Four formulas, and why they disagree
Four formulas, and why they disagree
This is the most important thing a calculator like this can tell you. Several formulas exist, they were developed for different purposes, and they disagree with each other by ten pounds or more at the same height. None of them measures health.
What they provide is a reference range. Whether a particular weight is healthy for a particular person depends on body composition, distribution of fat, fitness, blood markers, family history, and a good deal else that a height measurement cannot capture.
Devine (1974) — originally created for calculating medication dosages, not health targets. It remains the most widely used.
Men: 50 kg + 2.3 kg per inch over 5 ft
Women: 45.5 kg + 2.3 kg per inch over 5 ft
Hamwi (1964) — developed for diabetes management. Men: 106 lb + 6 lb per inch over 5 ft. Women: 100 lb + 5 lb per inch over 5 ft.
Robinson (1983) — a revision of Devine using updated population data, generally producing slightly lower figures.
Healthy BMI range — not a single number but the weight range corresponding to a BMI of 18.5 to 24.9. This is the most useful of the four precisely because it is a range rather than a point.
Each formula was built from a different population for a different purpose, decades apart. Devine was designed so clinicians could estimate drug dosages consistently; it was never validated as a health target. Hamwi came from dietetic practice with diabetic patients. Robinson attempted to correct Devine against newer data.
None was derived from outcome data linking a specific weight to better health. That is a genuine limitation, and it is why the BMI range — which at least draws on large population health studies — is usually the more defensible reference despite its own well-known flaws.
Every formula here uses height and sex, and nothing else. That means each shares the same blind spots as BMI.
Muscle mass. A trained athlete will often exceed every formula's output while carrying low body fat and excellent health markers. The formula reports a problem that does not exist.
Frame size. Bone structure varies meaningfully between people of the same height. This calculator applies a frame adjustment, but frame size is itself estimated rather than measured.
Fat distribution. Where fat is carried matters more for metabolic risk than how much. Visceral fat around the organs is associated with considerably higher risk than the same amount carried subcutaneously.
Age and ethnicity. Body composition changes with age, and health risk thresholds differ across populations — several health bodies use lower BMI cut-offs for people of South Asian descent, for instance.
If the goal is assessing health rather than hitting a number, several measures carry more information:
If your weight sits outside these ranges, that is information rather than a verdict. Many people are healthy at weights the formulas would flag, and some people within the ranges have poor metabolic health. The number is a prompt to look at the more informative measures above, not a target to pursue on its own.
If you are considering significant weight change, a doctor or registered dietitian can assess the factors these formulas cannot see and give advice grounded in your actual health rather than your height.
None is a health measure. The healthy BMI range is generally the most defensible reference because it derives from population health data and gives a range rather than a single number. Devine, Hamwi, and Robinson were built for clinical dosing and dietetic practice rather than health targets.
They were developed decades apart, from different populations, for different purposes. Devine was created for medication dosing, Hamwi for diabetes management, and Robinson as a revision of Devine. None was validated against health outcomes.
It may mean nothing on its own, particularly if you carry significant muscle. Waist circumference, waist-to-height ratio, body fat percentage, and blood markers give a far better picture of health than weight against a height formula.
Bone structure does vary between people of the same height, which is why frame adjustments exist. However, frame size is usually estimated rather than measured, so the adjustment adds only a rough correction.