Appetite Regulation Statistics: Key Research and Trends

The science of appetite has moved a long way from the old assumption that hunger is a matter of discipline. A body of controlled research now describes appetite as a regulated physiological system — one with measurable set points, hormonal feedback loops and predictable responses to weight loss. This page collects the statistics that anchor that view. It is intended as a citation-ready reference: each figure below is attributed to a specific study or surveillance dataset, with the year given inline, so that the numbers can be quoted and checked.

The figures are grouped by theme — prevalence, hunger hormones, satiety and ultra-processed food, food noise, weight regain, and the efficacy of GLP-1 medicines. For the mechanisms behind the numbers, our hunger hormones explained primer and the broader appetite regulation guide set out how the system works; this page concentrates on the evidence.

Obesity and appetite: how common is the problem

The scale of obesity is the necessary backdrop to any discussion of appetite, because it is the population-level consequence of a dysregulated appetite system meeting an abundant food environment.

• 40.3% of US adults had obesity over the August 2021–August 2023 measurement period, according to the CDC's National Center for Health Statistics (NCHS Data Brief No. 508, 2024). The figure was 39.2% in men and 41.3% in women.
• 9.4% of US adults had severe obesity over the same period — a category that has risen from 7.7% a decade earlier even as overall obesity has plateaued (CDC/NCHS, 2024).
• The highest prevalence fell in the 40–59 age band, at 46.4%, compared with 35.5% in adults aged 20–39 (CDC/NCHS, 2024).
• Globally, around 890 million adults were living with obesity in 2022 — roughly one in eight people — and more than one billion people of all ages, according to the World Health Organization's March 2024 fact sheet. The WHO reports that worldwide obesity more than doubled between 1990 and 2022.
• In 2022, 43% of adults worldwide were overweight (a group of 2.5 billion people), of whom the obese subset is counted above (WHO, 2024).

Two features of these figures repay attention. First, the US prevalence has plateaued rather than fallen: from 2013–2014 to 2021–2023 the age-adjusted obesity rate did not change significantly, even as severe obesity continued to climb (CDC/NCHS, 2024). A plateau at 40% is not a recovery. Second, the global trajectory is still rising — the WHO's statement that worldwide obesity more than doubled between 1990 and 2022 reflects rapid increases across low- and middle-income countries as their food environments change, not merely the wealthy nations where the trend began.

These are surveillance figures, drawn from large representative samples and updated on multi-year cycles, and they describe the outcome rather than the cause. They are also measured, not self-reported: the CDC's NHANES programme weighs and measures participants directly, which avoids the well-known tendency for people to under-report weight and over-report height in surveys. The interesting question is why appetite so reliably defends a high body weight once it is reached, and the hormone data below begin to answer it. For the wider context, see our metabolism guide.

Hunger hormones after weight loss

The single most influential dataset on appetite regulation comes from Priya Sumithran and Joseph Proietto's 2011 study in the New England Journal of Medicine. It is the study most often cited to explain why diets fail, and the numbers deserve to be quoted precisely.

• Fifty adults with overweight or obesity completed a ten-week very-low-energy diet and lost a mean of 13.5 kg (Sumithran et al., 2011).
• One year later, ghrelin — the principal hunger hormone — remained significantly elevated above its pre-diet baseline (P<0.001). The hunger signal had been turned up, not down, by weight loss (Sumithran et al., 2011).
• Over the same year, the satiety hormones leptin, peptide YY, cholecystokinin and others remained suppressed, and subjective appetite and hunger ratings stayed significantly raised (P<0.001) (Sumithran et al., 2011).

In other words, nearly every appetite hormone the investigators measured had shifted in the direction that promotes eating, and the shift had not faded a full year after the diet ended. This pattern is explored in our pieces on ghrelin and dieting and why diets fail (the biology). The earlier work of David Cummings and colleagues (2002, also in the NEJM) had already shown ghrelin rising after diet-induced weight loss; Sumithran's contribution was to show that the change persists rather than resolving as the body adapts to its new, lower weight.

It is worth being precise about what these numbers do and do not show. The study measured circulating mediators of appetite — leptin, ghrelin, peptide YY, gastric inhibitory polypeptide, glucagon-like peptide-1, amylin, pancreatic polypeptide, cholecystokinin and insulin — together with subjective ratings of hunger and fullness, at baseline, at the end of the ten-week diet, and at 62 weeks. The headline finding is directional and durable: the integrated change pushed toward greater hunger and reduced satiety, and the difference from baseline was still statistically significant a year out. What the study did not do is show that the hormones never normalise; it shows that, at one year, they had not. That distinction matters when the figure is quoted, and it is why we describe the shift as persistent rather than permanent.

The practical reading of these figures is that a dieted body is hormonally distinct from a never-dieted body of the same weight — a point developed in the hunger and satiety guide and our explainer on leptin resistance. Two people may weigh the same and yet experience hunger quite differently depending on whether they arrived at that weight by losing or by maintaining. This is the physiological core of the argument that obesity is a defended state, and it reframes the high prevalence figures above: a population is not simply failing to restrain itself, it is contending with a regulatory system that resists downward change.

Satiety and ultra-processed food

If hormones explain why weight is defended, the food environment helps explain why it climbs in the first place. The cleanest experimental evidence comes from Kevin Hall's 2019 inpatient trial in Cell Metabolism, which remains the most-cited controlled test of ultra-processed food and energy intake.

• Twenty weight-stable adults lived on a metabolic ward and were given either an ultra-processed or an unprocessed diet for two weeks, then crossed over to the other. The two diets were matched for presented calories, energy density, macronutrients, sugar, sodium and fibre (Hall et al., 2019).
• On the ultra-processed diet, participants ate about 508 kcal more per day (P=0.0001) than on the unprocessed diet, with the excess coming from carbohydrate and fat rather than protein (Hall et al., 2019).
• They gained roughly 0.9 kg over two weeks on the ultra-processed diet and lost a similar amount on the unprocessed one (Hall et al., 2019).

Because the diets were matched on paper, the extra intake cannot be attributed to nutrient composition alone; something about the form and palatability of ultra-processed food weakened satiety. The trial also recorded a behavioural clue: participants ate faster on the ultra-processed diet, and faster eating gives the gut's satiety signals less time to register before the next mouthful arrives (Hall et al., 2019). The investigators proposed that energy density, texture, eating rate and the weaker post-meal hormonal response together explain the gap, rather than any single villain nutrient. We treat the mechanisms in why ultra-processed food does not satisfy and the broader science of satiety.

The size of the effect is what makes the trial so frequently cited. A roughly 500 kcal/day surplus, sustained, is more than enough to drive the population trends in the prevalence section above — and the participants were not instructed to eat more, nor were they aware that intake was the outcome of interest. They simply ate to appetite, and appetite delivered a different answer depending on how the food was made. This is a tightly controlled, crossover design in which each person served as their own control, which is why a study of only twenty people carries such weight. It does not, on its own, prove that ultra-processed food causes obesity at the population level; it provides a clean mechanistic demonstration that the food environment can override appetite regulation even when the nutrition label looks identical.

Food noise and GLP-1 effects on cravings

"Food noise" — the intrusive, persistent preoccupation with food that many people with obesity describe — has only recently been measured systematically, largely because GLP-1 medicines appear to quieten it. The figures here come from a Novo Nordisk survey of 550 US adults on semaglutide, presented at the European Association for the Study of Diabetes (EASD) annual meeting in September 2025.

• The share of respondents reporting constant thoughts about food fell from 62% to 16% after starting treatment (EASD 2025).
• Those saying they spent too much time thinking about food fell from 63% to 15%; uncontrollable thoughts about food fell from 53% to 15% (EASD 2025).
• 64% reported improved mental health, 76% improved self-confidence, and 80% reported developing healthier habits on treatment (EASD 2025).

These are self-reported survey data from people already taking the medicine, not a randomised trial of food noise as an endpoint, and they should be read with that limitation in mind. There is no placebo arm here, the respondents knew they were on treatment, and recall of how one felt before starting is imperfect; the effect sizes are therefore best treated as indicative rather than precise. We flag this deliberately, because the food-noise figures are among the most widely shared appetite statistics and among the most often quoted without their caveats.

That said, the direction is consistent with the proposed mechanism — GLP-1 receptor agonists act on the brain's reward and appetite circuits, not only on the gut, which is why many patients describe a qualitative change in their relationship with food rather than merely feeling fuller after meals. The reduction in intrusive, uncontrollable thoughts about food maps onto what would be expected if the drug dampens the salience of food cues. The honest summary is that the food-noise concept is now well-described clinically and biologically plausible, but the quantitative survey figures await confirmation from controlled studies designed to measure it directly. Our article on how GLP-1 influences satiety and the explainer on food noise set out what is and is not established.

Weight regain after dieting

Regain is the rule rather than the exception, and the GLP-1 withdrawal trials have produced unusually clean numbers because they isolate the effect of removing a single intervention.

• In the STEP 1 trial extension (Wilding et al., 2022), participants who stopped semaglutide regained about two-thirds of their lost weight within a year. Specifically, they regained 11.6 percentage points of the weight they had lost, leaving a net loss of 5.6% from the original starting weight (Wilding et al., 2022).
• Cardiometabolic improvements — blood pressure, lipids, glycaemia — reverted in step with the weight, underlining that the benefits depend on continued treatment (Wilding et al., 2022).

The figures from the extension are worth stating in full, because they are easy to misread. From the original starting weight to week 68, the semaglutide group had lost a mean of 17.3% (the extension cohort's figure, slightly higher than the headline 14.9% trial average because of who continued into the extension). Following withdrawal of the drug and the lifestyle support, they regained 11.6 percentage points by week 120 — about two-thirds of what they had lost — ending at a net 5.6% below where they began (Wilding et al., 2022). So the medicine did not stop working in the sense of leaving no trace; rather, the loss it produced was conditional on continued use, and most of it returned once use stopped.

The regain is not a failure of resolve; it is the predictable result of the hormonal adaptations documented by Sumithran reasserting themselves once the counter-signal is withdrawn. A GLP-1 medicine supplies a satiety signal strong enough that the body tolerates a calorie deficit without mounting its usual counterattack; remove that signal and the underlying biology — elevated ghrelin, suppressed satiety hormones, a metabolic rate lowered by weight loss — is still there, waiting. We discuss this at length in weight regain after dieting is not your fault and why diets fail. The same physiology is why obesity is increasingly framed by clinicians as a chronic, relapsing condition requiring ongoing management, in the way that hypertension or type 2 diabetes are — treatments that control the condition while taken and lose their grip when stopped, rather than cures.

Appetite and weight regain by the numbers

The chart below brings together the headline figures from this page: the population prevalence statistics, the experimental hunger and satiety findings, and the trial efficacy of the two leading GLP-1 medicines. The contrast between diet-induced loss — which the body fights and largely reverses — and pharmacologically supported loss is the central story of the modern appetite literature.

GLP-1 trial efficacy: STEP and SURMOUNT

The pivotal trials of semaglutide and tirzepatide are the most-cited GLP-1 weight loss statistics, and the numbers are worth stating exactly, because they are frequently rounded or conflated.

STEP 1 — semaglutide 2.4 mg

• In Wilding et al. (2021, NEJM), 1,961 adults with overweight or obesity and without diabetes received once-weekly semaglutide 2.4 mg or placebo, both with lifestyle support, for 68 weeks.
• Mean weight change was −14.9% with semaglutide versus −2.4% with placebo (Wilding et al., 2021).
• 86.4% of the semaglutide group lost at least 5% of body weight (vs 31.5% on placebo); 69.1% lost at least 10% (vs 12.0%); and 50.5% lost at least 15% (vs 4.9%) (Wilding et al., 2021).

SURMOUNT-1 — tirzepatide

• In Jastreboff et al. (2022, NEJM), 2,539 adults with obesity or overweight received tirzepatide 5, 10 or 15 mg, or placebo, for 72 weeks.
• Mean weight change was −20.9% at the 15 mg dose and −19.5% at 10 mg, versus −3.1% on placebo (Jastreboff et al., 2022).
• 57% of the 15 mg group lost at least 20% of body weight (vs 3% on placebo), and 36.2% lost at least 25% (vs 1.5%) (Jastreboff et al., 2022).

Reading the trial figures fairly

A few caveats keep these numbers honest. Both trials enrolled adults without type 2 diabetes — the figures are higher than those typically seen when the same drugs are tested in people with diabetes, in whom weight loss is consistently more modest. Both delivered the medication alongside structured lifestyle support, so the placebo arms are not "no treatment" but diet-and-lifestyle alone, which is why the placebo groups still lost 2–3%. And these are means: in STEP 1 about one participant in seven on semaglutide lost little or no weight, while others lost a great deal, so the average conceals a wide spread. The trials also ran for 68 and 72 weeks respectively, long enough to demonstrate efficacy but not to settle questions of very-long-term maintenance, which is precisely where the withdrawal data above become relevant.

With those qualifications, the trials remain landmark results: the largest mean reductions recorded for pharmacotherapy in obesity, with tirzepatide's 20.9% approaching the range once reserved for bariatric surgery. The contrast with the diet literature is the point. A very-low-energy diet in the Sumithran cohort produced a meaningful initial loss that the body then defended against and largely reclaimed; the GLP-1 medicines produce a larger loss that holds for as long as treatment continues. The difference is not effort but whether the appetite system is fought or supplied. For how the drugs produce these results, see how GLP-1 influences satiety; for why the loss must be sustained with continued use, return to the regain figures above. A fuller curated overview sits in our appetite regulation hub and the appetite and hunger collection.

How this page is maintained and last updated

This reference was last reviewed in June 2026. It is updated when the underlying sources change — for example, when the CDC releases a new NHANES surveillance cycle, when the WHO revises its obesity fact sheet, or when major new trials (such as further STEP and SURMOUNT readouts, or head-to-head comparisons) are published in peer-reviewed journals. Each statistic is tied to a named, dated source in the references below, and figures that cannot be traced to a primary source are not included. Where a number is a trial mean, it is described as such, and survey-based figures are flagged as self-reported rather than experimental. If you cite this page, please cite the underlying primary source alongside it.

Key takeaways

• 40.3% of US adults (CDC, 2024) and roughly 890 million adults worldwide (WHO, 2024) have obesity — a regulated, defended condition, not a behavioural one.
• After a 13.5 kg diet-induced loss, ghrelin stays elevated and satiety hormones stay suppressed for at least a year (Sumithran, 2011).
• Ultra-processed food drove ~508 extra kcal/day in a matched-diet trial (Hall, 2019).
• GLP-1 treatment is associated with a 62%→16% drop in constant food thoughts (EASD 2025 survey), but stopping it returns about two-thirds of lost weight (Wilding, 2022).
• Trial efficacy: 14.9% with semaglutide (STEP 1) and 20.9% with tirzepatide 15 mg (SURMOUNT-1).