Global waist thresholds missed 87.5% of Qatari women with obesity by body fat

An analysis of 6,000 Qatar Biobank participants derived population-specific thresholds for body mass index and waist circumference — and found that applying the World Health Organization's global cut-points instead misclassified 54.6% of men and 43.0% of women in the sample [s1].

For waist circumference in women the gap was larger still: 87.5% of women who met an obesity definition by measured body-fat percentage were classified as not having obesity by the WHO waist threshold [s1].

What the study did

BMI and waist circumference are proxies. What clinicians actually care about is adiposity and the cardiometabolic risk that comes with it, and the thresholds in use — a BMI of 30, particular waist circumference numbers — are conventions calibrated largely on European-descent populations.

The Qatari analysis used a cross-sectional sample of 6,000 Qatar Biobank participants and ran receiver operating characteristic curve analyses to find the anthropometric cut-points that best aligned with measured body-fat percentage and with disease-specific thresholds for elevated blood pressure, diabetes, and dyslipidaemia [s1]. Having body-fat percentage available as a reference standard is what makes the exercise possible; without it, a cut-point can only be validated against outcomes, not against adiposity itself.

The numbers they derived

The optimal BMI cut-points were 25.2 kg/m² for men and 24.8 kg/m² for women [s1] — close to what WHO classifies as the overweight threshold rather than the obesity threshold. Waist circumference cut-points came out at 84.3 cm for men and 74.5 cm for women, also below global standards [s1].

The study sample is described as relatively young adults with high levels of adiposity [s1]. That combination is itself informative: a young population already carrying substantial body fat is exactly the group in which a threshold calibrated elsewhere is most likely to under-detect, because BMI conflates fat with muscle and bone and the relationship between them varies by population and by age.

The authors also report that disease-specific thresholds varied depending on which condition was being predicted, which they attribute to the distinct biological pathways underlying elevated blood pressure, diabetes and dyslipidaemia [s1]. That is a caveat against the idea of a single correct number: the waist circumference that best flags diabetes risk is not necessarily the one that best flags hypertension risk.

Why the misclassification figure is the headline

More than half of the men and 43% of the women in a national biobank sample being misclassified by the standard BMI cut-point is a large number, and its practical meaning is asymmetric. Misclassification in this direction — WHO thresholds calling people non-obese who have obesity by measured body fat — means people who might benefit from earlier intervention do not trigger a screening flag.

The 87.5% waist figure in women is the most extreme case in the paper. A screening measure that misses seven in eight of the people it is meant to identify is not functioning as a screening measure in that population.

The limits

This is cross-sectional. The cut-points were optimised against concurrent measurements of body fat and current disease markers, not against future cardiovascular events — which is the outcome that would ultimately justify a threshold. Cut-points derived this way can shift when validated prospectively.

The sample also comes from a biobank, and biobank volunteers are not a random sample of a population. The paper describes the cohort as relatively young; thresholds derived in a young sample may not transfer to older adults, in whom body composition changes.

And the study derives thresholds; it does not test them. Whether screening Qatari adults at a BMI of 25.2 rather than 30 improves outcomes is a separate empirical question that would require its own study. Lowering a threshold increases detection and also increases the number of people labelled with a condition, and this paper does not weigh those against each other.

None of the above is clinical guidance. Anyone's own measurements and what they mean are a matter for their clinician.

The wider argument

The paper sits in a long-running dispute about whether global anthropometric standards should be global at all. Lower BMI thresholds have already been adopted for some South Asian and East Asian populations on similar reasoning — that cardiometabolic risk appears at lower BMI than the European-calibrated cut-points assume.

What this study adds is an Arab-population dataset to that argument, with body-fat percentage as the reference standard rather than outcomes alone. The authors' framing is explicitly local: tailored cut-points as a tool for clinical and public health use in Qatar [s1], not a proposal to revise global standards.

What to watch next

Whether other Gulf biobanks replicate the derivation. Qatar, the UAE and Saudi Arabia all have population cohorts of comparable design, and cut-points that converged across several of them would carry considerably more weight than one country's numbers — particularly for a region where diabetes prevalence is among the highest recorded anywhere.

Sources

1. Deriving sex-specific anthropometric cut-points for obesity and cardiovascular disease risk in Qatari adults — International Journal of Obesity, 10 December 2025 (primary)