Nearly everyone who takes one of these drugs long enough watches the weekly drop get smaller and then stop. It arrives as a personal failure and it is a property of the curve. What the trials can say about when it happens is fairly precise. What they can say about getting past it is much thinner. That gap is where a good deal of upselling lives, including the dose ladders priced on the flat-dose board.
The curve has a knee, and it has been located
The longest weight analysis in this class followed treatment for four years in adults with cardiovascular disease and obesity. Loss ran on for roughly 65 weeks before it stopped falling, and what it then did was hold, not drift back, through week 208.[1]
Fifteen months is therefore a reasonable prior for when to expect the slope to flatten, and flattening is not the same event as regaining. Those two are constantly confused in the same sentence, and only one of them is a reason to change anything.
Before that point, the drug is usually still working
A withdrawal trial makes the earlier part of the curve visible. After a 20-week run-in on semaglutide, participants who continued treatment lost a further 7.9% of body weight between week 20 and week 68, while those switched to placebo put 6.9% back on.[2] A stall at month four or month six sits inside a window in which the average participant was still descending.
That is one reason a stall during the climb to a maintenance dose is a different object from a stall two years in. The escalation schedule and what it costs are set out in the titration article.
More drug: the one randomized test
The obvious response to a plateau is a bigger dose, and it has now been tried properly. STEP UP randomized 1,407 adults with a body-mass index of 30 or above and no diabetes, in a 5:1:1 ratio, to semaglutide 7.2 mg, semaglutide 2.4 mg or placebo for 72 weeks, all with a lifestyle intervention.[3]
Mean weight change was −18.7% on 7.2 mg against −15.6% on 2.4 mg, an estimated treatment difference of 3.1 percentage points (95% CI, −4.7 to −1.6; P < 0.0001), and −3.9% on placebo.[3] At the deep end the gap widened: the odds of reaching a 25% reduction were 2.4 times higher on the top dose than on 2.4 mg (95% CI, 1.6 to 3.5).[3]
The tolerability column is the part a dose-increase pitch tends to leave out. Gastrointestinal adverse events were reported by 70.8% of the 7.2 mg group against 61.2% on 2.4 mg and 42.8% on placebo, and dysesthesia — altered or unpleasant skin sensation — by 22.9% against 6.0% and 0.5%.[3] Nearly one participant in four on the top dose reported a symptom that one in sixteen reported a rung below.
Two limits travel with that result. Everyone was randomized at baseline, so this is a trial of a higher dose rather than a trial of rescuing someone who has stalled. And 7.2 mg is above the labeled ceiling most sellers on the seller index are dispensing against.
Switching molecules, and what that trial did not ask
The other standard suggestion is to change drug. A 72-week open-label trial randomized 751 adults with obesity and no diabetes to the maximum tolerated dose of tirzepatide or of semaglutide. Mean weight change was −20.2% (95% CI, −21.4 to −19.1) against −13.7% (95% CI, −14.9 to −12.6), and waist circumference fell 18.4 cm against 13.0 cm.[4]
Those participants were starting treatment, not stalling on it. The trial establishes that one molecule outperforms the other from a standing start; it does not establish what happens when a person who has plateaued on semaglutide moves across to the drug costed on the tirzepatide board. Nobody has randomized that question.
The physiology everyone invokes
Plateau explanations usually reach for metabolic adaptation, and the best-known measurement of it is small and extreme. Fourteen contestants from a televised weight-loss competition were followed for six years. Resting metabolic rate was 704 kcal per day below baseline at six years, and the adaptation component — the residual after adjusting for body composition and age — was −499 kcal per day.[5]
The correlations complicate the folk version. Adaptation measured at the end of the competition did not predict regain (r = −0.1, P = 0.75). The people holding the largest losses at six years were the ones showing the greatest concurrent metabolic slowing (r = 0.59, P = 0.025).[5] Slowing tracked the effort to stay smaller rather than sabotaging it.
Those were 14 people losing a mean of 58.3 kg under television conditions, with no drug involved. It is a reason to take adaptation seriously as a mechanism and not a number anyone should apply to themselves.
What has not been tested at all
Three things, stated plainly. No trial has randomized a dose increase triggered by a plateau, which is the exact clinical decision people face. No trial has tested a pause, a diet break or an intermittent schedule against continuous treatment. And no plateau evidence exists for compounded preparations, since every trial above used branded product.
There is also a plainer explanation that gets skipped. A flat scale can mean the dose has reached its effect, or that intake has drifted, or that a week was missed, and none of those look different from the outside. Distinguishing them is clinical work, and a person watching their weight stall belongs in a conversation with a prescriber rather than in a dosing experiment.
Reading a stall on a pricing page
Commercially, a plateau is when a plan’s dose policy stops being theoretical. If the answer to a stall is a higher rung, the question is whether the price climbs with it, and a plan that charges by milligram turns a clinical decision into a budget one.
The second question is what a seller promises about it. A program advertising that it breaks plateaus is describing something no randomized trial has demonstrated, whatever the drug in the vial. Continued treatment is what the durability evidence covers, and stopping is the one move with a measured consequence. How any figure on this page was established is set out in the methodology.