New biological BMI measures offer a more accurate representation of metabolic health

Researchers at the Institute for Systems Biology (ISB) have developed measures of biological body mass index (BMI) that provide a more accurate representation of metabolic health and are more versatile, informative and actionable than the traditional BMI equation used by long ago. The work was published in the journal Nature Medicine.

For decades, doctors have relied on BMI as a crude tool to classify individuals as underweight, normal weight, overweight, or obese. BMI results are calculated by dividing a person’s weight in kilograms by height in meters squared. About 30 percent of the population is misclassified by this approach. Despite its limitations, BMI continues to be profound and widely accepted in the clinic, as it is a major risk factor for a variety of chronic diseases, including diabetes, cardiovascular disease, and cancer.

For years, BMI has been the primary measurement for doctors to classify individuals based on their height and weight relative to the average person. However, this average person does not really exist. We now have the ability to use advanced molecular measurements as a more comprehensive representation of a person’s metabolic health, which can be used to make more accurate clinical recommendations for individuals.”

Noa Rappaport, PhD, ISB senior scientist and corresponding author of the paper

Rappaport and colleagues studied 1,000 individuals enrolled in a wellness program by performing multi-omic profiling, examining more than 1,100 blood tests such as proteins and metabolites, as well as genetic risk scores and gut microbiome composition collected at different time points. The researchers then created machine learning models that led to more accurate predictive variations of a biological BMI than traditional BMI measurements alone.

The team made several important findings, including:

• Those with a high biological BMI and traditional normal BMI were less healthy but able to lose weight more easily after a lifestyle intervention.
• Those classified as obese with a traditional BMI but with a biologically normal BMI were biologically healthier and had a harder time losing weight.
• When people made positive lifestyle changes, biological BMI was more responsive and fell earlier than traditional BMI.

With positive lifestyle changes, the findings suggest that even if someone isn’t losing weight, they can become biologically healthier.

This work is a valuable asset for understanding the molecular changes associated with obesity and metabolic health, and has the potential to significantly improve the development of predictive and preventive clinical approaches to treat metabolic disorders.”

Kengo Watanabe, PhD, lead author of the study and K. Carole Ellison Fellow in Bioinformatics.

Added Rappaport: “We have demonstrated the value of multi-omic profiling to uncover important insights into the complex relationships between obesity, metabolic health and chronic disease, and highlighted the need to consider a range of factors beyond traditional BMI measures in understanding and addressing these issues”.