Take a look at any of the evidence-based US obesity treatment guidelines. The key criteria for diagnosing overweight and obesity is based on the body mass index (BMI).
The guidelines also use BMI to stratify care options to decrease cardiovascular risk. For example, persons with BMI ≥30 are classified as having obesity, and antiobesity medications are recommended. Those with BMI ≥ 40 are classified as having severe obesity, and metabolic bariatric surgery may be appropriate.
But where did these cutoff points for more and less aggressive treatments come from? These BMI cutoffs are based primarily on mortality data collected from large non-Hispanic White populations, without data on potential differences by gender and ethnicity.
For example, it is certainly true that those with BMI ≥ 30 have more cardiovascular risk factors than those with BMI < 30. But Asian American individuals have more risk factors at lower BMIs than do White or African American individuals likely because of more visceral fat accumulation at lower BMIs.
Besides the variation in gender and ethnicity, BMI does not take the type and location of body fat into consideration. Adipose tissue in visceral or ectopic areas have much higher risks for disease than subcutaneous adipose tissue because of the associated inflammation. Measures such as waist circumference, waist-to-hip ratio, and skinfold measurements aim to capture this aspect but often fall short because of variation in techniques.
BMI does not account for muscle mass either, so fit athletes and bodybuilders can be classified as having obesity by BMI alone. More accurate body fat percent measures, such as dual-energy X-ray absorptiometry or MRI specifically for ectopic fat, are labor intensive, expensive, and not feasible to perform in a busy primary care or endocrinology clinic.
Assessing Risks From Obesity Beyond BMI
Clearly, better risk measures than BMI are needed, but until they are available, supplemental clinical tools can aid diagnosis and treatment decisions at obesity medicine specialty centers, endocrinology and diabetes centers, and those centers that focus on the treatment of obesity.
For example, a seca scale can measure percent body fat by bioelectric impedance analysis. This technique also has its limitations, but for persons who are well hydrated, it can be used as a baseline to determine efficacy of behavioral interventions, such as resistance-exercise training and a high-protein diet to protect muscle mass as the patient loses weight.
A lot also can be gleaned from diet and exercise history, social history, family history, and physical exam as well as laboratory analyses. For example, an Asian American patient with a BMI of 26 who has been gaining weight mostly in the abdominal region after age 35 years is likely to have cardiometabolic risk, and a family history can solidify that. An exam can show signs of acanthosis nigricans or an enlarged liver and generous abdominal adipose tissue. This would be the patient in whom you would want to obtain a hemoglobin A1c measurement in the chance that it is elevated at > 5.7 mg/dL, suggesting high risk for type 2 diabetes.
A Fibrosis-4 score can assess the risk for liver disease from aspartate transaminase and alanine aminotransferase and platelet count and age, providing clues to cardiometabolic disease risk.
In the next 10, years there may be a better measure for cardiometabolic risk that is more accurate than BMI is. It could be the sagittal abdominal diameter, which has been purported to more accurately measure visceral abdominal fat. But this has not made it to be one of the vital signs in a busy primary care clinic, however.