BMI has been used as a clinical standard for decades. It probably shouldn’t have been.
The measure is simple: divide weight by height squared, get a number, assign a category. Underweight, normal, overweight, obese. Developed by a 19th-century mathematician as a statistical tool for tracking population averages, BMI was never designed to assess the health of any individual person. Yet somewhere along the way — particularly after the World Health Organization declared obesity a global epidemic in 1997 — it became medicine’s default screening gate. Fast, cheap, and easy to administer, it filled a role that clinicians needed filled. Whether it was actually suited to that role is a different question.
The answer, increasingly, is no.
Francesco Rubino at King’s College London puts it plainly: “There is no logic, no medical coherence to using BMI to define a disease. It’s just not suitable.” After roughly 30 years as the clinical status quo, a genuine consensus is forming around that view. The metric distorts health assessments for a significant portion of the population, and the consequences are concrete. BMI thresholds currently determine access to knee surgeries, GLP-1 weight-loss medications, infertility treatments, bariatric procedures, and gender-affirming care. People outside the acceptable range may be denied treatment. High-risk patients with normal BMI scores may be overlooked entirely.
The core problem is what BMI cannot see. It measures volume, not composition. It makes no distinction between fat and muscle, which means a well-trained athlete often registers as overweight. It ignores where fat is stored — a critical oversight, since visceral fat packed around internal organs carries substantially different health risks than subcutaneous fat distributed across the arms, thighs, or hips. Men and women tend to store fat differently, yet BMI applies uniform cutoffs to both. The number flattens all of that complexity into a single figure and then draws hard clinical lines around it.
There is also a population-versus-individual problem baked into the measure’s origins. A tool designed to track statistical trends across large groups behaves very differently when applied to one person sitting in a doctor’s office. Population-level correlations between BMI and disease risk say almost nothing reliable about any specific individual’s metabolic health, cardiovascular fitness, or actual fat distribution.
What makes this more than an academic debate is its real-world weight. People have been denied medical procedures, pushed toward disordered eating, or simply received wrong information about their health — all on the basis of a number that was capturing something else entirely. Someone can fall within the normal BMI range while carrying dangerous levels of visceral fat or lacking sufficient body fat to sustain regular menstruation, with the cascading health consequences that follow. The number tells you neither story.
The search for better alternatives is now underway in earnest. Researchers are exploring measures that account for waist circumference, body fat percentage, and fat distribution — metrics that require slightly more effort but offer substantially more clinical signal. None has yet achieved the institutional momentum that BMI built up over decades of entrenchment.
What this moment is really forcing is a harder question: what does a healthy weight actually mean, and can any single number capture it? The evidence suggests the answer to the second part is almost certainly no. A person’s metabolic health, physical fitness, and disease risk are shaped by factors that resist reduction to one tidy figure. BMI has been useful as a blunt population instrument. As a tool for individual clinical decisions, its limitations have become too well documented to ignore.
Photo by Debby Hudson on Unsplash
This article is a curated summary based on third-party sources. Source: Read the original article
