Key takeaways:

• Area under the receiver operating curve was higher when BMI and waist circumference were used to diagnose obesity vs. The Lancet criteria.
• BMI alone produced the lowest area under the receiver operating curve.

The use of BMI or waist circumference to diagnose obesity may be more accurate than The Lancet Diabetes & Endocrinology commission’s definition for the disease, researchers reported in JAMA Network Open.

As Healio previously reported, a report published in The Lancet Diabetes & Endocrinology in 2025 proposed a new definition for obesity that stated healthcare professionals could diagnose the condition if a person has one measure of excess adiposity plus an elevated BMI, two measures of excess body fat without elevated BMI, or by using DXA scan or another direct body fat metric. In a diagnostic study utilizing data from the National Health and Nutrition Examination Survey, researchers found using BMI and waist circumference yielded a higher area under the receiver operative curve for diagnosing obesity than The Lancet commission’s definition.

“The overarching message is that more complicated does not necessarily mean more accurate,” Aayush Visaria, MD, MPH, instructor of medicine at Rutgers Robert Wood Johnson Medical School in New Brunswick, New Jersey, told Healio. “A simple approach that identifies obesity based on either an elevated BMI or an elevated waist circumference performed at least as well overall as the full Lancet definition for identifying excess total body fat, and it performed better for identifying visceral obesity.”

Comparing obesity definitions

Researchers collected NHANES data from 1,900 adults aged 20 to 59 years who had DXA scans during the 2017 to 2018 cycle. DXA criteria for obesity was defined as total body fat of 25% or higher for men and 35% or higher for women. Visceral obesity was diagnosed in men with a visceral fat area of at least 130 cm² or women with a visceral fat area of at least 100 cm². Researchers assessed five measures for diagnosing obesity:

• full Lancet definition;
• Lancet definition with waist-to-height ratio removed;
• BMI plus waist circumference;
• BMI alone; and
• waist circumference alone.

According to DXA scan measures, obesity prevalence in the study group was approximately 68%. The full Lancet definition for obesity had a sensitivity of 87.7% and specificity of 70.7%. When waist-to-height ratio was removed from the Lancet definition, sensitivity was 73.8% and specificity 88.4%. Using BMI and waist circumference was tied to a sensitivity of 74% and specificity of 89.5%. Waist circumference alone had a sensitivity of 71.4% and specificity of 91.6%. BMI alone had a sensitivity of 51.3% and specificity of 94.9%.

Area under the receiver operating characteristic curve (AUROC) was higher using BMI and waist circumference compared with the full Lancet definition (0.818 vs. 0.79; P < .001). When waist-to-height ratio was removed from the Lancet definition, AUROC increased to 0.818. Waist circumference alone was tied to AUROC of 0.815. BMI alone had the lowest AUROC of 0.73.

“BMI remains inexpensive and easy to obtain so it can still be used, but it should not be used in isolation,” Visaria said. “In our study, BMI alone had a sensitivity of approximately 51%, meaning it missed nearly half of adults who met the DXA body fat definition of obesity.”

For visceral obesity, using BMI and waist circumference performed better than the full Lancet definition (AUROC, 0.785 vs. 0.717).

Clinical implications

“For routine clinical care, our findings support using BMI together with waist circumference, with a person meeting the criteria for obesity if either measure is elevated,” Visaria said. “This approach provided the best overall balance between identifying people with excess adiposity and avoiding unnecessary diagnostic labeling.”

Visaria said studies in more diverse populations are needed to confirm the findings. Additionally, he said studies assessing other long-term outcomes with different obesity measures should be conducted.

“We need to improve implementation of waist circumference measurements in the clinic setting,” Visaria added. “Unfortunately, this is not well adopted despite evidence of its importance for many years.”

For more information:

Aayush Visaria, MD, MPH, is an instructor of medicine in the department of medicine at Rutgers Robert Wood Johnson Medical School; and core faculty at Rutgers Health-RWJBarnabas Health Center for Climate, Health, and Healthcare, Rutgers Institute for Health, Health Care Policy and Aging Research. Visaria can be reached at aayush.visaria@rutgers.edu.