BMI explained: what the number means and what it misses
Body Mass Index is one of the most widely used numbers in public health — and one of the most misunderstood. Developed in the early 19th century by Belgian mathematician Adolphe Quetelet as the "Quetelet Index", it was never intended as a measure of individual health. It was a statistical tool to characterise populations. Only in the 1970s, when researchers needed a simple cross-population comparison of body size, did the Quetelet Index get rebranded as BMI and slowly become the screening default for clinics, insurers and governments. Understanding both what BMI does well and what it misses is essential for using it sensibly.
How BMI is calculated
The formula is identical regardless of source: weight in kilograms divided by height in metres squared. A person who weighs 75 kg and stands 1.75 m tall has a BMI of 75 / (1.75 × 1.75) = 24.5. In imperial units, the equivalent is (weight in pounds × 703) / (height in inches squared). The squaring is the part most people forget — it means BMI scales non-linearly with height. Two people with identical body proportions but different heights produce slightly different BMIs, which is one of the formula\'s structural weaknesses.
The WHO categories
The World Health Organization adopted the current cut-offs in 1995: below 18.5 is underweight, 18.5 to 24.9 is the "healthy" range, 25.0 to 29.9 is overweight, and anything 30 or higher is obese (further subdivided into classes I, II and III). These thresholds were chosen because population studies show inflection points in mortality risk around them — the so-called "obesity J-curve" where both very low and very high BMIs predict higher all-cause mortality. The lowest-risk range in most large studies sits around BMI 22–23 for adults under 65.
What BMI predicts well
At population level, BMI correlates reasonably well with body fatness and tracks risk for cardiovascular disease, type 2 diabetes, hypertension, several cancers (colorectal, breast, endometrial, kidney), gallbladder disease, osteoarthritis and obstructive sleep apnoea. It is cheap, requires no special equipment, and reproduces across observers. As a triage tool — separating people who probably don\'t need further screening from those who probably do — it is genuinely useful.
What BMI gets wrong
BMI doesn\'t distinguish fat from muscle. A bodybuilder at 100 kg and 1.78 m has a BMI of 31.6 — formally "obese" — while carrying perhaps 8% body fat. A sedentary 75 kg woman of the same height has a BMI of 23.7, formally "healthy", while potentially carrying 35% body fat and significant metabolic risk. The phenomenon of "normal-weight obesity" — high body fat with a healthy BMI — is increasingly recognised in research, particularly in older women, and carries meaningfully elevated cardiometabolic risk despite a reassuring BMI.
BMI also ignores fat distribution. Visceral fat — the fat stored around internal organs — is far more metabolically dangerous than subcutaneous fat on hips and thighs. Two people with identical BMI and identical body-fat percentage can have very different risk profiles depending on where the fat sits. Waist circumference and waist-to-height ratio (target: under 0.5) are simple complements that capture this. A general rule: regardless of BMI, waist circumference above 102 cm in men or 88 cm in women indicates significant abdominal obesity.
The formula has known biases by ethnicity. Asian populations develop diabetes and cardiovascular disease at lower BMIs than European populations, which is why the WHO recommends lower BMI cut-offs (overweight ≥23, obese ≥27.5) for Asian adults. Conversely, populations of African descent often carry more lean mass per unit height, pushing BMIs upward without corresponding fat-mass increases. Age matters too — the lowest-mortality BMI in adults over 65 shifts slightly upward into the 23–27 range, partly because some buffer against sarcopenia and illness benefits older adults.
BMI and weight loss decisions
For most adults, BMI is a reasonable starting reference for "should I think about losing weight?" — but it should never be the only signal. A 30-year-old man with BMI 27, low waist circumference, normal blood pressure, normal lipids and a heavy lifting habit has very different risk than a 30-year-old man with BMI 24, a 95 cm waist, prediabetic fasting glucose and zero exercise — even though only one is "overweight" by BMI. Use BMI to start the conversation, then layer in waist circumference, body-fat percentage, blood markers and how your clothes actually fit. If BMI is in the obese range (30+) regardless of build, the conversation is worth having; lifestyle change at that level is high-value almost universally.
Setting a realistic weight target
Once you know your BMI and your goal range, target the middle of the healthy range (BMI 22), not the absolute lower bound. Aiming for BMI 22 gives a sustainable body composition for most people, leaves room for natural fluctuation, and matches the body-mass range with the best mortality data. To find your target weight in kilograms, multiply 22 by your height in metres squared: a 1.70 m adult lands at roughly 63.6 kg. Combine that with the rule that 5–10% total weight loss produces the bulk of metabolic benefits, and you have a clear, evidence-based target without needing to obsess about the BMI number itself.