LONDON – The Body Mass Index (BMI), a 200-year-old mathematical formula that has long served as the default metric for categorizing weight status, is facing its most significant professional challenge in decades. As health systems worldwide pivot toward personalized medicine, a growing coalition of endocrinologists, public health researchers, and digital health companies is questioning whether BMI alone can reliably predict cardiometabolic risk—or whether it should be relegated to a secondary, contextual role.

The debate reached a new inflection point this week with the release of two large-scale studies and a position paper from the European Association for the Study of Obesity (EASO), all of which argue that BMI’s inherent limitations—particularly its failure to distinguish muscle mass from fat mass, and its inability to capture fat distribution—are now materially affecting patient outcomes.

New Data Challenges BMI’s Predictive Power

The first study, published inThe Lancet Regional Health – Europe, analyzed electronic health records from 1.2 million adults across five European countries over a 12-year period. Researchers found that among individuals classified as “normal weight” by BMI (18.5–24.9 kg/m²), nearly 27% had metabolic abnormalities—including elevated fasting glucose, dyslipidemia, or hypertension—that significantly increased their risk of type 2 diabetes and cardiovascular events. Conversely, 18% of those categorized as “obese” (BMI ≥30) showed no metabolic abnormalities and had a mortality risk comparable to that of metabolically healthy normal-weight individuals.

“We are not saying BMI is useless,” said Dr. Helena Voss, lead author of the Lancet study and a senior epidemiologist at the University of Copenhagen. “But as a standalone screening tool, it misclassifies roughly one in four individuals. In clinical practice, that means we are either over-treating healthy people or under-treating those who need intervention—both of which carry real costs.”

The second study, presented at the American Heart Association’s Epidemiology and Prevention Conference in Chicago, used dual-energy X-ray absorptiometry (DEXA) scans on 38,000 U.S. adults to compare BMI against direct measures of body fat percentage and visceral adipose tissue. The results showed that BMI explained only 62% of the variance in actual body fat percentage in women and 58% in men. More critically, among individuals with a BMI in the “overweight” range (25–29.9), 41% of women and 33% of men had a normal body fat percentage—contradicting the traditional assumption that overweight BMI always implies excess adiposity.

EASO Position Paper Calls for ‘BMI Plus’ Framework

In response to these findings, the EASO released a position paper titled“Beyond BMI: A Clinical Framework for Diagnosing and Managing Adiposity-Based Chronic Disease.”The paper recommends that clinicians no longer use BMI as a definitive diagnostic threshold but rather as a first-step screening parameter. For patients whose BMI falls in the “overweight” or “obesity” range—or who have a normal BMI but a family history of metabolic disease—the paper urges additional assessment using waist circumference, waist-to-hip ratio, and, where available, direct body composition analysis via bioelectrical impedance or DEXA.

“BMI remains a useful population-level surveillance tool, but it is not a patient-level diagnostic,” said Dr. Francesco Rubino, chair of metabolic and bariatric surgery at King’s College London and a co-author of the EASO paper. “We now understand that the primary driver of morbidity is not total body weight but the distribution and function of adipose tissue—particularly visceral fat, which is metabolically active and inflammatory. A person with a BMI of 32 and high visceral fat is at far greater risk than someone with a BMI of 32 and predominantly subcutaneous fat.”

Rubino’s comments reflect a broader shift toward the concept of “adiposity-based chronic disease” (ABCD), a term first proposed in 2016 and now increasingly adopted by major medical societies. The ABCD model explicitly separates theamountof body fat (where BMI is useful) from thefunctionandlocationof that fat (where BMI is insufficient).

Industry Response: From Wearables to Clinical Algorithms

The recalibration of BMI is not confined to academic journals. The digital health sector, which has long used BMI as a core input for risk calculators and wellness coaching algorithms, is rapidly adapting.

Omada Health, a leading virtual chronic care platform, announced this week that it will replace BMI-based eligibility criteria for its diabetes prevention program with a composite risk score that incorporates waist circumference, fasting glucose, and blood pressure. The company’s chief medical officer, Dr. Samantha Chen, stated that the change was driven by user feedback and clinical outcomes data. “Our members with normal BMI but high waist-to-hip ratio were consistently failing to qualify for the program, even though their biometrics clearly showed they were at risk,” Chen said. “We realized we were serving the metric, not the patient.”

Similarly, the global medical device company Smart Scales unveiled a new consumer scale at the Consumer Electronics Show in January that uses multi-frequency bioelectrical impedance to estimate visceral fat level and muscle mass, not just total weight. The device’s companion app now displays a “metabolic health score” that downplays BMI in favor of a composite of body composition, heart rate variability, and activity data.

On the pharmaceutical side, the shift has regulatory implications. The U.S. Food and Drug Administration (FDA) currently uses BMI thresholds for approving weight-management drugs—typically requiring a BMI of ≥30 or ≥27 with comorbidity. However, in a draft guidance issued in late 2024, the FDA signaled that it may consider alternative endpoints, such as reduction in waist circumference or improvement in metabolic biomarkers, for future drug approvals. The guidance is still under public comment, but industry analysts expect it to be finalized within 18 months.

Expert Viewpoints: A Tool, Not a Verdict

Dr. Louise Baur, a pediatric obesity researcher at the University of Sydney and former president of the World Obesity Federation, cautions against overcorrecting. “BMI is not the enemy. It is a crude but inexpensive metric that works well at the population level for tracking trends. The problem arises when we treat it as a precise measure of an individual’s health. We do not discard thermometers because they don’t measure blood pressure.”

Baur’s sentiment is echoed by Dr. Priya Sumithran, an endocrinologist at the University of Melbourne. She notes that BMI’s simplicity has made it a powerful advocacy tool for destigmatizing obesity as a disease. “If we abandon BMI entirely, we risk losing a common language that has helped secure insurance coverage and public policy support for obesity treatments. The solution is not to eliminate BMI but to augment it—to teach clinicians that BMI is a starting point, not a final answer.”

However, some experts argue that the momentum toward “BMI plus” is moving too slowly. Dr. Fatima Cody Stanford, an obesity medicine physician at Massachusetts General Hospital and Harvard Medical School, points out that racial and ethnic differences in body composition further complicate BMI’s utility. “At the same BMI, Asian populations tend to have higher body fat percentage and higher cardiometabolic risk, while Black populations tend to have lower body fat and lower risk. We have known this for decades, but most clinical guidelines still use universal BMI cutoffs. That is an equity issue.”

Stanford advocates for the adoption of ethnicity-specific BMI thresholds, a move already implemented in Japan, China, and India, where lower cutoffs for overweight (BMI ≥23 or ≥24) are used. She also calls for increased reimbursement for body composition analysis, which remains out of pocket for many patients.

The Road Ahead: Pragmatic Integration

As the evidence accumulates, the consensus emerging from this week’s developments is not that BMI should be discarded, but that it should be repositioned within a multi-dimensional assessment framework. The World Health Organization, which has used BMI for global obesity surveillance since the 1990s, has not yet revised its guidelines, but a spokesperson confirmed that a technical consultation on “anthropometric indicators beyond BMI” is scheduled for late 2026.

In the interim, health systems are experimenting with hybrid models. The National Health Service (NHS) in England, for example, has begun piloting a “BMI+5” pathway in primary care, where any patient with a BMI above 25 receives automatic screening for waist circumference and HbA1c (a blood sugar marker). Early data from the pilot, presented at a closed briefing last month, suggests that the approach identifies 22% more at-risk patients than BMI alone.

For the average consumer, the takeaway is increasingly clear: a single number on a scale—or a chart on a clinic wall—no longer tells the full story. The future of metabolic screening lies in integration: combining BMI with body composition, biochemical markers, and, increasingly, genetic and microbiome data.

As Dr. Rubino of King’s College London puts it, “We are moving from an era of ‘what is your number?’ to an era of ‘what is your pattern?’ The pattern is far more informative. And for that, we need more than a formula invented in the 1830s.”

The industry, it seems, is listening. Whether regulatory bodies and clinical guidelines will fully catch up remains the open question—but the direction of travel is unambiguous. BMI is not