Semaglutide Discontinuation Rates 2026: Who Stops and What's Next

Nearly half of adults who started semaglutide for obesity stopped within a year, a 157,822-person Danish registry study found. A separate 2026 model of the trials estimates weight returns at about 1.04 kg per month after stopping. This is not medical advice — talk to your prescriber.

Nearly half stop within a year: what the Danish data show

Two papers published days apart at the end of August 2026 answer different halves of the same question. One counted how many people stop semaglutide. The other modelled what happens to body weight afterwards. They are separate literatures, and their authors say so.

How the 49% was counted

The cohort followed 157,822 first-time semaglutide-for-obesity initiators in Denmark between Dec. 1, 2022 and Nov. 1, 2024 — 108,551 women (69%) and 49,271 men (31%), median age 49, all semaglutide-naive adults without diabetes, for 12 months. Using a 60-day grace period, 49% had discontinued by 12 months: 77,483 people.

The grace period is not a footnote. The same cohort yields 58% at a 30-day gap and 45% at 90 days. Cite the 49% only with its basis attached, because the headline moves by 13 percentage points depending on how long a prescription gap counts as a stop.

The curve behind the headline

Discontinuation accumulates early. By 3 months 13% (20,208 people) had stopped, by 6 months 27% (42,477), and by 9 months 39% (60,918). The curve then flattens toward the 12-month figure — most of the attrition happens in the first two quarters.

Who was more likely to stop

Risk ratios here are relative, not absolute. Adults aged 18–29 had a risk ratio of 1.49 (95% CI 1.47–1.52) versus ages 45–59 — relative risk was 49% higher, not “49% stopped”. Men versus women: 1.16 (1.14–1.17). Prior psychiatric medication: 1.13 (1.12–1.14), a subgroup worth reading alongside the 2026 semaglutide psychiatric-hospitalization study. Prior GI medication: 1.10 (1.09–1.12). Prior bariatric surgery was the one lower-discontinuation group at 0.89 (0.86–0.93). No stratum of baseline characteristics fell below 41%.

About 1 in 4 who stopped restarted within 3 months

Discontinuation was not always permanent. One of four discontinuers reinitiated semaglutide within 3 months. A further 1,088 people (1.4%) filled liraglutide and 563 (0.7%) filled tirzepatide after stopping, though tirzepatide only reached the Danish market on Oct. 30, 2024, which shortens the window that switch had to happen in.

The registry’s limits bound everything above. It carries no BMI or other anthropometric data, no individual-level income, insurance or out-of-pocket data, and no weight-loss data after initiation. The authors also say the findings may not be fully generalisable beyond Denmark.

Why people stop

Cost

Semaglutide for obesity is, in the study authors’ words, “expensive … and is not currently reimbursed, creating a considerable economic burden for patients”. The paper puts the cost at roughly €2,200–€3,700 ($2,500–$4,200) a year in Denmark. That is the same arithmetic behind the employer and insurer coverage retreat, and it lines up with what GLP-1s cost per month in the United States.

Side effects and tolerability

Gastrointestinal adverse effects are common with this drug class, and pre-existing GI conditions may raise susceptibility. That plausibly explains the 1.10 relative risk for prior GI medication in the same cohort. The 2026 GLP-1 hair-loss studies belong to the same side-effect literature: bothersome, usually reversible, and a reason some patients raise the question of stopping with a clinician. That literature has a nutritional corner too: persistent vomiting and rapid weight loss are the setting a February 2026 Clinical Nutrition review flags for thiamine (vitamin B1) deficiency, and the 15 reported cases of Wernicke encephalopathy in people taking GLP-1 medicines are what that review found. The class has a vision-loss signal as well, and it is the one now in front of a federal court: see the NAION GLP-1 vision-loss lawsuits.

Access and supply

Availability is not only a pharmacy question. Compounding rules, the FDA delist fight and supply limits all sit between a prescription and a refill — the subject of our 2026 compounding ruling coverage.

Feeling “done”

The regain-model paper lists cost, tolerability, access restrictions, patient preference, treatment fatigue, and uncertainty about the optimal duration of therapy as the reasons people stop. Duration uncertainty is the one no one can resolve for a patient in advance: no source here sets a maximum treatment length.

What the cohort can attribute is associational. It measured who stopped, using prescription records — not why each person did, and not what they weighed afterwards. Reported side effects are one of the recorded reasons people stop, and the 2026 Penn analysis of 410,198 Reddit posts is the current evidence on underreported GLP-1 side effects — reported symptoms, not proven causes.

What happens after you stop

Read the next numbers with their type attached: this is a reconstructed aggregate-data Bayesian re-analysis of six already-published studies, ten intervention arms and 1,776 participants — not a new trial, and not new participants. No updated literature search was performed; the source review ran through February 2025.

Why regain follows at all

The mechanism is withdrawal of support, not a lapse in effort. In the analysis’s words: “As these agents act partly by reducing appetite and improving satiety during active treatment, withdrawal of therapy may remove the physiological support that enabled weight loss, resulting in progressive weight regain”.

The modelled trajectory

Mean weight loss at cessation was 15.35 kg (95% CrI 11.18–19.60). From there the model projects regain at 1.04 kg per month (95% CrI 0.80–1.29), half of the initial loss back by 7.50 months (5.05–10.57), and a return to baseline by 15.00 months (10.10–21.13).

The last figure needs its label: observed post-discontinuation follow-up ran 4 to 52 weeks, so the 15.00-month point is the authors’ own linear extrapolation beyond that observation window, not a measured outcome. The curve is a constant linear slope by choice, because follow-up was sparse.

What the curve looks like at 6, 9 and 12 months

Modelled mean weight change from baseline: −9.12 kg (95% CrI −13.64 to −4.68) at 6 months, about 41.4% of the initial loss regained; −6.01 kg (−10.82 to −1.26) at 9 months, about 62.1%; −2.90 kg (−8.04 to 2.22) at 12 months, about 82.8%. These are average-trajectory model probabilities, not individual-patient risk.

What the model does not say

Behavioural or lifestyle support after stopping showed a point estimate of −0.19 kg/month (95% CrI −0.57 to 0.17) and an 87.2% probability of slower regain — inconclusive, with a credible interval that includes zero. On the drug question, tirzepatide versus semaglutide came in at −0.04 kg/month (95% CrI −0.46 to 0.44), a 40.2% posterior probability of faster regain; the authors conclude the evidence “does not establish that either drug has an inherently faster post-discontinuation rebound pattern”.

Trial-predicted regain vs modelled real-world regain

The September 2026 estimates read best next to the numbers that came before them — including the one that comes from an actual off-treatment trial arm rather than a model.

One caveat binds the table: the September 2026 analysis reuses the same six studies and ten arms as the review it draws on — “study overlap was complete by design”. It is a re-estimate of the same evidence, not an additional body of evidence.

How long can you stay on it?

Neither paper sets a maximum duration, and nothing in them answers the question for an individual. What the data show is behavioural: in the Danish cohort 49% were off semaglutide by 12 months (45–58% depending on the grace period), and about 1 in 4 of those restarted within 3 months. The trials behind the regain model observed off-treatment follow-up of 4 to 52 weeks. Ozempic and Wegovy are brand names for semaglutide; how long to continue it is a clinical decision, and this page carries no dosing or stopping instructions.

Planning to stop: what the evidence can and cannot tell you

A supervised taper is not established to do better than stopping abruptly. The analysis states plainly that the data “do not determine whether tapering, lower-dose maintenance, switching, intermittent treatment, or any specific behavioural programme prevents regain”. No taper-versus-abrupt comparison was performed, and the paper’s own predictor for behavioural support after stopping was inconclusive.

The reframe matters more than the caveat. Regain after stopping is pharmacology — the withdrawal of appetite and satiety support the drug was providing — not personal failure, and not evidence that someone did the wrong thing. Nothing in either paper supports the idea that willpower or a particular programme prevents it, and nobody should read a model’s average trajectory as a verdict on their own discipline. If a prescriber and patient are weighing what comes after a GLP-1, our comparison of the approved oral options covers what is actually approved.

Frequently asked questions

Who is most likely to stop semaglutide?

In a 157,822-person Danish cohort (JAMA Network Open 2026;9(8):e2631371), 49% of first-time semaglutide users for obesity had discontinued within 12 months (60-day grace period; 45–58% across grace periods), and no baseline stratum fell below 41%. Relative risk was higher at ages 18–29 (RR 1.49 vs 45–59), in men (1.16), with prior psychiatric medication (1.13) and with prior GI medication (1.10); prior bariatric surgery was lower (0.89).

What happens when you stop semaglutide?

The appetite and satiety support the drug provided is withdrawn. A September 2026 re-analysis of trial off-treatment arms (Kow et al., Endocrinology, Diabetes & Metabolism 2026;9(5):e70325) puts average regain at 1.04 kg per month (95% CrI 0.80–1.29), with about 41.4% of the lost weight back by 6 months.

Will I regain the weight?

That model (Kow et al. 2026) projects half the initial loss regained by about 7.50 months (95% CrI 5.05–10.57) and a return to baseline by 15.00 months (95% CrI 10.10–21.13), the second figure extrapolated beyond 52 weeks of observed follow-up. It puts the regained share at about 62.1% at 9 months and 82.8% at 12 months. These are average-trajectory model probabilities, not individual-patient risk. This is not medical advice — talk to your prescriber.

How long can you stay on it?

Neither source sets a maximum duration, and the trials behind the model observed off-treatment follow-up of 4 to 52 weeks. In the cohort 49% were off by 12 months (45–58% across grace periods), and about 1 in 4 of those restarted within 3 months. Whether to continue is a decision for you and your prescriber.

Is a supervised taper better than stopping abruptly?

Not established. The analysis says the data “do not determine whether tapering, lower-dose maintenance, switching, intermittent treatment, or any specific behavioural programme prevents regain”. No taper-versus-abrupt comparison was performed, so nothing in either paper supports reading a taper as the proven route.

Does tirzepatide cause faster regain?

No clear difference was found. Adjusted meta-regression put tirzepatide at −0.04 kg per month versus semaglutide (95% CrI −0.46 to 0.44) — a 40.2% posterior probability of faster regain, which the authors read as no established difference between the two drugs.

More general questions about peptides, compounding and GLP-1 medicines are answered in our peptide FAQ.

Sources

1. Mailhac A, et al. Discontinuation of Semaglutide Therapy for Obesity Management. JAMA Network Open. 2026;9(8):e2631371 — 157,822 first-time initiators in Denmark, Dec. 1, 2022–Nov. 1, 2024.
2. Kow CS, et al. Weight Regain Trajectories After Discontinuation of Semaglutide or Tirzepatide: A Reconstructed Aggregate-Data Bayesian Longitudinal Meta-Analysis. Endocrinology, Diabetes & Metabolism. 2026;9(5):e70325 — six studies, ten arms, 1,776 participants; no updated literature search after February 2025.
3. Butler K. High Semaglutide Dropout Rates Signal Adherence Challenge. Medscape, Sept. 10, 2026 — via Wayback Machine capture 20260910132136 — the live Medscape page is behind a bot wall; the figures cited here come from the primary papers.
4. Wilding JPH, et al. Weight regain and cardiometabolic effects after withdrawal of semaglutide: The STEP 1 trial extension. Diabetes, Obesity and Metabolism. 2022;24(8):1553–1564 — a separate 2022 trial-extension estimate, not part of the September 2026 items.

PeptidesBeat is an independent editorial publication covering peptide policy, research, and industry developments. We do not sell peptides, recommend dosing, or provide medical advice. All content is informational. Peptides referenced may be subject to FDA restrictions; consult a licensed healthcare provider for any therapeutic question.