Stopping the semaglutide tablet: the five-week tail, and what comes back

Before you read. This page explains what the FDA label and the published withdrawal data say. It is educational and cannot assess your situation. Stopping a prescribed medicine is a decision for you and your prescriber, and it matters more than usual here: if you take this for type 2 diabetes, stopping changes your glucose control, and if you also take insulin or a sulfonylurea, the doses of those may need revisiting. Do not stop, restart or change a dose on the strength of a web page.

Quick answer

There is no taper in the label. There is also no withdrawal syndrome in it, and that is not an omission — it is a consequence of how slowly this drug leaves.

Two numbers do most of the explaining. Semaglutide's elimination half-life is approximately one week, and it is present in the circulation for about five weeks after the last tablet. Whatever "stopping" means, it does not mean a switch flipping. It means a five-week slope.

Which is why the phrase cold turkey imports the wrong picture. A drug you stop abruptly and clear in hours is one thing. A drug that is still measurably present more than a month later is a different one, and nothing in the label describes a rebound, a discontinuation reaction, or a reason to step down through the strengths on the way out.

What does come back is what the drug was doing. Glucose control, appetite, and the things that follow from appetite. Those return on their own schedule, and the schedule is the useful part of this page.

Keep tracking through the gap, not just during treatment →

The two numbers that govern everything after the last tablet

Both are in the label's clinical pharmacology section, and both are unusual for a swallowed drug.

Elimination half-life: approximately one week. Semaglutide is more than 99% bound to plasma albumin, which is the trick that makes it last — albumin binding shields the peptide from being broken down and from being cleared by the kidney. Absolute clearance is around 0.04 litres per hour, which is slow.

Present in the circulation: about five weeks after the last dose. That is the label's own phrasing, and it follows arithmetically from the first number. After one week roughly half remains, after two weeks a quarter, and by five weeks the level is low enough that the label stops describing it as present.

Neither number changes because you swallowed it rather than injected it. What the tablet changes is how much gets in on the way — under 1% of each dose, which is the whole reason the morning has rules. Once it is in, it behaves like semaglutide.

Time since last tablet	Roughly what is left
1 week	About half
2 weeks	About a quarter
3 weeks	About an eighth
5 weeks	The label stops describing it as present

Treat the middle rows as arithmetic rather than measurement. The label states the half-life and the five-week figure; the intermediate steps are what those two facts imply, not separate findings.

There is no taper instruction, and that is not an oversight

Look for one in section 2 and it is not there. The dosage section climbs — 3 mg, then 7 mg, then 14 mg, each step held for 30 days — and says nothing about coming back down.

For most medicines, a taper exists because stopping abruptly causes something: a rebound, a withdrawal syndrome, an adrenal problem. None of those is described for semaglutide, and the pharmacology explains why. The drug tapers itself. Stop on a Tuesday and your level a week later is roughly half what it was, a fortnight later a quarter. You cannot step down faster than the molecule does.

That is a genuine finding rather than a reassurance. It is also the reason people are sometimes surprised at how late things change — the week after stopping can feel like nothing has happened, because pharmacologically not much has.

The washout number the label gives you, in an unexpected place

Section 8.3, under Females and Males of Reproductive Potential, contains the clearest statement of what a full washout looks like:

Discontinue RYBELSUS or OZEMPIC tablets in women at least 2 months before a planned pregnancy due to the long washout period for semaglutide.

Two months. That is the interval the label considers adequate when the requirement is that the drug be gone, and it is comfortably longer than the five weeks quoted in the pharmacology section — a margin, deliberately.

It is worth knowing even if pregnancy is not the reason you are asking, because it is the only place the label puts a number on "cleared" rather than "present". If you have wondered how long semaglutide stays in your system, that sentence is the label's answer to the strict version of the question.

Side effects after stopping, honestly

This search has two different questions inside it, and they have different answers.

"Will stopping cause side effects?" The label lists no discontinuation reaction. The adverse reactions it does list — nausea, abdominal pain, diarrhoea, decreased appetite, vomiting, constipation — are reactions to the drug being present, and the label ties the gastrointestinal ones to dose escalation specifically. Removing the drug removes the cause.

"How long will the side effects I already have last after I stop?" Here the five-week tail cuts the other way. If you stopped because of nausea, the nausea does not end on the day you stop, because the drug is still there. Expect the fade to track the level rather than the decision. What the label reports and when covers the pattern during treatment; after stopping, the same symptoms fall away along the same slow curve.

One thing that genuinely can change quickly is glucose, and it is the reason the medical note at the top of this page is not boilerplate. If you take insulin or an insulin secretagogue alongside this — a sulfonylurea, for instance — the label already warns that the combination raises hypoglycaemia risk, and doses of those medicines are sometimes set with the semaglutide in mind. Removing one leg of a regimen without telling the person who built it is the actual risk in stopping, and it is not a gastrointestinal one.

The weeks after stopping are the weeks people record least and ask about most. Takely keeps logging when there is no dose to log — symptoms, appetite, weight, a line a day — so the question "when did that change?" has an answer later. Open Takely in your browser →

Appetite, and the part people are least prepared for

The label's route into this is oblique: decreased appetite is listed as an adverse reaction, reported by 9% of patients on 14 mg and 6% on 7 mg, against 1% on placebo.

Read that as a description of what the drug does rather than as a complaint, and its mirror image tells you what stopping does. The appetite suppression is the effect. It leaves when the drug leaves.

People are usually braced for weight to move and are less braced for the sensation to come back first — the background hum about food that had gone quiet. There is a whole literature on that experience, and what actually has evidence for quieting it is a separate question from what the medicine was doing.

What the withdrawal data show, and what they do not

The best-known evidence on stopping comes from the STEP 1 trial extension, and it needs its context stated before its finding.

STEP 1 studied once-weekly subcutaneous semaglutide 2.4 mg in 1,961 adults with obesity or overweight and without diabetes, for 68 weeks. At week 68 the treatment and the lifestyle programme were both withdrawn, and a subset of 327 participants was followed for a further year.

The finding: one year after withdrawal, participants had regained about two-thirds of the weight they had lost, and the cardiometabolic improvements seen during treatment had reverted towards baseline for most variables.

Now the caveats, which matter more than usual here. That was a different formulation at a different dose, in a population without diabetes, and the lifestyle intervention stopped at the same time as the drug — so the design does not separate the two withdrawals. It is not a study of stopping Rybelsus, and treating its number as a prediction for a tablet would be exactly the cross-trial reasoning this site declines to do elsewhere.

What it does establish is the shape of the thing, which the authors state plainly: the results confirm the chronicity of obesity and suggest that continued treatment is needed to maintain the improvements. Stopping is not a graduation, and nobody in that extension was told it would be.

Stopping for a reason, and stopping by accident

These get discussed as one topic and behave differently.

Stopping because it is not tolerable. The label's own numbers say this is not rare: 8% of patients on 14 mg and 4% on 7 mg discontinued because of gastrointestinal reactions, against 1% on placebo. If that is where you are, the conversation to have is about whether the dose comes down rather than whether the drug goes away, since the ladder exists precisely because reactions cluster at the steps.

Stopping because of a procedure. A hold for surgery or endoscopy is a different thing with a different logic, and holding it for a procedure covers what the current guidance suggests for a daily tablet.

Stopping because of supply or cost. This is the most common unplanned version and the least discussed one. It is worth naming as a medical event rather than an administrative one — an interrupted prescription has the same pharmacology as a decision to stop, and your prescriber cannot account for it if they do not know it happened. What the tablet costs and why the savings card may not help covers the version of this problem that arrives with a pharmacy quote.

Restarting, and the gap in the label

Here is a question the label does not answer, and it is better to say so than to improvise.

Section 2 tells you how to start from zero: 3 mg for 30 days, an initiation dose the label states is not effective for glycemic control, then 7 mg, then 14 mg if more is needed. It also gives a rule for a single missed dose — skip it, take the next one the following day.

What it does not give is a rule for resuming after a break of weeks or months. Whether you restart at the strength you were on or climb again is a clinical judgement, and the inputs are how long the gap was, what your glucose has done, and how you tolerated the ladder the first time. Ask before the gap rather than after, if there is any warning.

One thing worth writing down before you stop. Your current dose, the date of your last dose, and how the last dose increase went. Those three facts are what a restart conversation needs, and they are the ones that evaporate fastest — particularly if the break turns out to be longer than planned.

What is worth tracking in the weeks after

The instinct is to stop logging when there is no dose to log. It is the wrong instinct, because the questions that arrive in weeks three to eight are all about timing.

Four things are worth a line a day. Appetite — not weight, the sensation, because it moves first. Any symptom that was present at the end, so you can see it fade rather than wonder. Glucose, if you monitor it, which is the one your prescriber will ask about specifically. And the date of the last tablet, because every other question on this page is measured from it.

Keeping the weeks after on the same timeline as the weeks before is the only way "it changed around three weeks in" turns into something a clinician can use. Memory compresses gaps; a log does not.

Frequently asked questions

How long does Rybelsus stay in your system?

The label states that semaglutide's elimination half-life is approximately one week and that it is present in the circulation for about five weeks after the last dose. For a stricter definition of cleared, the label's pregnancy guidance is the one to look at: it instructs discontinuation at least two months before a planned pregnancy because of the long washout period.

Do I need to taper off Rybelsus?

The label contains no taper instruction and describes no withdrawal reaction. With a one-week half-life the drug declines gradually whether or not you step the dose down. Whether a taper makes sense for you is still a prescriber's call, since the reason for stopping shapes the answer.

Is it dangerous to stop Rybelsus suddenly?

The label does not describe a discontinuation reaction. The risk in stopping is generally about what the medicine was managing rather than the stopping itself — glucose control in type 2 diabetes, and the doses of any insulin or sulfonylurea that were set alongside it. That is why this belongs in a conversation rather than a decision.

Will the side effects stop as soon as I stop taking it?

Not immediately. The same slow elimination that makes a taper unnecessary also means symptoms fade along the drug's curve rather than the calendar. The label ties most gastrointestinal reactions to dose escalation, so people who stop from a stable maintenance dose are usually starting from a lower symptom burden anyway.

Will I regain the weight?

The best evidence on withdrawal comes from the STEP 1 extension, which studied a weekly injection at a different dose in people without diabetes and withdrew the lifestyle programme at the same time. In that setting, about two-thirds of the lost weight was regained over the following year and the cardiometabolic improvements largely reverted. It is not a study of this tablet, and the honest summary of the whole literature is that continued treatment is what maintains the effect.

Can I restart at the dose I was on?

The label does not say. It gives a starting ladder for people beginning from zero and a rule for one missed day, and nothing for resuming after a longer gap. That gap in the labelling is real, which is why the answer has to come from your prescriber rather than from a page.

Primary sources and further reading

Trial data and label sections both change. Everything above was checked against those sources on 8 September 2026.