Anorexia and Bulimia Cases Rose 71% Worldwide Since 1990, Study Finds

An estimated 18.1 million people worldwide were living with anorexia nervosa or bulimia nervosa in 2023, 71 percent more than in 1990, according to a Global Burden of Disease analysis published Oct. 28 in JAMA Psychiatry. The same model projects the combined case count will grow another 20 percent by 2050.

The study is the first substantial update to the eating disorder estimates inside the Global Burden of Disease framework since 2017, and its authors describe it as a correction to a long-standing gap rather than a new discovery about either illness. "Anorexia nervosa and bulimia nervosa contribute substantially to global health loss, and their prevalence is forecast to continue increasing," the researchers wrote, adding that the findings support strengthening prevention, early detection and treatment capacity in regions where case numbers are growing fastest.

What the 2023 snapshot shows

Of the 18.1 million people estimated to be living with one of the two disorders, 69 percent were female and 41 percent were between the ages of 10 and 24. Those proportions matter for how the numbers translate into service demand: the concentration in adolescents and young adults is the age band in which eating disorder treatment is most often delivered, and the one in which delays between symptom onset and care produce the longest illness durations.

Disability-adjusted life-years — the measure the study uses, which combines years of life lost to premature death with years lived in reduced health — rose 21 percent between 1990 and 2023 for the two diagnoses combined.

Bulimia carries most of the caseload

The totals are dominated by bulimia nervosa, which accounted for 14.2 million of the 18.1 million cases against 3.9 million for anorexia nervosa. That ratio is one of the more consequential numbers in the paper for anyone reading it as a planning document, because the two diagnoses reach care through different referral pathways: anorexia is typically identified through weight loss that is visible to family and clinicians, while bulimia can persist for years at a body weight that raises no alarm.

The researchers found that prevalence was generally higher in Australasia and other high-income regions. They are careful about what that means: the difference, they wrote, may reflect not only epidemiological variation but also differences in data availability — a caveat that undercuts any straightforward reading of the map as a map of risk.

A forecast built on acknowledged gaps

The 2050 projection — 4.7 million cases of anorexia nervosa and 17 million of bulimia nervosa — is an extrapolation from a model whose inputs the authors describe as uneven. Coverage of low-income countries in particular remains thin, and the paper's estimates for those regions rest on less country-level data than the estimates for wealthy ones.

Two categories were left out of the analysis entirely. Binge eating disorder and other specified feeding or eating disorders were not estimated, the researchers wrote, because the framework lacks data on their disability burden. The published total therefore describes the two named diagnoses and says nothing about how many people worldwide have a clinically significant eating disorder of another type.

The authors also flagged the limits of the underlying measure. A disability-adjusted life-year is a modeled quantity built from years of life lost and years lived with disability, and the disability weight assigned to a condition is an input to it rather than a measured outcome.

What changes at the level of care

For clinicians, the paper's practical content is its recommendation rather than its arithmetic. The authors call for eating disorder care to be integrated into broader mental health strategies, which in the American system would mean routing patients through general behavioral health settings rather than only through specialty programs — an approach that already exists in intensive outpatient programs and in the partial hospitalization level of care.

That recommendation sits alongside a body of recent work suggesting the diagnostic boundaries are less stable than the service categories built around them. A Danish cohort published in October found that off-schedule puberty raised anorexia odds in girls, locating part of the risk in developmental timing. The burden study does not attempt that kind of mechanistic question; its contribution is the scale of the trend line and the forecast behind it.

The paper's own hedge is worth keeping in view. A 71 percent increase over three decades sounds like an epidemic curve, and the researchers do not describe it as one. They describe a disorder group that has been chronically under-measured, now measured somewhat better, and projected forward on the assumption that the recent trajectory continues.

Ruhai Bai of Nanjing University of Science and Technology is the paper's first author. Damian F. Santomauro of the Institute for Health Metrics and Evaluation at the University of Washington is among the co-authors.