BMI Limitations: What It Misses and What to Measure Instead

BMI is weight divided by height squared. It was designed in the 1830s to describe populations, and it does that job reasonably well. Applied to one person it carries a specific, predictable set of errors — and knowing them is the difference between a useful screening number and a misleading one.

The formula, and two worked examples

Metric: BMI = kg ÷ m². Imperial: BMI = 703 × lb ÷ in², where 703 is the unit-conversion constant, not a health factor.

Example 1 (metric). 100 kg, 1.80 m. Square the height first: 1.80² = 3.24. Then 100 ÷ 3.24 = 30.86. Rounded, BMI 30.9.

Example 2 (imperial). 154 lb, 70 in. 70² = 4,900. Then 703 × 154 ÷ 4,900 = 22.09. Converting the same person to metric — 69.85 kg and 1.778 m — gives 69.85 ÷ 3.161 = 22.1, confirming the two formulas agree.

The conventional thresholds:

Working backwards is often more useful than the forward calculation. At 1.75 m, BMI 25 corresponds to 76.6 kg and BMI 18.5 to 56.7 kg — a 20 kg band described as "normal", which tells you how coarse the measure is.

What BMI cannot see

BMI has no information about body composition. It cannot distinguish:

• muscle from fat
• subcutaneous fat from visceral fat
• a heavy skeleton from a light one
• where on the body the mass sits

Those distinctions are most of what determines metabolic risk.

The five systematic errors

Muscular people

A 1.80 m rugby forward at 100 kg has a BMI of 30.9 — "obese" — while possibly carrying 12% body fat. Studies of professional athletes routinely find a majority classified as overweight or obese by BMI alone.

Older adults

Muscle declines with age while fat increases. Someone whose weight has not changed in twenty years may have swapped a substantial amount of muscle for fat, with BMI reporting no change at all.

Height extremes

Because the formula squares height rather than cubing it, BMI under-reads tall people and over-reads short people. Height scales roughly linearly, but body mass scales closer to a cube.

Make that concrete. A 1.60 m person at 57.6 kg has a BMI of 22.5. Scale that body geometrically to 2.00 m — every dimension multiplied by 1.25 — and the mass multiplies by 1.25³ = 1.95, giving 112.5 kg. The BMI of that identically proportioned person is 112.5 ÷ 4.00 = 28.1, in the overweight band, purely because of height. Nothing about the body changed except its scale.

Children and adolescents

The adult cut-offs do not apply at all. A growing child's healthy BMI changes year by year, so paediatric assessment uses age- and sex-specific percentile charts rather than fixed numbers. A BMI of 19 can be high for a six-year-old and low for a seventeen-year-old. Applying the adult table to a child produces a meaningless answer.

Ethnicity

Metabolic risk rises at lower BMI in South Asian and East Asian populations, which is why several health authorities recommend an overweight threshold around 23 rather than 25 for these groups. Using the standard cut-off understates risk for a very large share of the world's population.

Measurements that work better

Waist circumference

Directly reflects abdominal fat, which is the fat most strongly linked to cardiovascular and metabolic disease. Risk rises above roughly 94 cm for men and 80 cm for women, with higher-risk thresholds at 102 cm and 88 cm.

Waist-to-height ratio

The simplest useful rule available: keep your waist under half your height. It needs no tables, adjusts automatically for body size, and outperforms BMI as a predictor in most published comparisons.

The arithmetic is one division. At 180 cm tall with a 94 cm waist the ratio is 94 ÷ 180 = 0.52, just over the 0.50 line; the half-height target is 90 cm. A 102 cm waist on the same frame gives 0.57. Because both measurements are in the same unit, the ratio works identically in inches: a 37-inch waist on a 71-inch frame is also 0.52.

Measure at the midpoint between the lowest rib and the top of the hip bone, at the end of a normal exhale, with the tape level and snug but not compressing. Measuring across the navel instead, or pulling the tape tight, moves the result by several centimetres and is the main source of error in every waist-based metric.

Waist-to-hip ratio

Distinguishes fat distribution patterns. Above 0.90 for men or 0.85 for women indicates central adiposity. See the waist-to-hip ratio calculator.

Body fat percentage

What BMI is attempting to approximate. Tape-based methods such as the US Navy formula are free and reasonably consistent — see the body fat calculator. Healthy ranges are roughly 10–20% for men and 18–28% for women, rising with age.

When BMI is still the right tool

• Population studies, where individual error averages out.
• Tracking one person over time, where composition is roughly stable and the

trend is informative.

• A first screen, cheap and requiring only a scale and a tape.

Use it as a prompt to look further, never as a conclusion. If you want a number that drives decisions, the TDEE calculator gives you something you can act on.