Obesity Research Timeline: How the Science Evolved

For most of recorded history, body fat was read as a moral signal — a sign of indulgence or, in leaner times, of prosperity. The idea that obesity might be a biological condition, governed by physiology rather than character, is surprisingly recent. It had to be assembled piece by piece, against a strong cultural intuition that weight is simply a matter of intake and willpower. The story below is the record of that assembly: a thread that runs from a nineteenth-century statistician, through a wartime starvation study and a prison overfeeding experiment, to the discovery of the hormones that regulate appetite and, at last, to medicines that act on them.

It is meant as a reference — a single place to check the chronology and its sources — rather than an argument. If you want the argument itself, our explainer on why obesity is a disease, not a failure of willpower makes the case at length. What follows is the evidence in the order it arrived.

The obesity research timeline

The milestones below condense nearly two centuries of inquiry into a single sequence. Each is anchored to the peer-reviewed literature or to the regulatory decision that marked it; the eras that group them are explained in the sections that follow.

Measuring obesity: Quetelet and the origins of BMI

The science of obesity began, oddly, with no interest in obesity at all. In the 1830s and 1840s the Belgian polymath Adolphe Quetelet — an astronomer and statistician by training — set out to describe what he called l'homme moyen, the average man. Gathering measurements across populations, he noticed that adult weight tended to scale with the square of height, and he proposed the ratio of weight to height squared as a convenient way to characterise a population's build. He never intended it as a measure of individual health, still less as a diagnostic tool. It was a piece of social physics.

The ratio lay largely dormant for over a century. It was the American physiologist Ancel Keys — whom we will meet again — who revived it. In a 1972 paper Keys compared several indices of relative weight and concluded that Quetelet's ratio performed best as a population measure; he gave it the name by which it is now universally known, the body mass index. That renaming mattered, because it supplied epidemiology with a single, cheap, reproducible number against which the health consequences of excess weight could be charted. BMI's well-known limitations — its blindness to muscle, frame and fat distribution — flow directly from its origins as a statistical average rather than a clinical measure, a tension that has never fully been resolved.

The regulation era: starvation, the lipostat and overfeeding

The next advance came from asking a different question: not how to measure body weight, but how the body itself controls it. The first great contribution was, again, Ancel Keys's. Between November 1944 and late 1945, at the University of Minnesota, Keys supervised 36 conscientious objectors through the Minnesota Starvation Experiment — a controlled study of prolonged semi-starvation designed to inform the refeeding of Europe's wartime famine victims. After a baseline period the men were held on roughly 1,600 calories a day for six months, losing about a quarter of their body weight. The findings, published in 1950 as the two-volume The Biology of Human Starvation, documented far more than physical wasting: an overwhelming preoccupation with food, depression, irritability, social withdrawal and a metabolic rate that fell further than the loss of tissue alone could explain. It was the first rigorous demonstration that the mind and metabolism mount a coordinated defence against weight loss.

If starvation revealed a defence on one side, a hypothesis was needed to explain it. In 1953 the Cambridge physiologist Gordon Kennedy proposed the lipostatic hypothesis: the brain, he argued, monitors the body's fat stores by sensing a circulating signal proportional to fat mass, and adjusts appetite and energy use to hold those stores near a defended level. It was an elegant idea years ahead of the means to test it — the signal Kennedy postulated would not be identified for another four decades. But it reframed body weight as a regulated variable, like temperature or blood glucose, rather than a passive tally of calories. This is the intellectual root of what later became known as set-point theory, and of the broader insight explored in our metabolism guide that the body actively works to maintain its mass.

The regulation came into still sharper focus in the opposite direction. Through the 1960s the Vermont physician Ethan Sims ran a series of deliberate overfeeding studies, encouraging volunteers — among them inmates at a state prison — to gain weight by eating well beyond their needs. The result, reported in 1971, surprised everyone: many subjects found it genuinely hard to gain, requiring enormous intakes to add modest weight, and most shed the surplus easily once normal eating resumed. Their metabolic rates had risen to resist the gain. Together, Keys's starvation work and Sims's overfeeding work bracketed the same conclusion from both ends: the body defends a weight, fighting both loss and gain. The lipostat was no longer just a hypothesis; it had a behavioural signature.

The genetics era: twins, adoptees and the discovery of leptin

A regulated system implies a set point, and a set point invites an obvious question: what determines it? By the 1980s the tools of human genetics could begin to answer. The decisive evidence came from Albert Stunkard and his collaborators. In 1986 they published a study of 540 Danish adoptees, comparing each person's weight class with that of both their biological and their adoptive parents. The adoptees resembled their biological parents across the whole range from thin to obese, and bore no resemblance to the families that had raised them. In 1990 Stunkard's group followed this with a study of identical and fraternal twins, some reared together and some apart, and arrived at a heritability for BMI of roughly 0.7 — comparable to that of height. Shared childhood environment, the data suggested, had little lasting effect. These findings did not deny that environment shapes the population's weight; rather, they showed that within a shared environment, who becomes heavier is strongly influenced by inheritance.

Genetics pointed to a biological mechanism, but the mechanism itself remained hidden until 1994. That year Yiying Zhang, Jeffrey Friedman and colleagues at Rockefeller University used positional cloning to identify the gene mutated in a strain of grossly obese mice — the ob gene — and the hormone it encodes. They named the hormone leptin, from the Greek leptos, meaning thin. Leptin is secreted by fat tissue in proportion to its mass and acts on the brain to signal that energy stores are adequate; mice unable to make it eat ravenously and become enormous. Here, at last, was the circulating fat signal Kennedy had postulated in 1953. The discovery electrified the field and, briefly, raised hopes of a simple cure. Those hopes faded when it emerged that people with common obesity usually have high leptin levels, not low ones — their brains have grown resistant to the signal, a phenomenon our explainer on leptin resistance examines in detail.

Leptin's discovery did more than fill a gap. It established, in molecular terms, that body weight is an endocrine variable — something hormones report on and regulate. The mechanistic backbone of why deliberate weight loss is so hard to sustain, set out in our piece on why diets fail for biological reasons, rests on the foundation leptin laid.

The hormonal era: energy gaps, ghrelin and persistent adaptation

If leptin reported on fat stores, the 1990s and 2000s filled in the rest of the signalling network — and, crucially, quantified how it behaves during and after weight loss. In 1995 Rudolph Leibel, Michael Rosenbaum and Jules Hirsch published a landmark study in the New England Journal of Medicine. They had volunteers gain and lose weight under controlled conditions and measured their energy expenditure at each point. The finding was stark: after losing 10 to 20 per cent of body weight, subjects burned significantly fewer calories than their new, smaller size predicted — an energy gap of a few hundred calories a day. The lighter body was, in effect, running on less fuel than expected, biasing it back towards its former weight. This adaptive fall in expenditure is the quantitative core of the difficulty described in weight regain and hunger hormones.

The appetite side of the ledger gained its key player in 1999, when Masayasu Kojima and colleagues in Japan isolated a hormone from the stomach that powerfully stimulated growth-hormone release and, it soon emerged, hunger. They named it ghrelin. Where leptin signals sufficiency, ghrelin signals deficit: its levels rise before meals and fall after eating, and they climb during dieting. Ghrelin gave the field a circulating hunger hormone to set against leptin's satiety signal, and it explained, in part, why weight loss provokes such insistent appetite — the theme of our explainer on ghrelin, the hunger hormone, and dieting. The wider physiology of how these signals are integrated is covered in our appetite regulation guide.

The era's most sobering result came in 2011. Priya Sumithran, Joseph Proietto and colleagues at the University of Melbourne put overweight and obese volunteers through a ten-week diet, then measured their appetite hormones — leptin, ghrelin, peptide YY and others — at the end of the diet and again a full year later. The hormonal profile remained shifted towards hunger long after the weight had been lost; the body was still, twelve months on, behaving as though it were starving and pressing for the lost weight to be regained. The persistence was the point. It showed that the metabolic and hormonal defence Keys had glimpsed in 1944 does not relent with time, and supplied a physiological account of why most diets are followed by regain.

The disease and treatment era: recognition and the GLP-1 medicines

By the 2010s the accumulated science had outrun the prevailing cultural view. A condition governed by inherited set points, defended by falling metabolism and shifting hormones, and resistant to sustained willpower did not fit the model of a lifestyle choice. In June 2013 the American Medical Association formally recognised obesity as a disease — a contested vote, taken against the cautious advice of its own science council, but a symbolically decisive one. The reclassification did not settle the biology, but it changed the framing: obesity became a medical condition warranting treatment and, in principle, insurance coverage, rather than a personal shortcoming to be scolded away.

Recognition without effective treatment, however, would have rung hollow. The treatments arrived from a separate line of research — the decades-long investigation of the gut hormone GLP-1, told in full in our survey of the future of obesity science. GLP-1 is one of the satiety signals released after a meal, and medicines that mimic it, developed first for type 2 diabetes, turned out to produce weight loss of a magnitude the field had never seen from a drug. In 2021 the regulators approved high-dose semaglutide for chronic weight management, the first of a class delivering average losses approaching 15 per cent of body weight, with later dual-agonist drugs reaching higher still. For the first time, the regulated system the previous eras had described could be acted upon pharmacologically — the appetite signals turned down, the defended weight nudged lower.

This is why the timeline matters as a whole rather than as a list. The GLP-1 era did not appear from nowhere; it is the practical consequence of everything before it. Without Kennedy's lipostat and Keys's starvation work there is no concept of a defended weight; without Stunkard there is no genetic mandate; without leptin, ghrelin and Sumithran's hormones there are no targets to aim at; without the AMA's 2013 ruling there is no medical framework in which to deploy the drugs. The science had to establish that obesity was a regulated, biological, treatable condition before the treatment could be understood for what it is.

What the timeline reveals

Read end to end, the chronology carries a few lessons worth stating plainly. The first is that the body defends its weight, and does so by several independent mechanisms — falling metabolism, rising hunger hormones, persistent appetite — that the research uncovered one at a time across half a century. The second is the long lag between hypothesis and proof: Kennedy postulated a fat signal in 1953, and it took until 1994 to hold it in hand. The third is that the shift from blame to biology was not ideological but evidential; it followed the data, slowly, often against resistance.

The story is not finished. Leptin resistance remains incompletely understood, the durability of the new medicines is still being established, and the question of how environment and genes combine to produce the modern prevalence of obesity is far from closed. For the wider evidence base, our weight-loss research category collects the explainers, and the weight-loss research hub gathers the deeper material in one place. What began as a statistician's idea of the average man has become one of the more consequential stories in modern medicine — and, on present evidence, it is still being written.