Body mass index, or BMI, is a screening measure calculated from body mass and height. For adults, BMI is calculated by dividing weight in kilograms by height in metres squared. The resulting value is expressed in kilograms per square metre, although the units are often omitted in ordinary usage.
In the commonly used adult classification, a BMI below 18.5 is classified as underweight, 18.5 to less than 25 as healthy weight, 25 to less than 30 as overweight and 30 or above as obesity, with obesity sometimes further divided into classes. These categories are population-level screening thresholds rather than precise measurements of an individual's body fat, health or nutritional status. BMI is therefore useful partly because it is simple and inexpensive, not because it captures every aspect of body composition.
The mathematical concept underlying BMI has a long history. The measure was developed from the work of Belgian mathematician and astronomer Adolphe Quetelet, who studied relationships between human measurements and population characteristics in the nineteenth century. Quetelet used what was known as the Quetelet Index, based on weight divided by height squared. The term body mass index became widely established later, particularly through twentieth-century epidemiological and public-health research. The historical development is important because the measure was designed as a population-level statistical indicator rather than as a diagnostic instrument capable of precisely determining an individual's adiposity or health status.
BMI is useful because height and weight are relatively easy to measure and standardise across large populations. Researchers can use BMI categories to investigate relationships between body size and outcomes such as cardiovascular disease, type 2 diabetes and mortality. Public-health organisations can also use BMI distributions to monitor population trends. At the individual level, BMI can serve as an initial screening indicator that prompts consideration of other measurements or risk factors. Its simplicity makes it particularly useful when more detailed assessments of body composition are unavailable. However, its usefulness as a screening tool should not be confused with the claim that it provides a complete assessment of an individual's health.
BMI does not directly measure body fat. It is possible for two people with the same BMI to have substantially different proportions of fat and lean tissue. A muscular person can have a relatively high BMI because muscle contributes to body mass, even when body-fat percentage is comparatively low. Conversely, someone can fall within a conventionally normal BMI range while having a relatively high proportion of body fat, particularly if muscle mass is low. BMI also does not reveal where fat is distributed. Visceral fat stored around internal organs is metabolically different from subcutaneous fat stored beneath the skin, and waist circumference can provide information about central adiposity that BMI cannot.
Age and sex affect the interpretation of body composition, which is why adult BMI thresholds should not simply be applied to children. For children and adolescents, BMI is interpreted in relation to age and sex using growth references or percentiles because body composition changes substantially during development. Pregnancy is another situation in which ordinary BMI interpretation is inappropriate because pregnancy-related changes in body mass include the fetus, placenta, amniotic fluid and maternal tissues. Older adults may also require additional context because changes in muscle mass and fat distribution can occur with ageing. BMI therefore needs to be interpreted according to the population and physiological context in which it is being used.
Ethnicity can also be relevant to health-risk interpretation because the relationship between BMI, body-fat distribution and metabolic risk varies among populations. Some ethnic groups can experience increased cardiometabolic risk at lower BMI values than the standard international adult thresholds indicate, while other populations can have different distributions of body fat and lean mass at the same BMI. Public-health and clinical guidance in some countries consequently uses ethnicity-specific considerations or alternative thresholds in particular circumstances. This does not mean that there is a universally superior BMI system; rather, it demonstrates that BMI is a statistical screening measure whose relationship with health outcomes depends partly on the population being assessed.
BMI also has important limitations when used to make individual judgments. It does not measure blood pressure, cholesterol, blood glucose, cardiovascular fitness, strength, diet quality, physical activity, mental wellbeing or most other dimensions of health. Nor does a BMI category by itself establish the cause of a person's weight, the presence of disease or the appropriate treatment. A person's waist circumference, body composition, medical history, family history, laboratory measurements and lifestyle may provide important additional information. BMI should therefore be interpreted as one indicator among several rather than as a definitive statement about an individual's health or worth.
The main distinction to retain is that BMI measures a mathematical relationship between weight and height, whereas body-fat percentage measures the proportion of body mass that is fat, and waist circumference provides an indicator of abdominal size and potentially central adiposity. These measurements answer different questions and can complement one another.
BMI remains widely used because it is cheap, reproducible and practical for population screening, but its limitations become more important when it is applied to individuals with unusual body composition or particular physiological circumstances. A sound interpretation therefore treats BMI as a screening tool that can identify a broad weight category and indicate whether further assessment may be useful, rather than as a direct measurement of body fat or a standalone diagnosis of health.
