BMI misclassifies two-thirds of children as overweight when compared to waist-to-height ratio
New research reveals that BMI significantly overestimates childhood overweight prevalence compared to waist-to-height ratio measurements. The study of 7,600 children found that 64% of children classified as overweight by BMI standards actually had normal fat levels when measured using waist-to-height ratio, highlighting potential issues with BMI-based obesity diagnosis in paediatric populations.
A large-scale study from the University of Eastern Finland and UK collaborators has found that body mass index (BMI) is almost three times more likely to classify children as overweight than waist circumference-to-height ratio (WHtR), potentially leading to significant misclassification in paediatric obesity assessment.
The research, published in the journal Obesity and Endocrinology, analysed data from 7,600 children drawn from the University of Bristol’s Children of the 90s cohort who were followed from ages 9 to 24 years. It represents the largest follow-up paediatric study comparing BMI and waist-to-height ratio in the world.
BMI significantly overestimates overweight prevalence
The study found stark differences between the two measurement methods. Of the 1,431 children classified as BMI-overweight at age 9 years, only 25% had WHtR measurements indicating high fat, and just 11% showed excess fat according to WHtR measurements. Remarkably, 64% of children deemed overweight by BMI standards actually had normal fat levels when assessed by WHtR.
Conversely, when looking at children with high fat measurements via WHtR, the correlation with BMI was much stronger. Of 517 children classified as having high fat via WHtR, 70% were BMI-overweight, 24% had BMI-obesity, and only 6% had normal BMI.
Similar patterns were observed as the children aged into adolescence and young adulthood, with BMI consistently overestimating overweight prevalence by 2.3 to 2.8-fold compared to WHtR measurements.
Clinical implications and diabetes risk
The researchers validated their WHtR cutpoints by examining their ability to predict type 2 diabetes risk in 3,329 US adults. They found that WHtR measurements classified as “high fat” were associated with higher odds of prediabetes, while “excess fat” measurements predicted significantly higher odds of type 2 diabetes.
“Waist-to-height ratio is an affordable and universally accessible, accurate and precise tool for detecting high and excess fat in children and adolescents,” says Andrew Agbaje, who is a physician and associate professor of Clinical Epidemiology and Child Health at the University of Eastern Finland. His earlier research has also shown that waist-to-height ratio predicts both total body adiposity and central adiposity.
Limitations of BMI for childhood obesity assessment
The findings highlight significant limitations of using BMI alone for assessing childhood obesity. While dual-energy X-ray absorptiometry (DEXA) is considered the gold standard for accurate body fat measurement, it is expensive and not universally available in primary healthcare settings.
In the paper’s discussion, the authors note: “BMI is a poor diagnostic tool in the paediatric population, since it could falsely lead to the initiation of unnecessary interventions while excluding those who are truly in need of urgent care.”
This research builds on growing concerns about BMI limitations in adults as well. The Lancet Commission on Obesity and the European Association for the Study of Obesity (EASO) have recently recommended that obesity in adults should not be diagnosed with BMI alone but confirmed with additional measures, such as waist-to-height ratio.
New nomenclature proposed for childhood obesity
The study authors propose a significant shift in how childhood obesity is classified and described. Instead of using BMI-oriented terms like “overweight” and “obesity,” they suggest adopting WHtR-based classifications such as “adiposopathy grade 1” (for high fat) and “adiposopathy grade 2” (for excess fat).
The authors argue this change in nomenclature could significantly impact how children with excess adiposity are treated, potentially reducing stigma while improving clinical care.
“Overweight and obesity in children and adolescents should not be diagnosed with BMI alone but could be confirmed with waist-to-height ratio where a DEXA scan is lacking,” Agbaje concludes.
The research was supported by numerous foundations, including the Jenny and Antti Wihuri Foundation, the Finnish Cultural Foundation, the Orion Research Foundation, and the Novo Nordisk Foundation, among others.
Reference
Agbaje, A. O. (2025). BMI triples overweight prevalence in 7600 children compared with waist-to-height ratio: The ALSPAC Study. Obesity and Endocrinology, wjaf002. https://doi.org/10.1093/obendo/wjaf002
