A figure presented at ENDO 2026 in Chicago has stayed with me for weeks. In the largest claims analysis of its kind, researchers tracked more than 60,000 adults who started a GLP-1 for type 2 diabetes. By the two-year mark, most had stopped at some point. Here is the part our industry tends to skip over: among the patients who discontinued, 53.6% started again within a year, and 65.9% were back within two.
Read that slowly. The patient your dashboard quietly files under “discontinued” is, more often than not, already finding a way back to therapy. Usually on their own, without anyone who knows their history reaching out.
That single fact pulls a thread through everything our commercial playbook assumes.
We built the machine for two moments
Pharma commercial strategy is organized around winning the prescription and protecting the refill streak. Enormous budgets fund the start, and a second wave funds adherence while the patient stays on. The instant the streak breaks, attention moves elsewhere. The file closes, the reporting logs a discontinuation, and the system goes looking for the next new patient.
The ENDO data says that reflex is expensive. A lapse looks a lot like halftime. A large group of motivated, recoverable patients forms in exactly the moment we have trained ourselves to walk away.
The front door keeps leaking for years
There is a second population sitting alongside the comeback patients, and it is larger than most brands model. IQVIA’s U.S. Medicine Use Trends 2026 looked at 99 medicines launched between 2020 and 2024. In their first year, roughly 7 million new prescriptions were written for these products, and about 65% went unfilled. Payers rejected close to half outright. The rest cleared approval and were then abandoned by patients, most plausibly at the counter when the out-of-pocket price appeared.
And it persists. Four years after launch, more than half of these prescriptions are still going unfilled, well above the 29% rate for established medicines. A physician decided the patient needed the therapy. The patient agreed in the room. Then the months passed in silence and nothing happened.
Stack these two groups together, the patients who lapsed and the patients who never started, and you get a recoverable pool far bigger than the new-acquisition target most teams chase each quarter. Most of that pool is stuck on access and affordability, which means it responds to navigation and a trusted voice.
Recovery has a financial case
Coverage of AMCP 2026 added the piece access leaders need. A five-year actuarial model of one million lives found that semaglutide coverage ran ROI-positive for employers, PBMs, health plans, and government payers, with the savings concentrated in reduced comorbidity spending rather than the pharmacy line.
The system-level picture was more mixed, but the direction for sponsors is clear enough: keeping a patient on therapy, or bringing one back, lands on the favorable side of the ledger once you measure beyond the pharmacy budget. That is the language commercial and access leaders respond to, and it is the case a win-back program should be built to prove.
Engagement is the active ingredient
There is a comfortable objection waiting here: “We already text our patients.” A large randomized trial published in JAMA tested refill-reminder texts against usual care and found no difference in twelve-month adherence. An early lift of a few points faded by the end of the year.
The secondary analysis is where it gets interesting. Patients who replied to the texts, who treated the thread as a conversation, held meaningfully higher adherence at twelve months. The predictive signal was the act of engaging, the back-and-forth with a sender worth answering. A one-way blast slides into background noise the brain filters out, while a responsive exchange with someone the patient trusts behaves more like a relationship and tends to hold.
Designing for the return
So far the comeback has been happening by accident. Patients restart on their own timeline. The never-started drift back whenever something finally clears their path. Meanwhile, the patients who reply to a generic text quietly outperform the ones who went silent. The opportunity in front of patient-access and DTP teams is to stop leaving all of that to chance.
At Hoot, the moment a patient lapses is the moment a familiar voice should appear: their own physician, on video, arriving by text on the phone already in their hand. The 60-day refill gap is the highest-yield trigger we know of, because it lines up precisely with the window when the reinitiation curve is steepest. The same channel reaches the never-started patient at first fill, pairing short physician-led education with the practical next step on copay and prior authorization, since most of the leak comes down to cost and navigation hurdles.
None of this asks the patient to download an app or learn a new behavior. It uses the device already in their pocket and a relationship they already have with the person who prescribed.
The brands that win the next few years will be the ones that treat a lapse as the start of a conversation worth having. If your team measures the start and the streak but has no number for the comeback, that is the first gap worth closing. The patients are already trying to return. The question worth sitting with is whether anyone is there when they do.