A Useful Tool with Real Limitations
BMI has been a cornerstone of public health research and clinical screening for decades, and with good reason — it is cheap to calculate, requires no specialist equipment, and correlates with health outcomes at the population level. However, its use as an individual health indicator has attracted significant scientific criticism. The core problem is that BMI uses only two variables (weight and height) to stand in for a complex biological reality that includes body composition, fat distribution, metabolic health, and fitness.
Understanding what BMI does and does not measure helps you interpret your own result more accurately and know when you might need additional assessment.
The Muscle Mass Problem
The most frequently cited limitation of BMI is its inability to distinguish fat from muscle. Because both contribute to total body weight, a heavily muscled person will register a higher BMI than someone of the same height with average muscle and more fat. Elite athletes, bodybuilders, rugby players, and many regular gym-goers frequently fall in the 'overweight' BMI range despite having very low body fat percentages and excellent metabolic health.
This is not a trivial edge case. Research has shown that BMI misclassifies a meaningful proportion of the population — particularly active men. A 2008 study by Romero-Corral et al. in the International Journal of Obesity found that among US adults, BMI misclassified body fat status in about 19% of men and 25% of women. For overweight BMI in particular, a substantial number of people classified as overweight have normal body fat and are not at elevated health risk.
Ethnic and Racial Variation
Standard BMI thresholds were derived largely from data on European populations. Studies across different ethnic groups have consistently found that people of South Asian, East Asian, and some other backgrounds develop metabolic complications — insulin resistance, type 2 diabetes, cardiovascular disease — at lower BMI values than people of White European heritage at the same BMI.
The WHO Expert Consultation on BMI published recommendations that for Asian populations, a BMI of 23.0 should be considered the threshold for overweight (rather than 25.0) and 27.5 for obesity (rather than 30.0). Many diabetes and cardiovascular risk assessment programmes in the UK and internationally now apply these revised thresholds for patients of South Asian heritage. Using standard thresholds alone for these populations risks missing people at genuine metabolic risk.
Age and Sex Differences
As discussed elsewhere, women carry a higher percentage of body fat than men at the same BMI due to essential fat requirements related to hormonal function. This means the same BMI does not represent the same level of body fatness in a man and a woman — which limits the comparability of BMI across sexes without adjustment.
Similarly, older adults tend to have less muscle mass and more fat than younger adults at the same weight, so a BMI of 24 in a 65-year-old may correspond to a higher body fat percentage than the same BMI in a 25-year-old. This age-related compositional drift is one reason some researchers and clinicians argue for age-adjusted BMI thresholds, though no universal standard has been adopted.
Fat Distribution Is Invisible to BMI
Perhaps the most clinically significant limitation of BMI is that it provides no information about where fat is distributed in the body. Two people with identical BMIs can have very different fat distributions — one carrying most fat subcutaneously in the limbs and hips, the other concentrated viscerally around the abdominal organs. Visceral fat is metabolically far more dangerous, associated with insulin resistance, dyslipidaemia, hypertension, and increased cardiovascular risk independent of total body fat.
This means that a person with a normal BMI but a large waist circumference — indicating central adiposity — may have a higher metabolic risk profile than someone with a slightly elevated BMI but a lean waist. Without a waist measurement, this risk is entirely invisible to BMI alone.
Alternatives and Complements to BMI
Waist circumference is the single best practical complement to BMI for assessing metabolic risk. It reflects visceral fat more directly than any other simple measurement. WHO risk thresholds are >94 cm (men) and >80 cm (women) for increased risk, and >102 cm (men) and >88 cm (women) for substantially increased risk. Waist-to-height ratio (waist divided by height, with a threshold of 0.5) is increasingly recommended as a single metric that captures both central adiposity and body size, and performs better across ethnic groups than absolute waist cutpoints.
Body fat percentage measured via DEXA, BIA, or validated circumference-based methods provides the most direct assessment of body composition. Waist-to-hip ratio (waist divided by hip circumference) is a useful indicator of fat distribution pattern. In clinical settings, these measurements are used alongside blood biomarkers — fasting glucose, HbA1c, lipid panel, blood pressure — for a complete cardiometabolic risk assessment.
The Right Way to Use BMI
BMI is a screen, not a diagnosis. An unusual result — in either direction — should prompt a consultation with your GP rather than self-diagnosis.
Despite its limitations, BMI remains a valuable first-line screening tool at the population and individual level. Its value lies in flagging people who may benefit from further assessment — not in making definitive health judgements. Treat your BMI result as a starting point for a conversation with a healthcare professional, not as a diagnosis. Combine it with waist circumference at a minimum, and seek a fuller assessment if you have additional risk factors or fall in a category where BMI is known to be less reliable.