GLP-1 maintenance struggles: what the science says about keeping weight off

What did @caleysvensson actually say?

After nearly two years of stable weight maintenance on a GLP-1 medication, she had surgery in October, went off the drug, and has since gained about 11 pounds despite titrating back on. Her take: going off the medication "knocked" her out of maintenance, and the fix is bumping her dose from 10 mg to 12.5 mg. She credits a theory she picked up on TikTok, but says her doctor validated it.

To her credit, she frames this as personal experience, not medical advice. She is transparent about the uncertainty: "I don't know if it's going to work, but I will keep you posted." That kind of epistemic humility is rarer than it should be in the GLP-1 content space, and it matters.

Does the science back this up?

Mostly yes, though the mechanism is more complicated than "knocked out of maintenance" implies. The evidence is fairly consistent that GLP-1 receptor agonists do not produce durable weight loss once stopped, and resuming them does not always restore prior efficacy immediately.

The STEP 4 trial (Rubino et al., 2021, JAMA) is the clearest data point here. Participants who stopped semaglutide after 20 weeks regained roughly two-thirds of their lost weight within a year, while those who continued lost more. The drug is doing active suppressive work on appetite and energy regulation continuously. When you remove it, those signals return, sometimes faster than expected.

Less studied is what happens when you stop mid-maintenance, not mid-loss. There is limited controlled trial data on re-titration after a medically necessary drug interruption. Some clinical guidance suggests patients may need to restart from a lower dose and titrate back up, which can mean the "maintenance dose" is no longer sufficient when re-introduced. Her experience tracks with that clinical pattern, even if the TikTok framing of it is imprecise.

What did they get wrong (or right)?

The phrase "knocked myself out of maintenance" deserves scrutiny. It implies a discrete, resettable state, which is not quite how the biology works. GLP-1 drugs appear to work continuously on appetite-regulating pathways in the brain and gut. There is no fixed "maintenance mode" that gets toggled off. What is more likely is that her physiology resumed its baseline hunger signaling during the drug-free period, and the re-titration did not fully re-establish the suppression her body had adapted to at 10 mg.

She also attributes this theory to TikTok, which should give any clinician pause. The idea itself is not wrong, but it is oversimplified. The more accurate framing, supported by Wilding et al. (2022, Diabetes, Obesity and Metabolism), is that GLP-1-driven weight regulation requires continuous drug exposure. A gap does not permanently reset the body, but it can require a higher dose to re-establish prior suppression, particularly if weight regain has occurred.

What she got right: she talked to her doctor before changing her dose. That is the correct sequence of events, and she said so explicitly.

What should you actually know?

If you are on a GLP-1 medication and facing a planned or unplanned interruption, such as surgery, there are a few things the data support.

• Weight regain after stopping is well-documented and not a personal failure. The STEP 1 extension data (Wilding et al., 2022, Diabetes, Obesity and Metabolism) showed most weight loss reversed within a year of stopping semaglutide.
• Re-titration after a gap may not return you to prior efficacy at the same dose. Some clinicians restart from a lower dose and titrate up, which can temporarily feel like the drug is not working.
• There is no clinical consensus on what to do when a maintenance dose stops working after re-introduction. This is genuinely under-studied territory.
• Dose increases should be driven by clinical assessment, not TikTok pattern-matching, even when the underlying logic is directionally correct.

Her situation is not unusual. But the solution she arrived at, increasing her dose with physician input, is the appropriate clinical response, not something to replicate without a similar conversation with your own prescriber.