Why up to 65% of GLP-1 patients quit within a year

Most clinics running GLP-1 programs already know they lose patients. Fewer know how fast, or how early. Real-world persistence data puts the curve at roughly 26% discontinuation by 3 months, 31% by 6 months, and 36.5% within a year across the general GLP-1 population. For patients being treated for weight loss alone, without a type 2 diabetes diagnosis, the one-year discontinuation rate climbs as high as 64.8%, according to claims data reviewed by the Blue Cross Blue Shield Health Index. Some analyses put it closer to two-thirds.

That gap between diabetic and non-diabetic patients is the part worth sitting with. The drug is the same. The dose titration is often similar. What’s different is the support structure around the patient — and increasingly, the research backs that up rather than pointing at the medication itself.

It’s rarely just the side effects

Cost and GI side effects are real factors: every extra percentage point of out-of-pocket cost per 30-day supply measurably increases the odds a patient discontinues, and adverse events raise the risk further. But a recent Cleveland Clinic review of real-world GLP-1 discontinuation found something clinics tend to underweight — patients who lose meaningful weight are significantly less likely to stop, and patients treated by providers with specific obesity-medicine expertise are more likely to complete even the first 12 weeks of treatment. Persistence isn’t just about tolerating the drug. It’s about whether the patient is getting enough signal, early enough, that the program is working and someone is paying attention.

That’s the part a prescription alone can’t deliver. A patient who feels nauseous in week 2, isn’t losing weight as fast as they hoped in week 6, or simply goes quiet after a follow-up no-show in week 10 is drifting long before they formally discontinue. Most clinics don’t find out until the next scheduled visit — often 4 to 12 weeks later, by which point the patient has frequently already stopped filling the prescription.

Restart rates say the door isn’t actually closed

Here’s the more useful finding buried in the same data: discontinuation isn’t always permanent. Among patients who stopped, 47.3% of those with type 2 diabetes and 36.3% of those without reinitiated therapy within a year. In other words, roughly a third to nearly half of “lost” patients come back on their own, eventually, usually after a gap long enough to lose momentum, regain weight, or lose confidence in the program.

That reinitiation rate is the real opportunity. If a third to a half of patients who drift are recoverable anyway, the entire economic question becomes how early you catch the drift, not whether you can eventually win the patient back once they’ve already gone quiet for two months.

What actually closes the gap

This is the specific problem behavioral, trigger-based outreach is built to solve, and it’s why Adherix built Drift Correction as a core part of the platform rather than a bolt-on feature. Instead of waiting for the next scheduled visit to find out a patient has gone silent, the engine detects the pattern — no reply in 48 hours, a missed check-in, a stalled weigh-in — and fires a targeted re-engagement message automatically, before the gap turns into a permanent dropout. No coordinator has to notice; no patient has to make it to a follow-up call to be caught.

For a clinic running a panel of a few hundred active GLP-1 patients, even a modest improvement in early-drift recovery is worth real money in protected program revenue. You can run your own numbers with the ROI calculator, or upload a de-identified patient list to the free cohort audit and see your actual 90-day retention curve and estimated revenue leakage in about a minute.

The fastest way to see how the detection and outreach actually works is to try it on your own phone — enroll yourself as a demo patient and watch the same sequence a real patient would get, drift trigger included.

Try the live demo →