The Lowest Effective Dose Strategy for GLP-1 Weight Loss

How to Dose GLP-1s Without Painting Yourself Into a Corner

Most people start a GLP-1 the way the box tells them to, which is a fixed titration schedule that moves the dose up every four weeks whether or not the current dose is still working, and by the time they've been on it six months they're sitting at a high dose with nowhere left to go and a bill that reflects it.

The drug is not what's causing this, since the trouble comes from spending your dose range before you actually need it.

Think about the whole arc of a fat loss phase for a second, because the dose is only one piece of it. You start eating less, your body responds by making you hungrier and lowering the amount of energy you burn at rest, and the drug is there to blunt the hunger side of that response so you can keep the deficit going long enough to matter. That means the drug's job changes over time. Early on you barely need it because the deficit is fresh and adherence is easy. Later you need more of it because the body has adapted.

So if you burn through your dose range in month one, you have nothing left for month six, which is exactly when you need it.

So nowadays, the way that I suggest that people do it is they say, hey, start with a half a milligram, see how it makes you feel. If you're losing weight and seeing progress, fucking sweet. You just saved yourself a bunch of money, and you still have the entire dose range sitting there unused for later.

And then from there, you just stay on the lowest dose possible until you're no longer getting the result that you were seeking.

The literature hasn't actually tested this approach. The trials titrate on a calendar, not on how the patient feels, so I have never seen a study that compared a fixed schedule against a hold-until-it-stops-working approach. This is what I do with clients and what I've watched play out, not what a published protocol says.

What the trials do tell you is that lower doses are not placebo. In SUSTAIN 4, semaglutide at 0.5 mg weekly dropped A1c by 1.21 percent and body weight by 3.47 kg over 30 weeks, and while the 1.0 mg group did better, the low dose was clearly doing work on its own (Aroda et al., 2017), and that gap between the two doses is what tells you a half milligram already counts as a genuine, working dose rather than a token amount.

Maybe your appetite suppression will start to creep back in, say the food noise, maybe you reach a weight loss stall or whatever the case may be. That's your signal to move, and not before.

And since you started at such a low dose, you can increase by a half a milligram or a milligram per week. A jump that small still registers, because when you've been sitting at 0.5 mg, going to 1.0 mg is a doubling of the dose your receptors have been seeing, and the body responds to relative change more than absolute numbers.

And you're not going to end up having to scale up to six, eight, 10, 12 milligrams. Then when it comes time for you to come off, just taper off the same way that you tapered on.

Coming down slowly matters for the same reason going up slowly does. Your appetite has been suppressed by something external, and if you remove it in one step, the hunger arrives in one step too, usually into a body that has spent months getting more efficient at storing what you feed it. Walking the dose back down over weeks gives you time to see how much of your eating behavior was you and how much was the drug.

Research: Aroda VR, Bain SC, Cariou B et al. Efficacy and safety of once-weekly semaglutide versus once-daily insulin glargine as add-on to metformin in insulin-naive patients with type 2 diabetes (SUSTAIN 4), published in Lancet Diabetes Endocrinol back in 2017.

References:

Aroda VR, Bain SC, Cariou B et al.. Efficacy and safety of once-weekly semaglutide versus once-daily insulin glargine as add-on to metformin (with or without sulfonylureas) in insulin-naive patients with type 2 diabetes (SUSTAIN 4): a randomised, open-label, parallel-group, multicentre, multinational, phase 3a trial. Lancet Diabetes Endocrinol. 2017. https://pubmed.ncbi.nlm.nih.gov/28344112/

Liu M, Wang X, Li M et al.. Safety and efficacy of GZR18, a long-acting GLP-1 analog, in Chinese patients with type 2 diabetes: A randomized, double-blind, phase 1b/2a trial. Cell Rep Med. 2026. https://pubmed.ncbi.nlm.nih.gov/42320483/