Global Obesity Surge Outpaces Underweight Decline

The World Just Crossed a Threshold Nobody Noticed

In 1990, if you had to guess which global health problem was bigger, underweight or obesity, you would have picked the wrong one. By 2022, the answer was clear, and the gap was not small. Obesity now dominates underweight in 177 countries for women and 145 for men (Phelps et al., 2024). That is not a slow shift. It is a landslide.

For decades, the story of global nutrition was a tragedy of scarcity. Millions of people did not have enough to eat. That story is still true in parts of South Asia and Africa. But a new tragedy has layered itself on top: a surge in obesity so fast and so widespread that it has overwhelmed the progress made against underweight. The combined burden of both conditions has increased in 162 countries for women and 140 for men (Phelps et al., 2024). We are not solving one problem and creating another. We are adding a new crisis while the old one refuses to leave.

How 222 Million People Told Us What We Were Missing

The study that produced these numbers is a monster of global health research. The authors, led by Nowell H Phelps and colleagues at Imperial College London, pooled data from 3663 population representative studies covering 222 million people across 200 countries and territories (Phelps et al., 2024). That is not a survey. That is a census of the world's body mass index.

They tracked two groups separately: adults (age 20 and older) and school aged children and adolescents (age 5 to 19). For adults, they used standard BMI thresholds. Underweight meant a BMI below 18.5. Obesity meant a BMI of 30 or higher. For children and adolescents, they used World Health Organization growth reference standards. Thinness meant a BMI more than two standard deviations below the median. Obesity meant more than two standard deviations above.

The researchers then asked a simple question with a complicated answer: In every country, for every year from 1990 to 2022, what fraction of the population was underweight, what fraction was obese, and how did those fractions change? They applied Bayesian statistical models to estimate trends with enough rigor that they could assign a posterior probability to each finding. When they say something changed with a probability of at least 0.80, they mean the signal is real, not noise.

The Numbers That Should Scare You

Let me give you the headline numbers because they deserve to sit in your brain without interpretation first.

For adults globally, the obesity rate more than doubled for women and nearly tripled for men between 1990 and 2022 (Phelps et al., 2024). For children and adolescents, the increase was even sharper. Obesity prevalence among school aged children and adolescents increased fourfold for girls and fivefold for boys (Phelps et al., 2024).

But here is the part that makes this study different from every other obesity report you have read. The authors did not just track obesity. They tracked underweight and obesity together, as a combined burden. They wanted to know whether the world was getting healthier overall or simply trading one form of malnutrition for another.

The answer is not good. The combined prevalence of underweight and obesity increased in 81 percent of countries for women and 70 percent for men (Phelps et al., 2024). For children, the picture was similar. Combined prevalence of thinness and obesity increased among girls in 140 countries and among boys in 137 countries (Phelps et al., 2024). Only a handful of countries managed to reduce the combined burden. For adults, that list included 11 countries for women and 17 for men (Phelps et al., 2024). For children, it was 5 countries for girls and 15 for boys (Phelps et al., 2024).

The countries that succeeded are instructive. They include some in Western Europe, parts of East Asia, and a few in Latin America. The common thread appears to be nutrition policies that simultaneously reduced underweight through food security programs while preventing obesity through regulation of processed foods and sugar. But the authors caution that the evidence for what drives success is not yet strong enough to prescribe a single formula.

Where the Two Burdens Collide

The most striking finding in the paper is not a global average. It is a map. The countries with the highest combined prevalence of underweight and obesity in 2022 were not the poorest countries or the richest. They were island nations in the Caribbean and Polynesia and Micronesia, plus countries in the Middle East and North Africa (Phelps et al., 2024). In these places, you can find underweight children living alongside obese adults in the same household. You can find a mother who does not get enough protein and a father who gets too much sugar.

This is the double burden of malnutrition, and it is more common than most people realize. The authors found that in 2022, obesity prevalence was higher than underweight in 89 percent of countries for women and 73 percent for men (Phelps et al., 2024). But the converse was still true in 16 countries for women and 39 for men (Phelps et al., 2024). Those countries are concentrated in South Asia and sub Saharan Africa. India and Pakistan, for example, still have high rates of thinness among children, even as adult obesity rises.

The authors put it plainly: "In almost all countries for both adults and school aged children and adolescents, the increases in double burden were driven by increases in obesity, and decreases in double burden by declining underweight or thinness" (Phelps et al., 2024). That sentence is a one line summary of the entire global nutrition transition. Where underweight is falling, it is because of genuine progress in food security. Where obesity is rising, it is because of a flood of cheap, calorie dense, nutrient poor foods that did not exist in most of the world 30 years ago.

What Obesity Looks Like When It Arrives Fast

The speed of change matters. Obesity did not creep up over centuries. It exploded in a single generation. For adults, the global age standardized prevalence of obesity rose from 8.8 percent for women in 1990 to 18.5 percent in 2022. For men, it rose from 4.8 percent to 14.0 percent (Phelps et al., 2024). That is a doubling in 32 years. For children and adolescents, the change was even more dramatic. Obesity prevalence among girls went from 1.7 percent to 6.9 percent. Among boys, from 2.1 percent to 9.3 percent (Phelps et al., 2024).

Think about what those numbers mean for a child born in 1990 versus a child born in 2022. The 1990 child had roughly a 2 percent chance of being obese by school age. The 2022 child has roughly a 7 to 9 percent chance. The risk has quadrupled or quintupled depending on sex. That is not a gradual drift. That is a structural change in the food environment.

The authors note that the increase was not uniform. Some countries saw much larger jumps than others. The highest obesity rates for adults in 2022 were in island nations like Nauru, the Cook Islands, and Palau, where more than 60 percent of adults were obese (Phelps et al., 2024). For children, the highest rates were in Chile, Qatar, and several Pacific island nations (Phelps et al., 2024). These are places where traditional diets have been replaced by imported processed foods, often subsidized by global trade agreements.

The Children Are Not Okay

I want to pause on the children because the data for them is particularly unsettling. The authors found that in 2022, obesity in school aged children and adolescents was more prevalent than thinness in 133 countries for girls and 125 countries for boys (Phelps et al., 2024). The converse, where thinness still outranks obesity, was true in only 35 countries for girls and 42 for boys (Phelps et al., 2024).

This means that for the first time in human history, the typical child in most of the world is more likely to be overweight than underweight. That is a reversal of millennia of evolutionary pressure. For most of our species' existence, the biggest nutritional threat was not having enough food. Now, for most children, the threat is having too much of the wrong food.

The exceptions are revealing. Thinness remains more common than obesity among children in parts of South Asia, including India, Bangladesh, and Nepal (Phelps et al., 2024). These are countries where poverty and food insecurity persist even as obesity rises among wealthier urban populations. The authors found that in India, for example, thinness among children has declined but remains high, while obesity has increased from very low levels (Phelps et al., 2024). The result is a country where both forms of malnutrition coexist at significant levels.

What This Study Does Not Tell Us

Every good study leaves questions open. This one leaves several.

First, the study measures body mass index, not body composition. BMI is a useful population level metric, but it does not distinguish between fat mass and muscle mass. A muscular athlete can have a high BMI without being obese. Conversely, a person with normal BMI but high body fat can be metabolically unhealthy. The authors acknowledge this limitation but note that BMI remains the best available metric for global comparisons (Phelps et al., 2024).

Second, the study does not tell us why the trends are happening. It describes the what with extraordinary precision, but the why requires inference. The authors point to changes in food systems, urbanization, and declining physical activity as likely drivers, but they did not test these mechanisms directly (Phelps et al., 2024). That is not a weakness of the paper. It is a boundary. The paper is a map, not a diagnosis.

Third, the study does not address the health consequences of these trends. Obesity and underweight are risk factors, not diseases themselves. The authors note that both conditions are associated with adverse health outcomes throughout life (Phelps et al., 2024), but they do not estimate how many deaths or disabilities are attributable to the shifts they document. That would require a separate analysis.

Fourth, the study cannot tell us what to do about it. The authors recommend "a healthy nutrition transition that enhances access to nutritious foods" (Phelps et al., 2024), but they do not specify which policies work best. That is a question for implementation science, not descriptive epidemiology.

The Countries That Got It Right

The exceptions to the global trend deserve attention. A small number of countries managed to reduce the combined burden of underweight and obesity. For adults, these included some Western European nations, Japan, South Korea, and a few others (Phelps et al., 2024). For children, the list was even shorter.

What did these countries do? The authors do not claim to have a definitive answer, but the pattern is suggestive. Countries that succeeded tended to have strong food regulation, universal health care that includes nutrition counseling, and social safety nets that reduce food insecurity without promoting overconsumption of processed foods. Japan, for example, has maintained relatively low obesity rates through a combination of school lunch programs, portion control norms, and a food environment that does not flood consumers with cheap sugar.

But the authors are careful not to overinterpret. The number of success stories is small, and the data do not allow for controlled experiments. We cannot randomly assign countries to nutrition policies and see what happens. We have to learn from natural experiments, which are messy and confounded.

What This Actually Means

The paper by Phelps and colleagues is not just another study confirming that obesity is rising. It is a definitive accounting of a global transition that most people have not fully absorbed. Here is what the findings imply for anyone who makes policy, treats patients, or simply tries to feed a family.

• ▸The old model of malnutrition as a problem of scarcity is obsolete for most of the world. We need to think about nutrition as a system that can produce both underweight and obesity simultaneously, sometimes in the same household. Interventions that only address one side of the problem will fail.

• ▸Children are the canary in the coal mine. The fourfold to fivefold increase in childhood obesity over 32 years means that metabolic diseases that used to appear in middle age will increasingly appear earlier. Pediatricians and school health programs need to be prepared for conditions they rarely saw 20 years ago.

• ▸Trade policy is nutrition policy. The countries with the highest combined burdens are often small island nations that import most of their food. Global trade agreements that flood these markets with cheap processed foods are directly contributing to the obesity surge. Addressing the problem will require rethinking trade rules, not just individual behavior change.

• ▸The success stories are real but fragile. Countries that reduced the combined burden did so through sustained investment in food security and regulation. These policies take years to show results and can be reversed by political changes or economic shocks. There is no permanent victory in nutrition.

• ▸Individual responsibility is a necessary but insufficient frame. Telling people to eat better and exercise more ignores the fact that the food environment in most countries has been engineered to make the unhealthy choice the easy choice. The authors do not say this explicitly, but the implication is clear: changing the environment is more effective than changing minds.

The world has crossed a threshold. For the first time, obesity is the dominant form of malnutrition globally. Underweight has not disappeared. It persists in the places that can least afford to fight two wars at once. But the direction of travel is unmistakable. The question is whether we can learn to treat malnutrition as what it has become: a problem of excess as much as deficiency, and a problem that requires a unified response rather than two separate ones.