The BMI chart is a rough screening tool, not a precise measure of health. It correlates reasonably well with body fat at a population level, with studies showing a correlation of 0.86 between BMI and body fat percentage measured by DEXA scans. But for any individual person, it can be significantly off, sometimes classifying healthy, muscular people as overweight or giving a clean bill of health to someone carrying dangerous fat around their organs. The American Medical Association formally recognized these limitations in 2023, recommending that BMI be used alongside other measures rather than as a standalone assessment.
What BMI Actually Measures
BMI is a simple formula: your weight in kilograms divided by your height in meters squared. That’s it. It doesn’t measure fat, muscle, bone density, or where your body stores weight. It was invented in the 1830s by a Belgian statistician named Adolphe Quetelet, who was interested in defining the “average man” across populations. He had no particular interest in studying obesity. The formula was designed to identify statistical patterns in large groups, not to diagnose anything in an individual.
This origin matters because BMI still works better as a population-level tool than a personal one. When researchers study thousands of people, BMI trends track meaningfully with body fat and health outcomes. But zoom in on one person, and the number can mislead in several important ways.
Where BMI Gets It Wrong
The biggest blind spot is body composition. BMI treats all weight the same, whether it comes from muscle, fat, bone, or water. Someone who strength trains regularly can easily land in the “overweight” category while having low body fat and excellent metabolic health. Research on children and adolescents has shown that those with a high BMI driven by lean muscle mass don’t have the elevated cholesterol, blood sugar, or insulin resistance you’d expect from their BMI category. Their muscle mass appears to be protective, something BMI completely ignores.
The reverse problem is just as concerning. You can have a “normal” BMI while carrying a dangerous amount of visceral fat, the type that wraps around your liver, kidneys, and intestines. Visceral fat drives high blood pressure, high cholesterol, and high blood sugar, which are the starting points for diabetes, heart disease, and stroke. Subcutaneous fat (the kind just under your skin) is far less harmful on its own. BMI makes no distinction between the two. A person with a healthy BMI who stores most of their fat around their organs faces real health risks that the chart won’t flag.
BMI Becomes Less Reliable With Age
Starting in your 30s, your body gradually loses skeletal muscle and replaces it with fat. By your 70s, you may have only half the muscle mass you had in your 20s. This process, called sarcopenia, makes BMI particularly misleading for older adults. Because muscle weighs more than fat, someone losing muscle and gaining fat can maintain the same weight and the same “healthy” BMI while becoming measurably less healthy. Johns Hopkins Medicine notes that BMI is often less accurate for adults over 65 precisely because it doesn’t reflect these age-related shifts in body composition.
This has practical consequences. An older adult with a normal BMI might be told their weight is fine when they’ve actually lost significant muscle, which increases their risk of falls, fractures, and loss of independence. Meanwhile, an older adult who builds muscle through exercise may see a higher number on the scale, but those are protective pounds.
Ethnicity Changes the Risk Thresholds
BMI cutoffs were developed primarily from data on white European populations, and they don’t apply equally across ethnic groups. Research published through the American College of Cardiology found that the diabetes risk associated with a BMI of 30 in white adults kicks in at significantly lower BMIs for other populations: 28.1 for Black adults, 26.9 for Chinese adults, 26.6 for Arab adults, and 23.9 for South Asian adults.
The gap is even more striking at the “overweight” threshold. A BMI of 25 in white adults corresponds to equivalent diabetes risk at 23.4 for Black adults, 22.2 for Chinese adults, and just 19.2 for South Asian adults. That means a South Asian person with a BMI of 20, which the standard chart labels as solidly “normal,” may already face the same metabolic risk as a white person at the overweight cutoff. Using a single chart for everyone misses these differences entirely.
The “Metabolically Healthy Obese” Problem
About 20% of people classified as overweight or obese by BMI have completely normal metabolic markers: healthy blood sugar, healthy cholesterol, healthy blood pressure. These individuals fall into what researchers call “metabolically healthy obesity.” Their BMI says they’re at risk, but their actual metabolic profile doesn’t confirm it. This doesn’t mean excess weight is harmless for them long-term, but it does illustrate that BMI alone gives an incomplete picture of someone’s current health status.
Better Ways to Assess Body Fat Risk
Two alternatives capture what BMI misses, and both are things you can measure at home with a tape measure.
Waist-to-height ratio divides your waist circumference by your height. A ratio above 0.5 to 0.55 signals increased cardiovascular risk in younger and middle-aged adults, while a cutoff of 0.6 or higher is more appropriate for adults over 60. In research comparing different body measurements, waist-to-height ratio was the only one that independently predicted calcium buildup in coronary arteries after accounting for other risk factors. It also identified heart failure risk in older adults that BMI-based screening missed entirely.
The Body Roundness Index (BRI) is a newer metric that uses your height, waist size, and hip size, leaving weight out of the equation entirely. By focusing on body shape rather than body weight, it better estimates how much visceral fat surrounds your organs. A 20-year study of about 33,000 people found that those with a BRI of 6.9 or higher had nearly 50% greater risk of early death compared to those in the middle range. Interestingly, adults 65 and older with very low BRIs also had higher mortality risk, suggesting that too little body fat is also dangerous in older age.
What BMI Is Still Good For
None of this means BMI is useless. The AMA’s 2023 policy still recognizes it as a reasonable initial screener for metabolic health risks, particularly at higher levels. A BMI above 35 continues to be a meaningful risk marker regardless of body composition. And at a population level, the 0.86 correlation with body fat percentage means it tracks real patterns, just not with enough precision to tell any one person how healthy they are.
The most practical approach is to treat BMI as one data point among several. Pair it with a waist measurement, pay attention to your blood pressure and blood sugar numbers, and consider your muscle mass and activity level. A single number calculated from your height and weight was never designed to capture the full complexity of metabolic health, and expecting it to do so sets up both false alarms and false reassurance.