Beyond BMI: What Body Composition Numbers Really Tell You

The Body Mass Index was not invented by a physician. It was devised in the 1830s by Adolphe Quetelet, a Belgian astronomer, mathematician, and statistician, as part of his broader project to define 'the average man' - a statistical ideal of human characteristics. Quetelet was not studying health. He was studying population statistics, and he needed a simple way to compare body weights across different heights.

The formula he settled on - weight in kilograms divided by height in metres squared - was chosen for its mathematical convenience, not its medical accuracy. It provides a single number that roughly correlates with body size, which was useful for population-level statistics in an era before computers.

The index was largely forgotten until the 1970s, when physiologist Ancel Keys popularised it in a study of heart disease across populations. Keys gave it the name 'Body Mass Index' and argued it was superior to older measures like the Broca index (height in centimetres minus 100 equals ideal weight in kilograms). But even Keys acknowledged that BMI was a population-level screening tool, not an individual diagnostic instrument.

BMI measures exactly one thing: the ratio of your total body weight to the square of your height. It cannot distinguish between fat, muscle, bone, water, or any other component of body mass. Two people with identical heights and weights will have identical BMI values, regardless of whether one is a bodybuilder and the other is sedentary.

This is the fundamental limitation. BMI assumes that weight is a reasonable proxy for body fat, which is a reasonable assumption at the population level but a poor one for individuals. The correlation between BMI and body fat percentage is moderate - roughly 0.7 to 0.8 in most studies - meaning BMI explains about half the variance in body fat between individuals. The other half is unaccounted for.

The standard thresholds - 18.5 for underweight, 25 for overweight, 30 for obese - were chosen based on their statistical association with health outcomes in large epidemiological studies. They are not sharp boundaries. The health risk associated with weight increases gradually, not in steps, and the thresholds are simplifications of a continuous relationship.

Athletes are the most cited example of BMI's limitations, and for good reason. Muscle is denser than fat, so a muscular person will have a higher weight - and therefore a higher BMI - than a less muscular person of the same height and body fat percentage. A professional rugby player with 12 per cent body fat might register as obese on the BMI scale.

Less commonly discussed is the opposite problem. BMI can underestimate body fat in people who have lost muscle mass, which includes many older adults. A sedentary 70-year-old might have a 'normal' BMI while carrying a dangerously high proportion of visceral fat around their organs - a condition sometimes called normal-weight obesity.

BMI also does not account for fat distribution, which matters more than total fat for metabolic health. Visceral fat (stored around the abdomen and organs) is strongly associated with insulin resistance, cardiovascular disease, and type 2 diabetes. Subcutaneous fat (stored just under the skin, including on the hips and thighs) carries significantly less metabolic risk. Two people with the same BMI can have very different visceral fat levels and therefore very different health risks.

Pregnancy, amputation, and certain medical conditions can all make BMI misleading. For children and teenagers, BMI must be interpreted using age- and sex-specific percentile charts, not adult thresholds, because body composition changes dramatically during growth.

The standard BMI thresholds were derived primarily from data on European-descended populations. Research over the past two decades has shown that these thresholds do not translate well to all ethnic groups.

People of South Asian descent tend to develop metabolic complications at lower BMI values than the standard thresholds suggest. The World Health Organisation recommends that Asian populations use a lower overweight threshold of 23 rather than 25, and an obesity threshold of 27.5 rather than 30, because cardiovascular risk rises earlier in this group.

Conversely, some Polynesian populations have higher average muscle mass and bone density, which can inflate BMI without a corresponding increase in metabolic risk. Applying European-derived thresholds to these populations may overclassify them as overweight or obese.

These differences do not invalidate BMI as a tool, but they do mean that interpreting a BMI value requires context. A single number without that context is incomplete.

Several alternatives to BMI exist, each measuring something more specific but also more difficult or expensive to obtain.

Waist-to-height ratio is the simplest. It measures abdominal fat relative to height, and research suggests it is a better predictor of cardiovascular risk than BMI. The guideline is that your waist circumference should be less than half your height. It requires only a tape measure and is something you can do at home.

Body fat percentage can be estimated through several methods, each with different accuracy and cost. DEXA scans (dual-energy X-ray absorptiometry) are considered the gold standard and can distinguish fat, muscle, and bone with high precision, but they require specialised equipment and a clinical visit. Skinfold calipers, when used by a trained professional, can provide a reasonable estimate. Bioelectrical impedance scales, common in consumer devices, are convenient but can be inaccurate, especially if hydration levels fluctuate.

Waist circumference alone is also useful. For men, a waist circumference above 94 cm (37 inches) indicates increased risk, and above 102 cm (40 inches) substantially increased risk. For women, the corresponding figures are 80 cm (31.5 inches) and 88 cm (34.5 inches).

BMI is not useless. It is a quick, free, and non-invasive screening tool that, at the population level, correlates with health outcomes. For someone with no reason to suspect their body composition is unusual, a BMI in the normal range is reassuring and a BMI well above it is a signal to investigate further.

But it is a starting point, not a verdict. If your BMI suggests a concern, or if you have reason to believe it does not reflect your actual body composition, the next step is a more direct measure - waist-to-height ratio at minimum, and ideally a body composition assessment from a qualified professional.

Most importantly, BMI is not a measure of health, worth, or fitness. It is one data point among many, and it should be treated as such.