Health Isn’t a BMI Number

For decades, we’ve let a simple height-weight calculation determine who is “healthy” and who isn’t. The problem? It was never designed for that—and it’s been getting it wrong all along.

When Starre Vartan walked into her pre-employment medical, she felt great. She was an avid trail runner and swimmer, strong and fit, excited to start work as a field geologist. Then the doctor glanced at her chart.

“Lose 10 to 15 pounds,” he said. The proof was her body mass index (BMI), which placed her squarely in the “overweight” category.

Never mind that her heart was strong. Never mind that her blood markers were pristine. Never mind that her disease risk sat comfortably below average. The number said she was overweight, so she must be unhealthy.

Vartan spent her twenties chasing a “healthy” BMI through disordered eating. Only later did she discover the truth: the doctor wasn’t wrong about her chart. He was wrong about BMI.

A population tool, misapplied

BMI was never meant for this. Created in the 1830s by Belgian mathematician Adolphe Quetelet, it was designed as a statistical tool for populations—a quick way to document average height and weight across groups of people. That’s it.

Fast forward to the 1990s. Obesity rates were rising, the World Health Organisation had declared a global epidemic, and clinicians wanted a fast, cheap way to screen individuals. BMI was convenient. It spread. And despite early critiques, it became medicine’s default gatekeeper.

Today, that single number determines who qualifies for knee surgeries, GLP-1 medications like Wegovy and Mounjaro, infertility treatment, gender-affirming care and bariatric procedures. People outside the “acceptable” range may be denied care. High-risk patients with “normal” BMIs may be overlooked entirely.

“There is no logic, no medical coherence to using BMI to define a disease,” says Francesco Rubino at King’s College London. “It’s just not suitable.”

What BMI misses

The fundamental flaw is simple: BMI cannot distinguish between fat and muscle. That’s why athletes routinely land in the “overweight” category. The very thing that makes them fit—muscle mass—also makes them look “unhealthy” on paper.

It also cannot show where fat is stored. Visceral fat—the kind wrapped around organs in the belly—is genuinely dangerous, linked to heart disease, diabetes and dementia. Subcutaneous fat on the arms, hips and thighs carries far less risk. BMI lumps them together.

Age, sex and ethnicity matter too. Women typically store fat differently than men. South Asian, Chinese and Black populations develop diabetes and heart disease at significantly lower BMIs than white Europeans. A tool developed on 19th-century European bodies cannot account for this.

“BMI is a good metric of volume—if a body is a cylinder,” says Diana Thomas, a mathematician at the United States Military Academy. “But I’m no cylinder. I have hips, I have curves.”

The shift away from BMI

In 2024, the Lancet Global Commission on obesity delivered a landmark report. Its conclusion: BMI alone should never be used to diagnose obesity. More than 75 international medical organisations immediately signed on.

The commission proposed a new framework. Just as there is pre-diabetes and diabetes, there should be “pre-clinical obesity” (when excess fat is a risk factor) and “clinical obesity” (when it becomes a stand-alone illness). And diagnosis must consider more than a number on a scale.

What works better

Researchers have identified several alternatives that actually capture health risk:

Waist-to-hip ratio divides waist circumference by hip circumference. It predicts heart attacks better than BMI and is a stronger indicator of mortality.

Weight-adjusted waist index (WWI) divides waist circumference by the square root of body weight. It captures central fat storage—the kind linked to inflammation, high blood pressure and metabolic disease.

Body roundness index (BRI) uses height, waist and weight to model body shape. Studies show it predicts visceral fat more accurately than any single measurement.

For Vartan, these new metrics tell a different story. Her WWI places her in the healthiest category. Her waist-to-hip ratio says “low risk.” Her BRI shows “excellent” visceral fat levels.

Quite a different message than her BMI.

A more honest measure of health

Countries are finally adapting. Canada updated its guidelines in 2020. The UK’s NICE now encourages waist-to-height ratio in routine care. The American Medical Association declared BMI “imperfect” in 2023.

The shift isn’t about dismissing obesity-related disease. It’s about recognising that health is complex, bodies are individual, and no single number can capture either.

“Why does everything have to be easy?” asks Thomas. “My body’s complicated. Embrace that complexity.”

Health isn’t a BMI number. It never was.