The most under-discussed phase of any GLP-1 protocol isn't the dose-escalation, the side effects, or the early weight loss. It's what happens after you hit your goal.

The STEP-4 trial — published in JAMA in 2021 — is the cleanest published data we have on what happens when patients stop semaglutide. The headline: patients who stopped at week 20 regained, on average, two-thirds of the weight they'd lost within the following 48 weeks. Patients who continued kept losing.

Tirzepatide data from the SURMOUNT extension studies tells a similar story. Discontinuation without a plan is essentially a slow-motion rebound.

But "stay on full dose forever" isn't the only alternative. The maintenance protocol that holds results is more nuanced than either extreme — and here's how it actually works.

Why the rebound happens (biologically).

Three things change when you lose meaningful weight, GLP-1 or not:

  • Resting metabolic rate drops. A 200 lb body uses more calories at rest than a 165 lb body. After meaningful weight loss, you need 200–400 fewer calories per day to maintain.
  • Hunger hormones rebound. Leptin (satiety) drops; ghrelin (hunger) climbs. This is documented to persist for years after weight loss, sometimes permanently — a phenomenon called "metabolic adaptation."
  • Food noise returns. GLP-1s dramatically dial down the constant background thinking about food that most overweight patients have lived with for years. When you stop the medication, that mental chatter often comes back.

All three forces push toward rebound. The maintenance protocol is about neutralizing each of them deliberately rather than expecting willpower alone to do the job.

The maintenance dose framework.

Most patients don't need to stay on their full weight-loss dose to maintain. The published literature and our clinical experience converge on a few common patterns:

Tier 1 — Reduced GLP-1 dose, continued indefinitely: the most reliable maintenance approach. Reduce to roughly half your weight-loss dose (e.g., 1.0 mg/week semaglutide instead of 2.4 mg, or 5–7.5 mg tirzepatide instead of 15 mg). For patients with severe metabolic disease or who lost 60+ lb, this is the default.

Tier 2 — Maintenance dose for 6–12 months, then taper: for patients who lost 25–50 lb. The brain and body need time to recalibrate at the new weight. Holding a reduced GLP-1 dose for 6–12 months while building habits often allows successful tapering after that.

Tier 3 — Slow taper to discontinuation: for patients who lost <25 lb, who built strong nutrition and exercise habits during the active loss phase, and who have minimal metabolic disease. A multi-month taper (rather than a cold stop) gives the body time to adjust.

The honest framing
GLP-1s are not a 6-month course. For most patients, weight maintenance after meaningful loss involves some ongoing medication, at some reduced dose, for some extended period. The protocol that fails is the one that didn't plan for the back end.

The four habits that have to be in place before you taper.

Independent of which tier you're in, four things have to be solid before you reduce dose meaningfully:

1. Protein floor of 1 g per pound of target body weight, daily. Covered in body recomp on GLP-1s. This is the biggest single variable preventing muscle-mass loss during the weight-loss phase and preventing rebound after.

2. Resistance training 3x/week. Walking + zone 2 cardio is great for cardiovascular health and largely useless for maintaining muscle. Resistance training is the keep-this-muscle signal. The 3x-per-week resistance template is the minimum.

3. Steady step count at 7,000+ daily. Walking is the most underrated maintenance lever (covered in the 7,000-step rule). Patients who keep daily movement up after stopping a GLP-1 rebound dramatically less than patients who don't.

4. A specific eating pattern, not a vague "healthier diet." What you actually eat day-to-day matters more than the abstract "I'll eat better." A protein-and-fiber-first framework — like the 4 rules we use — that you can articulate in one sentence is more durable than a complex meal plan.

If any of those four isn't in place, holding the maintenance dose longer (Tier 1 or Tier 2) is the right call. If all four are solid, a slow taper (Tier 3) becomes a real option.

What to monitor during maintenance.

Quarterly check-ins are the right cadence:

  • Body weight — but more importantly, body composition. Scale weight alone misses the picture. A DEXA or InBody scan every 6 months tells you whether you're holding muscle.
  • Fasting insulin and HbA1c — if these creep up, you're regressing metabolically even if the scale hasn't moved.
  • Lipid panel — should stay improved from your pre-treatment baseline.
  • Liver enzymes (ALT, AST) — should also remain improved, or at minimum stable.
  • Hunger and food-noise self-report — if patients describe returning food noise as "intrusive," dose probably needs to come up, not down.

What rebound looks like — and what to do.

Rebound isn't usually sudden. It's 1–2 lb per month for a year, until you've quietly regained 20–30 lb. By the time the scale shows the problem, you've often been off-protocol for 4–6 months.

The intervention isn't shame, it's escalation:

  • Re-titrate dose up. The same dose that worked before usually works again.
  • Reinforce the four habits. Often one or two have quietly degraded.
  • Don't restart from scratch. Patients who restart at maintenance dose typically need a shorter ramp than first-time starters.

What about "GLP-1 vacations"?

Some patients ask about taking 1–3 month breaks. The clinical reality:

  • A 4-week break usually doesn't produce meaningful regain if habits are solid.
  • A 12-week break typically produces 5–15 lb of regain in most patients.
  • A 24+ week break tends to produce two-thirds rebound regardless of habits.

There's no clinical benefit to vacations from these drugs — no rest needed for safety, no tolerance to break. The decision to pause should be deliberate (cost, side effects, life circumstances), not routine.

The takeaway

GLP-1 therapy is closer to hormone replacement therapy than it is to a course of antibiotics. The maintenance phase is the protocol — not a separate decision after the protocol is over.

Most patients who lose meaningful weight on a GLP-1 will benefit from staying on some reduced dose for an extended period, paired with the habits that hold the result independently. The cleanest framework is to plan that maintenance phase before you start, not after you stop.

Quit cold turkey at goal weight and you're betting against published data. Build the maintenance protocol into the plan from the start and the rebound math reverses.

Sources: STEP-4 trial in JAMA, 2021; SURMOUNT extension data on tirzepatide discontinuation; International Society of Sports Nutrition position stand on protein.

Plan the back end

Compounded GLP-1, with clinician oversight.

DirectCare AI builds the maintenance phase into your GLP-1 protocol from the start — taper, dose-reduction, and the nutrition and training plan that holds the result.

See if you qualify →

Editorial disclosure: This article is for informational purposes only and does not constitute medical advice. All treatments at DirectCare AI are prescribed by US-licensed clinicians based on individual medical evaluation. Compounded medications are not FDA-approved and are not reviewed by the FDA for safety, effectiveness, or quality. Always consult a US-licensed clinician before starting or changing any therapy.