Deurenberg Formula

Estimates body-fat percentage from BMI, age and sex — no tape measure needed. Fast and famous, with one honest catch its own authors pointed out.

Deurenberg (1991) turns BMI into a body-fat estimate with a simple linear equation. It's handy because it needs nothing but height, weight, age and sex — but because it's BMI-based it can't separate muscle from fat, and it runs high at the top of the BMI scale. WIMSTA shows it as one voice in a range, next to CUN-BAE and the tape-based methods.

The formula (adults)
body fat %=1.20·BMI+ 0.23·age− 10.8·sex− 5.4
BMI
weight (kg) ÷ height (m)²
sex
men = 1, women = 0
age
years — adults only (18+)
Run Deurenberg
Weight70 kg
Height168 cm
Age40 yrs
Deurenberg body fat
33.6 %
CUN-BAE, same inputs
35.2 %
Same inputs, different samples

Your BMI is 24.8. Deurenberg and CUN-BAE both read only BMI, age and sex, yet they disagree by 1.7 pp here — because one was fitted on Dutch adults and the other on Spanish adults. Push weight up and watch the gap widen: Deurenberg is known to overestimate body fat in obesity, by its own authors. That disagreement is the honest signal — the full calculator adds tape-based methods too.

It overestimates in obesity — by design

The equation is a straight line, but the real relationship between BMI and body fat bends at the high end. So above roughly BMI 30, Deurenberg tends to read a few points too high — a limitation the authors named themselves. It's a good reason never to trust a single body-fat number, and exactly why we surround it with other methods.

It was also derived on a Dutch sample and not validated in Asian populations, where the BMI-to-fat relationship differs.

Same inputs as CUN-BAE, different answer

Deurenberg and CUN-BAE both read only BMI, age and sex — yet they disagree, because one was fitted on Dutch adults and the other on Spanish adults. That's not a bug: two honest estimates from independent samples should differ, and the gap between them is real uncertainty you'd never see from one number alone.

Good for

A no-equipment body-fat estimate for adults of ordinary build (SEE 4.1 pp · R² 0.79), and tracking your trend.

Off for

Obesity (reads high), very muscular builds, Asian populations, under-18s and pregnancy.

One BMI formula is a guess. See the range.

Deurenberg, CUN-BAE, US Navy and RFM rarely agree on the same body. The calculator shows the honest range they land in — and where it sits against the healthy zones.

Open the body-fat calculator
Origin & verification · ✅ primary source

Deurenberg, Weststrate & Seidell (1991). Br J Nutr 65(2):105–114. Derived on n = 1,229, Netherlands, validated against densitometry (underwater weighing) (SEE 4.1 pp · R² 0.79). Coefficients verified against the primary source and covered by our test suite.

FAQ

What is the Deurenberg formula?

It estimates body-fat percentage from BMI, age and sex: body fat % = 1.20 × BMI + 0.23 × age − 10.8 × sex − 5.4 (sex is 1 for men, 0 for women). Because it uses only BMI, it needs no tape measure — but it also can't tell muscle from fat.

Why does Deurenberg overestimate body fat in obesity?

Its authors noted this themselves. The equation is a straight line fitted to BMI, but the real BMI-to-fat relationship curves at the high end, so above roughly BMI 30 the formula tends to read high. That's why WIMSTA shows it inside a range rather than alone.

Deurenberg or CUN-BAE — which should I use?

They read the same three inputs (BMI, age, sex) but were fitted on different populations — Deurenberg on Dutch adults, CUN-BAE on Spanish adults — so they legitimately disagree. Neither is "right"; the gap between them is useful information. WIMSTA runs both.

Is the Deurenberg formula accurate?

For adults of ordinary build it lands within about SEE 4.1 pp · R² 0.79. It was not validated in Asian populations, isn't for under-18s or pregnancy, and shares every BMI method's blind spot for muscle. Good as one voice in a range, not as a verdict.

How every formula is sourced: the method page.