The evidence
GLP-1s work. Staying on them is the hard part.
The medication is the easy bit. Keeping the weight off is where it falls down, because maintenance is about routine and behaviour, not appetite alone. That's the gap Healthcount is built for.
The maintenance gap, in the data
This isn't a willpower story. The biggest, most recent real-world studies all show the same thing.
64.8%
stop within a year
In a real-world study of 125,474 adults, almost two-thirds of people without type 2 diabetes had stopped their GLP-1 within 12 months.
36.3%
restart within a year
Of those who stopped, more than a third started again within the next year. The people who regained weight were the most likely to. That's the stop-start cycle.
~⅔
of lost weight regained
After stopping semaglutide, people in the STEP 1 trial extension put back about two-thirds of the weight they'd lost within a year. SURMOUNT-4 found the same pattern for tirzepatide.
Sources: real-world discontinuation and reinitiation study (125,474 adults, 2018 to 2023); STEP 1 trial extension (Wilding et al.); SURMOUNT-4 withdrawal (Aronne et al.). Full references below.
Why behaviour change is the real answer
GLP-1 medicines lower appetite, and that's what drives the early weight loss. When people stop, whether it's the cost, the side effects, or supply running out, appetite comes back and the body works hard to return to its old weight. The physiology of that is well documented. If the routines that keep weight steady were never built in the first place, the weight returns. That's the regain-and-restart cycle the data keeps showing.
The missing piece isn't another drug. It's low-burden support that spots drift early, before a small wobble turns into a full relapse, and helps people get back to their routine (and their clinician when they need one). No daily calorie obsession. No shame. A quiet layer that only speaks up when it matters.
How we measure impact
We measure what funders and clinical teams can actually trust: the signals that show whether a programme is keeping people on track, not vanity numbers. Every metric below maps to the gap above.
How it runs: a baseline period, then the intervention, with read-outs at 8 to 12 weeks, 3 months and 6 months. Reporting is aggregated and privacy-safe. Partners see cohort trends (minimum 10 members), never one person's data.
References
- Discontinuation and reinitiation of GLP-1 receptor agonists among US adults with overweight or obesity (retrospective cohort, 125,474 adults, 2018 to 2023). NCBI
- Wilding JPH et al. Weight regain and cardiometabolic effects after withdrawal of semaglutide (STEP 1 trial extension), Diabetes, Obesity and Metabolism, 2022.
- Aronne LJ et al. Continued treatment with tirzepatide for maintenance of weight reduction (SURMOUNT-4), JAMA, 2024.
- Sumithran P, Proietto J. The defence of body weight: physiological responses to weight loss.
General information for evaluation, not medical advice. Healthcount supports self-management and signposting. It does not diagnose, prescribe, or replace clinical care.
See it work with your cohort
Run a structured pilot with staged, privacy-safe read-outs, and measure the stop-start gap for your own members.