Newer bariatric surgery criteria flagged 2.8 times as many adults as the 1991 rules in 37 countries

Bariatric surgery guidelines changed in 2022. A new analysis puts a number on how much that change widened the group of people who would be sent for a surgical evaluation, and the answer is close to three times as many.

The study was published in Obesity Pillars on 29 August 2026. It applied both the old and the new criteria to World Health Organization survey data covering 37 countries and territories, then projected the results forward to 2030.

Across those settings, 17.1 million adults, or 3.6%, met the 1991 threshold. 48.5 million adults, or 10.4%, met the 2022 threshold. The ratio between the two was 2.84.

The two rule sets differ by about five BMI points

Body mass index, or BMI, is weight divided by height squared. It is the number most surgical programs still use to decide who gets evaluated.

In 1991 a National Institutes of Health panel wrote the first widely used criteria. As this study applied them, an adult met the 1991 threshold with a BMI of 40 or higher, or a BMI of 35.0 to 39.9 plus a weight-related metabolic condition.

In 2022 the American Society for Metabolic and Bariatric Surgery and the International Federation for the Surgery of Obesity and Metabolic Disorders lowered both numbers. Under those criteria, an adult met the threshold with a BMI of 35 or higher on its own, or a BMI of 30.0 to 34.9 plus a metabolic condition.

The study could count only two metabolic conditions, high blood pressure and diabetes, because those were the only ones recorded the same way across every survey. Both were counted when a person reported a previous diagnosis or reported taking medication for them.

Most of the extra 31.5 million have a BMI under 35

The gap between the two thresholds came to 31.5 million adults. Of those, 26.5 million had a BMI between 30.0 and 34.9 along with high blood pressure or diabetes. The remaining 4.9 million had a BMI between 35.0 and 39.9 with no recorded metabolic condition, which put them over the 2022 line but under the 1991 one.

So the change did not mainly reach people at the highest weights. It reached people with moderate obesity who also have a metabolic illness, and people in the mid to high 30s who are otherwise healthy.

Rates were higher among women and among older adults, and varied widely between settings.

The author also ran a second, exploratory scenario assuming obesity keeps rising rather than holding steady. In that version 64.2 million adults met the 2022 threshold by 2030, and rising obesity rather than population growth accounted for 51.7% of the increase.

The United States was not among the 37 settings

This matters for anyone trying to apply the figures to themselves. The data came from WHO STEPS surveys, which are run mostly in low and middle income countries. The included settings fell into five WHO regions: nine in Africa, seven in the Americas, five in the Eastern Mediterranean, nine in Europe and seven in the Western Pacific. None were in South East Asia. The analysis covered 138,695 survey participants standing in for 468.2 million adults aged 18 to 69.

The author is explicit that the 37 settings were included because usable data existed, not because they were picked to represent any region, and that no regional or global rate should be read out of them.

Meeting a threshold is not the same as being a surgical candidate

The study states this directly and more than once. Survey data cannot see previous treatment, contraindications, operative risk, a multidisciplinary assessment, cost, local surgical capacity, or what a patient actually wants. The estimates describe how many people would warrant a closer look, not how many should have an operation or how many will get one.

Other limits are worth knowing. Diabetes was identified from a self-reported diagnosis or medication use rather than a blood test, because blood measurements were not available across all surveys. BMI is an imperfect stand-in for body fat and for risk, and behaves differently across populations. The metabolic definition left out every weight-related condition other than high blood pressure and diabetes, which almost certainly undercounts. The surveys ran between 2015 and 2024 and were cross-sectional, so the 2030 figures rest on modeling assumptions layered on top of them.

The paper carries a single author, Victor J. Vera Ponce, who reported no competing interests. Article processing costs were covered by the Vice-Rectorate for Research of the National University Toribio Rodriguez de Mendoza of Amazonas in Peru, which the paper says had no role in the study.