BMI Calculator

For a person who is 70 kg and 170 cm tall: height in meters = 170 / 100 = 1.7 m. BMI = 70 / (1.7 × 1.7) = 70 / 2.89 ≈ 24.2, which falls in the Normal category. The healthy weight range for this height is about 53.5 kg to 71.9 kg.

BMI is a ratio of weight to height, and nothing more. It was developed in the 19th century as a population statistic - not a clinical test for individuals - and it retains that character today.

• It does not distinguish fat from muscle. A lean, heavily muscled person can register as overweight or obese purely because muscle is denser than fat.
• It does not describe fat distribution. Two people with identical BMI can have entirely different health profiles depending on whether fat is carried around the abdomen or elsewhere - and abdominal fat carries the higher risk.
• It says nothing about fitness. Cardiorespiratory fitness is independently associated with health outcomes and is invisible to this measurement.
• It is a screening signal, not a diagnosis. It flags a range worth investigating, not a condition.

Note. BMI is useful precisely because it is cheap, instant and needs no equipment. The mistake is treating the number as the conclusion rather than as the first question.

Obesity is further divided into classes at 30, 35 and 40. The thresholds were set by linking BMI to mortality and disease risk in large populations - they are statistical boundaries, not physiological switches. Moving from 24.9 to 25.0 does not change anything about a person; crossing it is a rounding event.

• Lower thresholds apply to some populations. Research indicates that cardiometabolic risk rises at lower BMI among people of South Asian, Chinese, Southeast Asian and Japanese descent. Many health bodies use 23.0 as the overweight cutoff and 27.5 as the obese cutoff for these groups, reflecting higher risk at the same BMI.
• Higher muscle mass warrants a more cautious reading of the normal range in trained individuals.
• Elderly adults often benefit from a slightly higher range, where a BMI below 22-23 is associated with increased risk rather than reduced risk.

The same person at 76 kg would sit at the top of the healthy range - a difference of 13.7 kg. Whether that gap represents fat, muscle or a mixture is exactly what BMI cannot tell you.

Note also how sensitive the figure is to height: at the same 90 kg, a height of 1.70 m gives 31.1 rather than 29.4, and 1.80 m gives 27.8. Two adults of identical weight can land in different categories purely on a 10 cm height difference.

1. Waist circumference - measured at the navel, after a normal exhale. It reflects abdominal fat directly, which is the metabolically relevant depot. Risk rises above roughly 94 cm for men and 80 cm for women, with higher thresholds in some guidelines.
2. Waist-to-height ratio - keeping the waist below half your height is a simple heuristic that works across populations and avoids gender-specific cutoffs.
3. Waist-to-hip ratio - compares distribution; an apple shape carries more risk than a pear shape at the same BMI.
4. Body composition - calipers, bioimpedance or DEXA estimate fat percentage directly. Each has its own error, but they answer the question BMI cannot.
5. Trend over time - a stable weight and waist matters more than the position of a single reading.

• Children and teenagers - adult cutoffs do not apply. Growth requires age- and sex-specific percentile charts, and the interpretation is entirely different.
• Pregnant women - weight change is expected and is monitored by a different measure entirely.
• Athletes and bodybuilders - muscle mass routinely produces overweight or obese readings on healthy individuals.
• People recovering from illness - rapid weight change needs clinical context rather than a category label.
• Anyone using it as a target - pursuing a BMI number rather than measurable health outcomes leads to decisions that the number cannot justify.

Note. If the calculator returns a number that surprises you, that is the expected outcome for a meaningful proportion of people. It is an argument for measuring something better, not an argument that the measurement is wrong.