BMI, or body mass index, is a simple formula that divides your weight by your height squared. It’s easy to calculate and widely used, but it has serious limitations that can lead to misclassification of your health status. The American Medical Association acknowledged these shortcomings in a 2023 policy statement, noting that BMI “loses predictability when applied on the individual level” and recommending it be used alongside other measures rather than on its own.
BMI Was Never Designed to Measure Health
The formula we call BMI was created in the 1800s by Adolphe Quetelet, a Belgian mathematician and astronomer. Quetelet had no interest in obesity. He was trying to define the characteristics of the statistically “average man” by applying probability math to large populations. His goal was to show that human physical traits follow a bell curve distribution, the same pattern found throughout nature. The index was a population-level statistical tool, not a diagnostic one, and repurposing it as a personal health metric stretches it well beyond its original design.
It Can’t Tell Muscle From Fat
BMI treats every pound the same, whether it comes from muscle, fat, bone, or water. A 200-pound person who is 5’10” gets the same BMI regardless of whether that weight is largely lean muscle or largely body fat. This is a particularly well-documented problem among athletes, military personnel, and anyone who carries significant muscle mass. Research on Czech military personnel found that BMI routinely misclassified physically fit soldiers as overweight or obese, even when their actual body fat levels were healthy.
The reverse problem is just as concerning. People with relatively low muscle mass can register a “normal” BMI while carrying an unhealthy amount of body fat. When researchers compared BMI classifications against DXA scans (which directly measure body composition), 22% of people in the “normal weight” BMI range were actually assigned a different category. About 11% of those were reclassified as overweight, and nearly 1% qualified as obese based on their actual fat levels.
Where Fat Sits Matters More Than How Much You Weigh
BMI gives you a single number. It tells you nothing about where your body stores fat, and that distinction carries real health consequences. Visceral fat, the kind that wraps around your organs deep inside the abdomen, is far more metabolically dangerous than the subcutaneous fat you can pinch under your skin. Visceral fat is strongly linked to cardiovascular disease, insulin resistance, and type 2 diabetes. Two people with identical BMIs can have dramatically different levels of visceral fat, and therefore dramatically different health risks.
You can’t see or feel visceral fat, which makes BMI even less useful here. A tape measure around your waist actually gives you more relevant information. Waist-to-height ratio, in particular, has outperformed BMI in predicting conditions like hypertension. In one prospective study of patients with type 2 diabetes, waist-to-height ratio had a discriminative power of 0.609 for predicting high blood pressure, compared to just 0.571 for BMI. That may sound like a small difference, but it reflects a meaningful gap in how accurately each measure identifies people at risk.
The Standard Cutoffs Don’t Work Across Ethnicities
BMI categories (underweight below 18.5, normal weight 18.5 to 25, overweight 25 to 30, obese above 30) were developed primarily from data on non-Hispanic White populations. These thresholds don’t translate cleanly to other groups because body frame size, fat distribution, and metabolic risk vary significantly across ethnicities.
The numbers are striking. A study examining diabetes risk across ethnic groups in England found that the same level of risk associated with a BMI of 30 in White adults appeared at a BMI of just 23.9 in South Asian adults. For the “overweight” threshold, a BMI of 25 in White populations corresponded to a BMI of 19.2 in South Asian populations. South Asian people tend to have smaller body frames and carry proportionally more visceral fat, which means they can develop metabolic disease at BMIs that would be classified as perfectly normal. Black, Chinese, and Arab populations also showed meaningful shifts, with equivalent risk cutoffs of 28.1, 26.9, and 26.6, respectively, compared to the standard obesity threshold of 30.
Using universal cutoffs means that people from some ethnic backgrounds get flagged too late, after metabolic damage has already begun, while others may be unnecessarily categorized as at risk.
BMI Becomes Less Reliable as You Age
Aging changes your body composition in ways that BMI completely ignores. After about age 50, most people gradually lose muscle mass and bone density while gaining fat, particularly visceral fat. Your weight might stay the same or even drop slightly, keeping your BMI in the “normal” range, while your actual body composition shifts in an unhealthy direction.
There’s also a well-documented phenomenon called the “obesity paradox.” In adults 65 and older, being slightly overweight by BMI standards is actually associated with lower cardiovascular mortality. An analysis of National Health Interview Survey data spanning over two decades found that cardiovascular death risk was lowest at a BMI of 26 in adults over 65, compared to 22 in younger adults. Older adults classified as overweight had a 7% lower risk of cardiovascular death than those with a “normal” BMI. Even those with moderate obesity showed no increased risk. This suggests that the optimal BMI shifts upward with age, yet the standard cutoffs remain the same regardless of whether you’re 30 or 75.
What Works Better
The AMA’s 2023 policy recommends supplementing BMI with more informative measures. Several options give a clearer picture of metabolic health:
- Waist circumference captures abdominal fat accumulation, which is more closely tied to disease risk than overall weight.
- Waist-to-height ratio adjusts for body size and has shown stronger predictive power for hypertension and cardiovascular risk than BMI. A general guideline is to keep your waist circumference below half your height.
- Body composition testing (such as DXA scans or bioelectrical impedance) directly measures fat mass versus lean mass, eliminating the muscle-versus-fat confusion that plagues BMI.
- Metabolic markers like blood sugar, blood pressure, cholesterol, and inflammatory markers tell you what’s actually happening inside your body, regardless of what the scale says.
BMI still has a role as a quick, inexpensive screening tool for tracking obesity trends across large populations. The AMA acknowledges it remains useful as an initial screener and notes that a BMI above 35 still carries clear risk. But for understanding your individual health, it’s one data point among many, and often not the most informative one.