BMI is a real measurement, but it’s a limited one. It tells you something about health risk at the population level, yet it can seriously mislead when applied to any single person. The formula is simple: your weight in kilograms divided by your height in meters squared. What that number actually means for your health, though, is far more complicated than the standard weight categories suggest.
Where BMI Came From
BMI was never designed to diagnose anything. In 1832, a Belgian statistician named Adolphe Quetelet created the formula as part of his project to define the characteristics of an “average man” using bell curves. He was interested in population-level statistics, not individual health.
The formula sat relatively unused for over a century until physiologist Ancel Keys revived it in 1972, officially coining the term “body mass index.” Keys studied over 7,400 men and promoted BMI as a convenient, easy-to-calculate tool for research. He was explicit that it was meant for analyzing large groups, not for clinical use on individual patients. Somewhere along the way, that distinction got lost. Doctors’ offices adopted it as a quick screening number, insurance companies used it to set premiums, and it became embedded in healthcare in ways its creators never intended.
What BMI Actually Measures
BMI captures one thing: the ratio of your weight to your height. It cannot distinguish between muscle, bone, fat, or water. This creates obvious problems. A muscular athlete and a sedentary person of the same height and weight get the same BMI score, despite having wildly different body compositions and health profiles. As Harvard Health Publishing notes, muscle and bone are denser than fat, so BMI overestimates body fat in people with high muscle mass and underestimates it in older adults who have lost bone density and muscle.
The current categories, set by the CDC, classify adults as underweight (below 18.5), healthy weight (18.5 to 24.9), overweight (25 to 29.9), or obese (30 and above). These cutoff points are treated as universal, but the evidence shows they shouldn’t be.
The Ethnic Bias Built Into the Scale
BMI thresholds were developed primarily from data on white European populations, and they don’t translate cleanly across ethnic groups. A large study published in Diabetes Care calculated the BMI at which different populations develop diabetes at the same rate as white adults at a BMI of 30. The results were striking: South Asian individuals hit that same diabetes risk at a BMI of just 24, Chinese individuals at 25, and Black individuals at 26. All of these fall within what the standard BMI chart labels “healthy” or “overweight,” meaning the current categories can give a false sense of security to millions of people.
This isn’t a small discrepancy. A South Asian person with a BMI of 25 may face diabetes risk comparable to a white person classified as obese, yet their BMI would place them in the “healthy weight” range. For populations where fat tends to accumulate around the organs rather than under the skin, a single weight-to-height ratio misses the danger entirely.
Why Fat Location Matters More Than Fat Amount
Not all body fat carries the same risk. Visceral fat, the type that wraps around your liver, intestines, and other organs, is far more metabolically dangerous than the subcutaneous fat stored just beneath your skin. Visceral fat breaks down more readily into fatty acids that flood the liver, reducing insulin sensitivity and promoting the kind of cholesterol profile that leads to heart disease.
BMI does correlate with total body fat, but its correlation with visceral fat specifically is weaker. MRI-based research in young adults found that BMI correlated strongly with overall abdominal fat (r = 0.824) and subcutaneous fat (r = 0.768), but the correlation dropped noticeably for visceral fat (r = 0.633). That gap matters because visceral fat is the type most closely linked to metabolic disease. Two people with identical BMIs can have very different amounts of visceral fat, and very different health trajectories as a result.
Normal Weight Doesn’t Always Mean Healthy
One of the most striking challenges to BMI comes from research on metabolic syndrome, a cluster of conditions including high blood pressure, high blood sugar, excess abdominal fat, and abnormal cholesterol levels. A CDC analysis using national health survey data found that normal-weight adults with metabolic syndrome had a 45% mortality rate over a 150-month follow-up period. Obese adults without metabolic syndrome had just an 8.7% mortality rate over the same period.
The cardiovascular numbers tell the same story. Compared to normal-weight adults without metabolic syndrome, normal-weight adults with it had more than double the risk of dying from heart disease (a hazard ratio of 2.12). Meanwhile, obese adults without metabolic syndrome had a hazard ratio of just 0.71, meaning they were actually less likely to die from cardiovascular causes than the normal-weight reference group. In other words, a “healthy” BMI combined with poor metabolic health is more dangerous than an “obese” BMI with good metabolic health. Relying on BMI alone would flag the wrong person.
What Medical Organizations Say Now
The medical establishment has started to formally acknowledge these limitations. In 2023, the American Medical Association adopted a policy stating that BMI is “an imperfect clinical measure” and should not be used as a sole criterion for medical decisions or insurance reimbursement. The AMA recommended using BMI alongside other indicators: visceral fat measurements, waist circumference, body composition analysis, relative fat mass, and genetic or metabolic factors. The policy also flagged BMI as particularly problematic in the context of eating disorders, where it fails to capture the full spectrum of disordered eating.
This doesn’t mean BMI is useless. At scale, across thousands of people, higher BMI does correlate with higher rates of heart disease, diabetes, and certain cancers. It works as a rough screening flag, a reason to look deeper. The problem is when it becomes the final word rather than the first question.
Measures That Predict Health Better
If BMI is a blunt tool, what’s sharper? Waist circumference is one straightforward option. It approximates abdominal fat more directly than a weight-to-height ratio and requires nothing more than a tape measure. But the strongest evidence points to waist-to-hip ratio, which compares the circumference of your waist to your hips.
A large analysis published in JAMA Network Open found that waist-to-hip ratio was linked more strongly and more consistently to death from any cause than either BMI or fat mass index. Using a genetic analysis method called mendelian randomization, the researchers also found evidence that the relationship between waist-to-hip ratio and mortality is likely causal, not just a correlation. This makes sense biologically: waist-to-hip ratio captures where your body stores fat, which matters more than how much you weigh overall.
You can measure your own waist-to-hip ratio at home. Wrap a tape measure around the narrowest part of your waist (usually just above the belly button) and the widest part of your hips. Divide the waist number by the hip number. For women, a ratio above 0.85 is generally associated with increased health risk. For men, the threshold is about 0.90.
What BMI Can and Can’t Tell You
BMI is real in the sense that it’s a real number derived from a real formula. It is not an invention or a scam. But it was designed for statisticians studying populations, not for doctors treating patients, and it shows. It can’t tell you how much of your weight is muscle versus fat. It can’t tell you where your fat is stored. It can’t account for your ethnicity, your age, your fitness level, or your metabolic health. A person with a BMI of 27 might be in excellent health, and a person with a BMI of 22 might be on a path toward heart disease.
If your doctor has flagged your BMI, it’s worth treating that as a starting point for a broader conversation, not a diagnosis. Ask about your waist circumference, your blood pressure, your blood sugar, and your cholesterol. Those numbers, taken together, paint a picture that BMI alone never could.