Set Point Theory and GLP-1s: Why Your Body Defends Your Weight

Set point theory proposes that your body has a preferred weight range it actively defends, the way a thermostat defends a room temperature. Push below it and the system fights to bring you back. The idea emerged from animal studies in the 1950s and has been refined considerably since.

The mechanism sits in the hypothalamus, a region of the brain that monitors signals about energy stores and adjusts hunger and energy expenditure accordingly. It reads leptin, a hormone released by fat cells that broadly signals sufficiency, and ghrelin, released by the stomach, which signals hunger. When fat mass falls, leptin falls with it, and the hypothalamus interprets this as a threat.

What it does next is the important part. It does not simply make you hungrier, though it does that. It also reduces spontaneous movement, lowers thyroid hormone output slightly, increases muscle efficiency so you burn fewer calories doing the same work, and heightens the reward value of food. The effect is coordinated and largely unconscious.

Many researchers now prefer settling point over set point, because the defended weight is not a fixed genetic number. It drifts upward over a lifetime in response to environment, sleep, stress, medication, and, importantly for women in midlife, hormonal change. That distinction matters practically: a defended weight that can drift up can, with sustained intervention, also settle down.

The evidence is unusually solid for a concept that sounds like folk wisdom. Three studies carry most of the weight.

Leibel, Rosenbaum and Hirsch (NEJM, 1995) put participants in a metabolic ward and measured energy expenditure precisely. After a 10% weight reduction, total energy expenditure fell about 15% below what their new, smaller body size predicted. Participants who gained 10% showed the mirror image, burning more than predicted. The body resisted movement in either direction.

Sumithran and colleagues (NEJM, 2011) followed 50 people through a 10-week very-low-calorie diet and then for a year. At 12 months, ghrelin was still elevated above baseline, leptin still suppressed, and subjective hunger still higher than before the diet began. The hormonal defence did not fade with time, which overturned the assumption that the body eventually accepts a lower weight.

Fothergill and colleagues (Obesity, 2016) tracked contestants from *The Biggest Loser* six years after the competition. Most had regained substantially, yet their resting metabolic rates remained roughly 500 calories per day below predicted. The adaptation had persisted even as weight returned.

Taken together, these findings reframe weight regain. It is not a character failure. It is a predictable physiological response that operates through hunger, satiety, and energy expenditure simultaneously, and it is measurable in a laboratory.

If you are watching the scale refuse to move, [this guide to why plateaus happen](/blog/glp1-weight-loss-plateau-why-it-stalls-and-what-to-do) covers the practical side.

GLP-1 medications work primarily on the hunger arm of the defence, which is why they succeed where willpower alone tends not to. They mimic glucagon-like peptide-1, a gut hormone released after eating that signals satiety to the same hypothalamic circuits that read leptin and ghrelin.

The result is that the brain's hunger signal is overridden rather than resisted. Women describe this as the quieting of food noise, the constant background negotiation about eating. That is not a metaphor for better discipline. It is a pharmacological effect on the circuit that produces the thoughts.

What GLP-1s do not do is reverse the metabolic slowdown. Energy expenditure still falls as you lose weight. Some of that is simply because a smaller body costs less to run, and some is the adaptive component described above. Research on whether GLP-1s blunt the adaptive portion is ongoing and not yet settled.

This is the cleanest way to understand the medication: it does not move your set point so much as it holds the defence at bay while you live at a lower weight. Some researchers argue that maintaining a lower weight for long enough may allow the defended range itself to drift downward, but this remains a hypothesis rather than an established finding.

It also explains a common frustration. Women on a GLP-1 who eat well and train hard still find the scale settling somewhere above their imagined goal. The medication changed the negotiation. It did not remove the other party.

The plateau arrives when the medication's appetite suppression and your body's defence reach a standoff, and the clinical trial data show it clearly.

In STEP 1 (NEJM 2021), the average weight curve on semaglutide 2.4 mg declined steeply for the first 40 weeks, then flattened noticeably by week 60, ending at 14.9% mean loss at week 68. In SURMOUNT-1 (NEJM 2022), tirzepatide participants were still losing weight at week 72, finishing at 20.9% on the 15 mg dose, with the curve beginning to flatten near the end.

Three things converge to create the plateau:

Your calorie needs have fallen. A woman who weighed 200 pounds and now weighs 160 needs meaningfully fewer calories daily. The intake that produced a deficit at the start now produces balance.

Adaptive thermogenesis has kicked in. Beyond the size effect, your body is running more efficiently than predicted.

The dose has stopped escalating. Appetite suppression is dose-dependent, and once you reach a maintenance dose, the pressure on the other side of the equation stops increasing while your body's counter-pressure continues.

A plateau after significant loss is a sign the system is working as designed, not that something has broken. The relevant question is whether you are at a weight that improves your health and that you can sustain. Body composition often keeps improving after the scale stops, which [this piece on body recomposition](/blog/body-recomposition-on-glp1-when-the-scale-stalls) explores.

When you stop, the defence you were holding at bay reasserts itself, and the trial evidence on this is consistent and unambiguous.

In STEP 4 (JAMA, 2021), participants took semaglutide for 20 weeks, then either continued or switched to placebo. Over the following 48 weeks, those who continued lost a further 7.9%. Those who switched to placebo regained 6.9%. In the STEP 1 off-treatment extension, participants had regained around two-thirds of their lost weight one year after withdrawal. SURMOUNT-4 (JAMA, 2024) found the same pattern with tirzepatide: continuing produced a further 5.5% loss, while withdrawal produced about 14% regain.

This is not a flaw in the drugs. It is confirmation of the physiology. The elevated ghrelin and suppressed leptin never went away, they were simply outvoted. Remove the medication and the original signal returns to full volume, meeting a body that now also has a slower metabolism.

The implication is that obesity behaves like a chronic condition rather than an acute one. Nobody expects blood pressure to stay low after stopping an antihypertensive, and the same logic applies here. If stopping is your plan, tapering deliberately and having a maintenance strategy in place matters enormously, which [this guide on tapering and regain](/blog/stopping-a-glp1-tapering-and-weight-regain-what-to-expect) covers in detail.

Menopause appears to shift where your body settles, and it does so through several routes at once.

The SWAN study found that women gain weight steadily across midlife and, more tellingly, that fat redistributes to the abdomen during the transition regardless of whether total weight changes. Visceral fat increases while lean mass declines. That shift alone changes metabolic rate, because muscle is more metabolically expensive than fat.

Estrogen also acts directly on hypothalamic circuits that regulate appetite and energy expenditure, and it influences insulin sensitivity. As estrogen becomes erratic in perimenopause and then falls, those regulatory signals become less reliable.

Sleep is the underrated factor. Night sweats and insomnia fragment sleep for a large proportion of women in the transition, and short sleep is independently associated with higher ghrelin, lower leptin, and increased appetite the following day. A defended weight that is being renegotiated nightly by poor sleep tends to drift up.

What this means practically is that a woman starting a GLP-1 at 52 is working against a moving baseline in a way a woman of 32 is not. Expecting the same trajectory sets you up for unnecessary discouragement. [This piece on menopause metabolism](/blog/glp1-menopause-metabolism-why-weight-loss-gets-harder) goes deeper on the mechanisms.

A true equilibrium and a fixable stall look identical on the scale, so you have to look at the inputs instead. Work through these in order.

How long has it actually been? Weight fluctuates by one to two kilograms across a week from fluid, sodium, glycogen, and, in perimenopause, cycle position. Anything under four weeks is usually noise rather than a plateau. Compare weekly averages, not individual mornings.

Has your waist changed? If the scale is flat but your waist measurement is shrinking and clothes fit differently, you are losing fat and gaining or preserving muscle. That is progress the scale cannot see, and it is common in women who start resistance training at the same time.

Are you actually eating enough protein? Appetite suppression makes protein the first casualty, because meat and other dense protein sources become unappealing. Under-eating protein accelerates muscle loss, which deepens the metabolic slowdown and makes the stall self-reinforcing.

Has your food intake quietly crept up? Not through willpower failure, but because appetite suppression fades between doses, or a dose has become less effective over months. Three or four days of honest logging usually answers this.

Is your sleep broken? In midlife this is the most commonly missed input. Fragmented sleep raises next-day hunger measurably.

Has anything else changed? New medications, particularly some antidepressants, beta-blockers, and steroids, can slow loss. So can starting HRT, which sometimes shifts fluid balance in the first weeks without changing fat mass.

If protein, sleep, training, and intake all check out and the waist has not moved in eight weeks, you are most likely at a genuine new equilibrium for your current dose. That is the point to have a dose conversation rather than a discipline conversation.

You cannot argue with the hypothalamus, but you can change several of the inputs it reads. These are the interventions with the best supporting evidence.

Protect muscle mass. This is the single highest-leverage action. Muscle is the metabolically expensive tissue you are at risk of losing, and losing it deepens the metabolic slowdown. That means resistance training two to three times a week and adequate protein, generally 1.2 to 1.6 grams per kilogram of body weight for women losing weight. [How much protein you actually need on a GLP-1](/blog/protein-on-glp1-how-much-you-really-need) has the specifics.

Do not chase the deficit deeper. Cutting calories further when you plateau typically increases adaptation and muscle loss rather than restarting progress. Eating too little is one of the more common self-inflicted problems on a GLP-1.

Prioritise sleep. Seven to nine hours meaningfully affects the hormones under discussion. In menopause, that may mean treating night sweats before blaming the plateau on diet.

Reconsider the goal. A 10 to 15% loss delivers most of the cardiometabolic benefit. Chasing a number from your twenties against active physiological resistance is a poor trade.

Talk about dose. Sometimes a plateau reflects a dose that has not been optimised rather than a true set point standoff. [This guide to dose escalation](/blog/glp1-dose-escalation-when-to-increase-and-when-to-hold) covers when increasing is reasonable.