What BMI actually measures and what it does not
BMI is a ratio of weight to height squared. That is the entirety of what it calculates. It has no way of knowing whether the weight behind that number comes from muscle, fat, bone density, water, or any combination of these. This single limitation is responsible for most of the situations in which BMI produces a result that feels inconsistent with how someone actually looks, feels, or performs, because body weight and body composition are related but they are not the same thing.
The formula was developed in the 19th century as a statistical tool for describing population averages, not for assessing individual health. When it was adopted by healthcare systems decades later as a screening metric, its simplicity made it attractive, but that simplicity comes with trade-offs that become significant in specific groups of people.
Where BMI consistently gets it wrong
The most well-documented failure case is muscular individuals. Muscle tissue is denser than fat, so someone who has built significant lean mass through training will weigh more at a given height than the population averages BMI was calibrated against. The formula interprets that additional weight as a sign of excess fat and moves them into the overweight or obese category, even when their actual body fat percentage is low and their metabolic health is excellent. This is not an edge case, it affects recreational gym-goers, serious athletes, manual labourers, and anyone who has consistently built lean mass over time.
The opposite failure is equally important and arguably more dangerous from a health perspective. Someone can have a BMI that falls cleanly within the normal range while carrying a high proportion of body fat relative to lean mass, particularly visceral fat stored around the abdominal organs. This pattern, sometimes described as being metabolically obese at a normal weight, is associated with insulin resistance, elevated triglycerides, and cardiovascular risk in the same way that high-fat-mass obesity is, but BMI would give no indication that anything is worth investigating. The person looks fine on paper while carrying a meaningful and undetected metabolic risk.
Age introduces another consistent distortion. As people get older, body composition naturally shifts even when total weight stays stable, with muscle mass declining and fat mass increasing as part of normal ageing. This means the same BMI value represents quite different body compositions at 30 and at 65, and applying identical health interpretations to both ages produces misleading conclusions. Older adults who have lost muscle and gained fat can have a completely normal BMI while having a body composition that raises genuine health concerns.
Ethnic background creates further inconsistency. Research has consistently shown that people of Asian descent carry higher metabolic risk at lower BMI values than the standard classification thresholds suggest, largely because the categories were developed primarily from data on Western European populations. Several major health organisations now recommend lower BMI cut-off points for Asian populations to account for this difference, which acknowledges directly that the standard categories do not translate uniformly across all groups.
Why these limitations matter in practice
The practical consequence of these limitations is that BMI can generate false reassurance in people who need to pay attention, and unnecessary alarm in people who are genuinely healthy. Both outcomes have real costs. Someone with normal weight obesity who gets a clean BMI result may feel no motivation to investigate their body composition further, missing a metabolic risk that earlier intervention could address. A muscular athlete classified as overweight or obese may make decisions about their training, diet, or weight that are driven by a number that does not accurately represent their health status.
Using BMI as a sole criterion for clinical decisions, insurance assessments, or personal health goals amplifies these problems by treating a rough population-level screening tool as though it has individual diagnostic precision that it simply does not have.
What to use instead
Body fat percentage is the most direct replacement for the specific thing BMI is trying to estimate, which is whether excess fat is present in a quantity that affects health. By measuring what proportion of total body weight is actually fat tissue, it sidesteps the muscle-versus-fat confusion entirely and gives you a number that means the same thing regardless of whether you are an athlete, an older adult, or someone who has never set foot in a gym. The methods range from basic bioelectrical impedance available in home scales to DEXA scanning in clinical settings, with varying levels of precision depending on what you have access to.
Waist circumference adds something that body fat percentage alone cannot tell you, which is where fat is distributed. Visceral fat stored around the abdominal organs carries significantly higher metabolic risk than fat stored subcutaneously beneath the skin, and waist circumference correlates reasonably well with visceral fat accumulation without requiring specialist equipment. A measurement above 94 centimetres for men or 80 centimetres for women indicates elevated risk according to major health guidelines, and tracking it over time gives useful information about whether central fat is changing in response to lifestyle adjustments.
The ABSI, or A Body Shape Index, takes the waist circumference concept further by incorporating it into a formula that accounts for height and weight together, producing a shape-based health risk indicator that is less distorted by total mass and more sensitive to the body shape patterns most strongly associated with metabolic outcomes. For people whose BMI is misleading because of muscle mass or unusual fat distribution, ABSI often provides a more meaningful risk picture.
Waist-to-height ratio offers a simpler version of the same idea. The general guideline that your waist circumference should be less than half your height is easy to apply, requires no calculation beyond a tape measure, and research suggests it predicts cardiometabolic risk at least as well as BMI across different populations and age groups.
How to use BMI sensibly going forward
The most practical approach is to keep BMI as one data point while adding at least one or two measurements that capture what BMI cannot. Pairing BMI with body fat percentage tells you whether high weight reflects lean mass or fat mass. Adding waist circumference tells you whether fat is concentrated in the high-risk central area or distributed more peripherally. Combining those three measurements gives you a picture of body composition and fat distribution that is genuinely informative in a way that BMI alone is not, without requiring clinical equipment or specialist access for most of the pieces.
BMI is not useless, but treating it as the primary or sole indicator of health is where it causes problems. Knowing its specific failure modes, and knowing which measurements address those gaps, lets you use it for what it is actually good at while not letting it mislead you in the situations where it consistently falls short.