Why half of GLP-1 patients stop in the first year, and what changes the trajectory.

1. Why persistence matters more than peak weight loss

Most public conversation about GLP-1 therapies focuses on how much weight people lose. The clinically meaningful question is different: how long do they stay on treatment, and how stable is the result. Obesity is a chronic disease. Anti-obesity medications work for as long as they are taken. When semaglutide is stopped after a year of effective treatment, participants in the STEP-1 trial extension regained two thirds of their lost weight within twelve months, and most cardiometabolic improvements reverted toward baseline.³ The decisive variable for outcomes is not the peak loss at month six. It is whether the person is still on therapy at month twelve, eighteen, twenty-four.

This is also where the evidence base diverges most sharply between the trial bubble and routine care.

2. The real-world numbers, against the trial bubble

In STEP-1, the pivotal trial of weekly semaglutide 2.4 mg, 94% of participants completed the 68-week trial and 81% remained on the assigned dose throughout.² SURMOUNT-1, with tirzepatide 5, 10 and 15 mg, showed similar discipline. These numbers reflect a study population that received the medication free of charge, with structured visits, study coordinators, and protocol-driven escalation.

Real-world cohorts tell a very different story. A 2025 analysis of US electronic-health-record data on 125,474 adults newly prescribed a dual-labeled GLP-1 receptor agonist (liraglutide, semaglutide, or tirzepatide) for overweight or obesity reported that, within twelve months, 46.5% of patients with type 2 diabetes and 64.8% of patients without type 2 diabetes had discontinued therapy.¹ Other US health-system datasets summarised in the Rodriguez analysis report one-year discontinuation rates ranging from 37% to 81% depending on insurance status, indication, and how discontinuation is defined.¹ The pattern is consistent: a steep drop in the first three months, a slower decline thereafter.

The interpretation matters. Trial completion rates describe what is biologically possible with a medication. Real-world persistence describes what actually happens in a life that includes a job, a family, a budget, side effects on a Tuesday morning, and a consultation eight weeks away.

3. Reason one: side effects without a system to manage them

The most consistent self-reported reason for stopping is gastrointestinal. Nausea, vomiting, persistent constipation, reflux, sulfur burps. The clinical reality of these symptoms on GLP-1 therapy is well documented; a 2023 retrospective cohort study using a US claims database found significantly elevated rates of pancreatitis, gastroparesis and bowel obstruction with GLP-1 receptor agonists used for weight loss compared with bupropion-naltrexone (hazard ratios 3.7 to 9.1).⁴ Most events are mild to moderate and time-limited, but they do not feel that way at 7 a.m. on a working day, four days into a new dose step.

The Rodriguez analysis confirmed the link directly: experiencing a moderate or severe gastrointestinal adverse event raised the hazard of discontinuation by 19% to 38% in the year following the first prescription.¹ What changes the trajectory is rarely the molecule itself. It is whether the person knows that the third week after a dose escalation is normally the worst, that hydration and protein-forward small meals reduce nausea, that constipation needs daily attention, and that holding a dose for an extra month is a legitimate clinical option. Patients who lack this granular information often interpret each symptom as a sign that "the medication is not for me", when in fact it is a sign that the medication is doing exactly what it was designed to do, in a body still adjusting. Our companion piece, GLP-1 side effects, month by month, walks through the realistic timeline.

4. Reason two: cost, supply, and the access gap

The second reason most often given is financial. List prices for branded GLP-1 therapies remain high in most markets. Reimbursement is fragmented. Even where the medication is partially covered, the patient share, plus accessories and consultation fees, can reach a level that is not sustainable for a chronic indication. The Rodriguez cohort showed that, among adults with type 2 diabetes, those earning more than $80,000 a year were 28% less likely to discontinue (HR 0.72, 95% CI 0.69-0.76) than those earning under $30,000.¹ The same income gradient was not statistically significant in adults without diabetes, where coverage gaps and out-of-pocket cost may already filter the population that initiates therapy.

Supply matters as well. The shortages of 2023 and 2024 forced periodic pauses on a meaningful share of patients, not because they decided to stop, but because the pen was not available at the pharmacy on the day they needed it. Each forced pause is also a moment when nausea reasserts itself if the dose is later restarted at the same level, which in turn becomes a self-reported "side effect" in the next survey. Cost, supply, and tolerance are not separate problems; they are linked.

5. Reason three: weight stigma and social pressure

This is the reason most under-reported in studies and most central in patient interviews. Many people on AOMs report being asked, sometimes weekly, whether they are "really" doing it through the medication, whether they will "stop now that they have lost the weight", whether the result is "earned". Weight stigma does not disappear when the scale moves. In some social settings it intensifies, because the visible change makes the chronic disease visible. The mental load this creates is real, and it correlates with discontinuation. We discuss it in detail in the mental load of being on an AOM.

The same Rodriguez analysis offered another quiet signal: among the 41,792 patients who discontinued and had a documented discontinuation-weight measurement, 36.3% (no diabetes) to 47.3% (with diabetes) reinitiated GLP-1 therapy within twelve months of discontinuation, often after a measurable regain of weight.¹ Many of those re-initiations are people who paused under social pressure, watched the chronic disease return, and re-engaged with treatment. The trajectory is rarely linear.

6. Reason four: the silence between two consultations

Obesity care happens in slots of fifteen to twenty minutes, every two to three months. Between those slots, the patient is alone with a chronic disease, an injection routine, a body that changes faster than identity, a family that comments, and a workplace that asks. The questions that matter most in obesity care almost never line up with the calendar of the consultation. They arrive on a Saturday morning, after a difficult dinner, before a holiday, the day a pen was lost. None of these moments are emergencies, but each of them, untreated, contributes to the gradual erosion of persistence.

This is the gap into which most discontinuation falls. The published behavioural literature makes this point repeatedly. In the SCALE IBT randomised trial, adults receiving liraglutide 3.0 mg combined with intensive behavioural therapy in primary care lost 7.5% of body weight at 56 weeks, compared with 4.0% on placebo plus the same behavioural support.⁵ The signal generalises: structured, evidence-based behavioural support combined with anti-obesity medication consistently outperforms medication in isolation on persistence and on outcomes.

7. What "support that actually moves persistence" looks like

The literature on behavioural interventions in obesity is precise about what works. The components with the strongest evidence are not motivational pep talks. They are specific:

• Self-monitoring of intake, weight, side effects and dose, at a frequency the patient can sustain (not daily for everything, not nothing).
• Structured problem-solving applied to the specific situations the patient names: the airport day, the family dinner, the post-injection 24 hours, the holiday week.
• Protein-forward eating with adequate fibre and hydration, framed around satiety and rhythm rather than restriction. We unpack this in eating on GLP-1 without counting.
• Regular resistance training to protect lean mass during weight loss, since approximately 25 to 40% of weight lost on GLP-1 therapy is lean tissue (about 25% in the SURMOUNT-1 body-composition substudy on tirzepatide; closer to 35-40% in the STEP-1 DXA substudy on semaglutide).
• Early triage of side effects, with clear criteria for what can wait until the next consultation, what warrants a phone call, and what is a red flag.
• Realistic framing of plateaus, so a flat scale at month four is not interpreted as treatment failure. See plateau and rebound on AOM.

None of this requires a daily clinician. It requires a system that lives in the patient's pocket, captures the right information at the right moment, and presents it back to the prescribing physician at the next consultation in a form that takes thirty seconds to read.

8. What changes the trajectory

Boli Care is built around a single observation: the medication works, and the system around the medication does not. We do not replace the prescribing physician. We carry the structured information that physicians do not have time to gather: graded side effects, dose history, eating rhythm, activity, mood, social load, between consultations, every day, in three languages. The Boli care team includes patient experts, moderators, and medical advisors, and provides daily support around the clinical reality of treatment.

Patients gain a consistent rhythm and a place to put their questions when the consultation is still six weeks away. Physicians gain a structured signal at the next visit. The trajectory of persistence improves not because we add motivation, but because we close the information gap that produces most of the silent dropouts.

References

1. Rodriguez PJ, Goodwin Cartwright BM, Gratzl S, et al. Discontinuation and Reinitiation of Dual-Labeled GLP-1 Receptor Agonists Among US Adults With Overweight or Obesity. JAMA Network Open 2025;8(1):e2457349. doi:10.1001/jamanetworkopen.2024.57349. PMID 39888616.
2. Wilding JPH, Batterham RL, Calanna S, et al. Once-Weekly Semaglutide in Adults with Overweight or Obesity. New England Journal of Medicine 2021;384(11):989-1002. doi:10.1056/NEJMoa2032183. PMID 33567185.
3. Wilding JPH, Batterham RL, Davies M, 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. doi:10.1111/dom.14725. PMID 35441470.
4. Sodhi M, Rezaeianzadeh R, Kezouh A, Etminan M. Risk of Gastrointestinal Adverse Events Associated With Glucagon-Like Peptide-1 Receptor Agonists for Weight Loss. JAMA 2023;330(18):1795-1797. doi:10.1001/jama.2023.19574. PMID 37796527.
5. Wadden TA, Tronieri JS, Sugimoto D, et al. Liraglutide 3.0 mg and Intensive Behavioral Therapy (IBT) for Obesity in Primary Care: The SCALE IBT Randomized Controlled Trial. Obesity 2020;28(3):529-536. doi:10.1002/oby.22726. PMID 32090517.