Technical content for health and fitness professionals. It reproduces formulas and protocols published in the scientific literature, with the sources at the end of the page. It is not a recommendation and does not replace the judgment of the professional who runs the assessment.
The Siri and Brozek equations do the same thing: they turn body density, calculated from skinfolds or measured in a laboratory, into body fat percentage. Siri (1961) uses body fat (%) = (4.95 ÷ density − 4.50) × 100; Brozek et al. (1963) use body fat (%) = (4.57 ÷ density − 4.142) × 100. Across the most common density range in adults, the difference between them is under 1 percentage point, and what matters in an assessment is always using the same one.
In the assessor's routine, the conversion appears at the end of almost every skinfold protocol: the density equation from Jackson and Pollock 7-site or from Durnin and Womersley gives the density, and Siri or Brozek gives the percentage.
The formulas in full
Body density is in g/cm³ (the same as g/ml).
Siri (Siri, 1961):
Body fat (%) = (4.95 ÷ density − 4.50) × 100
Brozek (Brozek et al., 1963):
Body fat (%) = (4.57 ÷ density − 4.142) × 100
Both start from the two-compartment model: the body is divided into fat mass and fat-free mass, each with a fixed density. Siri assumes 0.90 g/cm³ for fat and 1.10 g/cm³ for fat-free mass (Siri, 1961). Brozek et al. (1963) revised these assumptions and arrived at slightly different constants. When the fat-free mass of the person being assessed departs from the assumed density, as in children and older adults, both are wrong, and there are specific conversions for some groups (Heyward and Wagner, 2004).
Worked example
In the example from the 7-site skinfold article, a 30-year-old man with a sum of 100 mm reached a density of 1.0653532 g/cm³.
- Siri: 4.95 ÷ 1.0653532 = 4.64635; 4.64635 − 4.50 = 0.14635; × 100 = 14.63%.
- Brozek: 4.57 ÷ 1.0653532 = 4.28966; 4.28966 − 4.142 = 0.14766; × 100 = 14.77%.
The difference is 0.13 percentage points, much smaller than the error of the skinfold equation itself.
Siri and Brozek side by side
Values calculated with the two formulas above:
| Density (g/cm³) | Siri | Brozek | Difference (Siri − Brozek) |
|---|---|---|---|
| 1.09 | 4.1% | 5.1% | −0.9 |
| 1.08 | 8.3% | 9.0% | −0.6 |
| 1.07 | 12.6% | 12.9% | −0.3 |
| 1.06 | 17.0% | 16.9% | 0.1 |
| 1.05 | 21.4% | 21.0% | 0.4 |
| 1.04 | 26.0% | 25.2% | 0.7 |
| 1.03 | 30.6% | 29.5% | 1.1 |
| 1.02 | 35.3% | 33.8% | 1.5 |
The curves cross near 1.06 g/cm³. In people with low body fat, Brozek gives a little more; in people with more body fat, Siri gives more, and the difference grows as density falls.
Siri or Brozek: which one to use?
- The one that goes with the density equation. Jackson and Pollock (1978) converted density with Siri, and that is the conversion that goes with their equations in this site's calculators.
- Special groups. In older adults, Guerra et al. (2010) compared the two with DXA in 60 people aged 60 to 92: both overestimated body fat, and Brozek agreed better (mean difference of 4.0 points, against 5.7 with Siri). For children, older adults and some ethnic groups, Heyward and Wagner (2004) collect specific conversion formulas.
- Consistency. Switching conversions between assessments creates a change that did not happen in the body. Record the formula used in the client's file.
Limits
- Density error. The conversion inherits the error of the density equation. In skinfold equations, this error is on the order of 3 to 4 percentage points against the reference method (Heyward and Wagner, 2004), much larger than the difference between Siri and Brozek.
- Two-compartment model. Both formulas assume a fat-free mass with fixed composition. Hydration, bone mass and the proportion of muscle change this density from person to person (Heyward and Wagner, 2004).
In PounceFit, body assessment software, the skinfold protocols that go through density (Jackson and Pollock, Guedes and Durnin and Womersley) convert with the Siri equation. To see the conversion inside a complete protocol, use the 7-site calculator or the Durnin and Womersley one.
Frequently asked questions
What is the difference between the Siri and Brozek equations?
Both convert body density into body fat percentage using the two-compartment model, but with different constants. Siri (1961): body fat (%) = (4.95 ÷ density − 4.50) × 100. Brozek et al. (1963): body fat (%) = (4.57 ÷ density − 4.142) × 100. Across the usual density range, the results are close (Heyward and Wagner, 2004).
Which one gives the higher percentage?
It depends on the density. By the formulas themselves, the two cross near a density of 1.06 g/cm³, about 17% body fat. Above that density (leaner people), Brozek gives a slightly higher value; below it (people with more fat), Siri gives the higher value.
Can I switch from Siri to Brozek between two assessments?
It is not advisable. The switch changes the number without the body having changed. Use the same conversion in all of the client's assessments, preferably the one that goes with the chosen density equation, such as Siri in the Jackson and Pollock (1978) protocols.
Which is better in older adults?
In a study of 60 older adults aged 60 to 92, both overestimated body fat compared with DXA, and Brozek came closer: a mean difference of 4.0 points against 5.7 points with Siri (Guerra et al., 2010). It is a small study with one specific density equation and does not set a general rule.
Sources
- Siri WE. Body composition from fluid spaces and density: analysis of methods. In: Brožek J, Henschel A, editors. Techniques for measuring body composition. Washington: National Academy of Sciences; 1961. p. 223-244.
- Brožek J, Grande F, Anderson JT, Keys A. Densitometric analysis of body composition: revision of some quantitative assumptions. Annals of the New York Academy of Sciences. 1963;110(1):113-140. doi:10.1111/j.1749-6632.1963.tb17079.x
- Guerra RS, Amaral TF, Marques E, Mota J, Restivo MT. Accuracy of Siri and Brozek equations in the percent body fat estimation in older adults. Journal of Nutrition, Health and Aging. 2010;14(9):744-748. doi:10.1007/s12603-010-0112-z
- Jackson AS, Pollock ML. Generalized equations for predicting body density of men. British Journal of Nutrition. 1978;40(3):497-504. doi:10.1079/BJN19780152
- Heyward VH, Wagner DR. Applied body composition assessment. 2nd ed. Champaign: Human Kinetics; 2004.
