The defended-setpoint model (why diet-induced loss is resisted and regain is the norm) is held on Weight-Loss Maintenance and Metabolic Adaptation and its intervention payoff on The Disease Model of Obesity. This page answers the next question: what actually counters the defense, and does holding weight down long enough wear it away? The organising distinction the evidence forces is that four different things get called mitigation, and they are not the same act:
- Reset — the defended level itself falls, so less counter-force is needed over time.
- Override — the defense is unchanged but overpowered from outside for as long as the agent is applied.
- Durably lower — a one-time intervention produces a large loss that holds without a standing external agent; whether it truly resets the defended level, or merely holds it down by other means, is inferred, not proven.
- Reduce the provocation — blunt a driver of the defense, or avoid engaging it at all.
The held evidence supports override and (for one intervention) a durable loss that holds without a standing prop; it gives no support for reset by attrition — the defended level weakening merely because reduced weight is maintained over time.
Does holding weight low weaken the setpoint over time? [EXTRACTED]
No held human evidence that maintenance duration weakens the defense — and the nearest measurement runs the other way. Sumithran 2011 tracked the appetite-hormone profile to one year into maintenance and found the shift still present: «One year after initial weight reduction, levels of the circulating mediators of appetite that encourage weight regain after diet-induced weight loss do not revert to the levels recorded before weight loss…» (Sumithran et al., 2011). Twelve months in — twice a six-month window — hunger is still up and satiety still down. So six months of maintenance should be expected to leave the defense intact, not worn down.
Four bounds make the reset question genuinely open rather than answered no — they are the reason this page will not assert a permanent defense either (see Weight-Loss Maintenance and Metabolic Adaptation for the full limits):
- Persistence is measured only to 12 months. Permanent defense is an extrapolation beyond the data; the candidate long-term dataset (Biggest Loser 6-year follow-up) was assessed and cut below the quality bar, so the fabric holds nothing on multi-year persistence.
- The still-lighter-body confound. At 12 months participants were still ~8% below baseline, so every hormone difference is confounded with genuinely being a smaller person. The competing reading — that these are simply the hormone levels appropriate to a lighter body, not an adaptation over and above it — is excluded for leptin only: «Reductions in leptin lev-els from baseline at weeks 10 and 62 remained sig-nificant when adjusted for fat mass…» (and leptin was still ~35% below baseline at a year) (Sumithran et al., 2011). For the other signals, an adaptation to weaken is assumed, not established — you cannot wear away a defense whose existence-beyond-current-size is itself unproven.
- VLCD-specific. The exposure was aggressive (~500 kcal/day liquid formula). Whether a slower, food-based deficit provokes the same persistence — or a weaker, more resettable one — is not shown.
- The drug mirror shows override, not reset. Stopping semaglutide after a year of suppression releases the weight back up (below), which is what an un-reset, merely-overridden setpoint does.
Bottom line on the direct question: holding weight low for six months has no held evidence of weakening the setpoint, and the mechanism by which time-at-reduced-weight would reset it is not established in anything the fabric holds.
The mitigation taxonomy — sorted by what each does to the setpoint
Override it — GLP-1/GIP drugs (best-evidenced; not a reset) [EXTRACTED]
The drugs do not reset the setpoint; they resupply the depleted satiety signal the defense runs on, overpowering it while taken. The tell that it is an override is regain on withdrawal: stopping semaglutide after 68 weeks returned most of the lost weight — «Following treatment withdrawal, semaglutide and placebo participants regained 11.6 (SD: 7.7) and 1.9 (SD: 4.8) percentage points of lost weight, respectively, by week 120…» (Wilding et al., 2022).
Uniquely among the counters, this route has hard-outcome evidence: SELECT cut major cardiovascular events — «hazard ratio, 0.80; 95% confidence interval, 0.72 to 0.90; P<0.001» (Lincoff et al., 2023) — but only in established cardiovascular disease (secondary prevention), with no hard-outcome evidence in primary prevention -> Semaglutide for Cardiovascular Risk in Obesity. Decision consequence: this is a continued medication (a blood-pressure-style commitment), not a course that fixes the setpoint and is then stopped.
Durably lower it — bariatric surgery (the one held durable large loss) [EXTRACTED]
Surgery is the only held intervention that produces a large loss which holds without a standing external agent: ~14-25% sustained at 10 years -> Does Weight Loss Reduce Cardiovascular Events, with an all-cause mortality benefit — «The unadjusted overall hazard ratio was 0.76 in the surgery group (P = 0.04)… and the hazard ratio adjusted for sex, age, and risk factors was 0.71 (P = 0.01).» (Sjöström et al., 2007). This is the closest thing the fabric holds to a durably lowered defended weight. Two guards: the durable loss is the observable — the fabric does not prove a mechanistic setpoint reset — and it is evidenced in the severe-obesity stratum via a matched non-RCT cohort (SOS is not randomised). No head-to-head of drug vs sustained lifestyle vs surgery on hard outcomes is held.
Reduce the provocation — blunt a driver, or don’t engage it
- Preserve lean mass (protein + resistance training). Lean mass is metabolically active, so preserving it during a deficit preserves some resting energy expenditure — one arm of the defense — and the deficit itself raises the protein requirement directionally -> GLP-1 and Lean Mass. (Adaptive thermogenesis proper is the expenditure drop beyond what body-composition change predicts, so this addresses the composition-predicted part, not necessarily that residual.) It blunts one driver; it is not shown to reset the setpoint.
- A gentler deficit. Because the held persistence is VLCD-specific, a slower food-based loss might provoke less adaptation — held as an open question, not a recommendation.
- Prevention — the cheapest mitigation. The defense is provoked by weight gain and then guards the higher level, so acting before the setpoint moves avoids engaging it at all. This is a mechanism-based directional expectation (a route-(b)-flavoured claim with no interaction evidence held that the same lever works better earlier), not an evidenced subgroup effect -> The Disease Model of Obesity.
Synthesis
The single load-bearing move: the held evidence supports countering a defended setpoint far better than weakening it by waiting. Drugs override it (and revert on stopping); surgery durably lowers weight without a standing prop (in severe obesity; whether the setpoint itself is reset is inferred, not proven); protein/resistance training blunts one arm; prevention avoids provoking it. None of these is a reset by maintenance duration, and the direct question — does six months at reduced weight wear the defense down? — has no held support and a 12-month measurement pointing against it.
This inverts the intuitive plan. Hold the line long enough and it gets easier is the reset model; the evidence fits an override / durably-lower model instead, where either you keep a standing prop in place (drug, environment) or you make a one-time structural change (surgery) — and simply enduring does not retire the defense. It is the same lesson The Disease Model of Obesity draws for the willpower model, applied to the time axis: duration is not the lever people assume it is.
Honest limits. The reset-question answer rests on a single small VLCD cohort (Sumithran, 34 completers — and the hormone panel is conditioned on >=10% achieved loss, so it is measured only in strong responders) plus RCTs never designed to test setpoint reset; the adaptation over and above current body size is cleanly established only for leptin; and the loop is open — no operation here grades any of this against a realized long-term outcome (the wiki grades coherence and source-fidelity, not validity). What would move the answer: a hormone-panel or energy-expenditure study across multiple years of maintained reduced weight, and one after slower food-based loss.