Does GLP-1 microdosing work for weight maintenance?

What "GLP-1 microdosing" actually means

Search for GLP-1 microdosing and you will find a specific promise: keep taking semaglutide or tirzepatide at a fraction of the normal dose, spend less, feel fewer side effects, and hold onto the weight you already lost. It sounds like a reasonable middle ground between full-dose treatment and stopping cold. The problem starts with the word itself. Microdosing has no clinical definition in obesity medicine. Nobody at the FDA, Novo Nordisk, or Eli Lilly uses the term. It came out of social media and got picked up by compounding pharmacies marketing custom low-dose vials.

That matters because the doses being sold as "microdoses," often 0.05 to 0.25 mg of semaglutide a week, sit below even the lowest FDA-approved starting dose of 0.25 mg, let alone the 2.4 mg weekly maintenance dose Wegovy is actually approved and studied at. Nobody manufactures a pen that low. Every microdose product on the market is compounded to order, which is a different regulatory and safety category than anything that went through a Phase 3 trial.

The dose numbers, side by side

That last row is the whole story. Semaglutide and tirzepatide trials, STEP, SURMOUNT, SURPASS, all tested doses inside the approved range. Nothing in the published record establishes what happens, good or bad, at a fraction of that dose over months or years.

Does it actually work?

Dr. Amanda Velazquez, director of obesity medicine at Cedars-Sinai, has treated patients who try it. Her read on the outcomes is not encouraging: very low, infrequent dosing tends to produce modest results, in the range of 5 to 6% weight loss, compared with the 15 to 20% seen at standard therapeutic doses. Some of that gap is expected. GLP-1 drugs work in a roughly dose-dependent way, so a much smaller dose should logically do less. The part that should give anyone pause is the maintenance claim specifically. Nobody has published data showing a microdose holds weight that was lost at a full dose, which is the exact use case it is being marketed for.

Dr. Jody Dushay, an assistant professor of medicine at Harvard Medical School and attending endocrinologist at Beth Israel Deaconess, was more direct when STAT News asked her about it in May 2026: "Microdosing GLP-1s is not a thing. There are no legitimate long-term data to support it." She draws a sharp line between that and the legitimate dose adjustments specialists already make within the approved range, using a pen's built-in dose clicks or spacing standard doses slightly further apart under medical supervision. Both of those stay inside data that actually exists. Microdosing, as sold, does not.

The real risk sits with the source, not just the dose

Because no manufacturer sells a below-starting-dose pen, anyone microdosing is buying compounded semaglutide or tirzepatide, usually from a pharmacy or an online vendor rather than a hospital pharmacy with tight oversight. We cover the shrinking legal window for compounding in detail in our guide to compounded GLP-1s in 2026, but the short version matters here directly: a 503A pharmacy is only supposed to compound a custom dose for a documented individual medical need, not because a patient wants a cheaper or smaller version of an approved drug. The FDA has separately documented dosing errors tied to compounded GLP-1 products, including fainting episodes traced back to measurement mistakes with unmarked, multi-dose vials, exactly the format most microdose products ship in.

Put together, someone microdosing is often taking on two unknowns at once: a dose nobody has studied, from a supply chain with a documented error record, for a use case, long-term maintenance, that nobody has tested either.

What the evidence actually supports instead

There is a real, data-backed version of "take less to keep the weight off," and it is not the same thing as microdosing. It is stepping down to a lower dose that is still inside the FDA-approved range, for example moving from 15 mg to 5 mg of tirzepatide, rather than chasing a dose below the lowest tier a manufacturer makes. We go through the trial data behind this in our guide to preventing GLP-1 weight rebound, but the headline numbers are worth repeating here. The SURMOUNT-4 trial found people who continued tirzepatide at any studied dose held onto far more of their weight loss than those switched to placebo, where 82% regained at least a quarter of what they had lost. The STEP 1 extension found a similar pattern with semaglutide: people who stopped entirely regained about two-thirds of their lost weight within a year. Both trials point the same direction. Continuing a real, approved dose, even a lower one, beats stopping. Neither trial tested anything below the approved floor.

What to actually do if cost or side effects are the reason you want less drug

Cost and side effects are the two honest reasons most people look into microdosing, and both have better answers than an unmarked vial. If cost is the issue, Medicare now covers Wegovy, Zepbound, and Foundayo for $50 a month for eligible enrollees, and manufacturer savings programs and Zepbound's LillyDirect pricing have both come down; our cost and eligibility cheat sheet walks through current numbers. If side effects are the issue, a slower dose-escalation schedule within the approved range, which your prescriber can adjust directly, has actual trial-adjacent support behind it, unlike an off-label low dose from a compounder. Either path keeps you inside the range where the safety and efficacy data actually exists.

The bottom line

"Microdosing" sounds like a smart, minimal-effective-dose strategy, and the instinct behind it, use less once you have reached your goal, is not wrong. It is the specific doses being sold under that name, below anything the FDA approved or any trial tested, that the evidence does not support. If maintaining your results at a lower cost or lower side-effect burden is the goal, the better-supported move is a lower dose still inside the approved range, prescribed and monitored by the person who already knows your history, not a product with no trial and an unresolved safety record behind it.