following description reflects approaches discussed in clinical practice rather than a schedule appropriate for any particular patient.
A general description of monitoring approaches discussed in clinical practice, presented for education rather than as a protocol. Assessment intervals and their content are determined by the treating clinician. Individual results vary.
There is a persistent instinct to regard indefinite medication as a lesser outcome than a durable result achieved without it, as though continuing treatment indicates that the underlying problem was never properly resolved. That framing invites a comparison worth examining, because medications for blood pressure are not withdrawn once blood pressure normalizes and lipid lowering therapy is not stopped once cholesterol falls, since in both cases the normalized value is the evidence that the treatment is working rather than a signal that it can be removed.
Obesity has substantial evidence supporting a similar model as a chronic condition with strong biological drivers in which withdrawal of effective treatment is commonly followed by return of the condition, and SURMOUNT-4 is a fairly direct demonstration of that pattern. None of which settles what any individual should do, because cost, tolerability, personal preference, and clinical circumstances all bear on whether continued therapy is the right choice, and discontinuation remains a legitimate decision that many people make for entirely reasonable reasons.
The narrower point is that the decision deserves to be made deliberately, with the expected outcome of each path understood beforehand, rather than arrived at by default when a prescription lapses or a goal weight is reached and no plan exists for what follows.
The maintenance phase lasts longer than the reduction phase and receives far less attention, yet it is the period during which the durability of the entire result is determined.
Frequently Asked Questions
A 3 minute intake is all it takes. A physician reviews your information and identifies the protocol matched to your specific health profile.
Get Started