Durnin-Womersley 4-Site Skinfold Calculator

What is the Durnin-Womersley method?

The Durnin-Womersley 4-site skinfold method, published by J.V.G.A. Durnin and J. Womersley in 1974 in the British Journal of Nutrition, is one of the oldest and most cited skinfold equations still in active use. The method predicts body density from the sum of four skinfold measurements, biceps, triceps, subscapular, and suprailiac, using age- and sex-specific equations. Body density is then converted to body fat percentage using the Siri or Brozek two-component equations. The Durnin-Womersley method has been used in thousands of research studies and remains a reference standard for skinfold-based body composition assessment.

The choice of four sites reflects a deliberate balance between practicality and accuracy. The biceps and triceps represent the upper extremity, the subscapular represents the back, and the suprailiac represents the trunk. Together, these sites sample subcutaneous fat from multiple regions, providing a more representative estimate of total body fat than a single site could. The inclusion of the subscapular and suprailiac sites also captures trunk fat, which is metabolically significant and strongly correlated with cardiovascular risk.

Measurement sites

The four sites used in the Durnin-Womersley method are:

• Biceps: front of the upper arm, at the level of the mid-point between the acromion (shoulder) and olecranon (elbow) processes, on the midline of the biceps belly
• Triceps: back of the upper arm, at the same level as the biceps measurement, on the midline of the triceps belly
• Subscapular: 1-2 cm below the inferior angle of the scapula, taken at a 45-degree angle to the spine (following the natural cleavage lines of the skin)
• Suprailiac: just above the iliac crest at the midaxillary line, taken horizontally or slightly obliquely

All measurements are taken on the right side of the body with the subject standing relaxed. Each skinfold is grasped between thumb and index finger, lifted away from the underlying muscle, and measured with calipers placed about 1 cm from the fingers. The measurement is taken 2-3 seconds after releasing the caliper.

Age- and sex-specific equations

Unlike the Jackson-Pollock equations, which use a single set of coefficients with age as a continuous variable, the Durnin-Womersley method provides separate equations for five age groups in each sex. This captures the nonlinear relationship between subcutaneous fat, body density, and age. The equations use log₁₀ of the sum of the four skinfolds (in mm):

• Men 17-19: BD = 1.1620 − 0.0630 × log(sum)
• Men 20-29: BD = 1.1631 − 0.0632 × log(sum)
• Men 30-39: BD = 1.1422 − 0.0544 × log(sum)
• Men 40-49: BD = 1.1620 − 0.0700 × log(sum)
• Men 50+: BD = 1.1715 − 0.0779 × log(sum)

• Women 17-19: BD = 1.1549 − 0.0678 × log(sum)
• Women 20-29: BD = 1.1599 − 0.0717 × log(sum)
• Women 30-39: BD = 1.1423 − 0.0632 × log(sum)
• Women 40-49: BD = 1.1333 − 0.0612 × log(sum)
• Women 50+: BD = 1.1339 − 0.0645 × log(sum)

The coefficients generally increase (in absolute value) with age for men, reflecting that older men have higher body fat at any given subcutaneous fat thickness. For women, the pattern is more variable but generally shows less age dependence.

Converting body density to body fat

Once body density is estimated, it is converted to body fat percentage using a two-component equation. The Siri equation, body fat % = (495 ÷ body density) − 450, is the most commonly used. The Brozek equation, body fat % = (457 ÷ body density) − 414.2, gives similar results and is sometimes preferred for very lean individuals. The two equations differ in their assumptions about the densities of fat (0.9007 g/cc) and fat-free mass (1.100 g/cc for Siri, 1.100 g/cc for Brozek).

For Durnin-Womersley specifically, the original 1974 paper used the Siri equation, and this remains the standard conversion. Modern practice generally follows this convention to maintain comparability with the original validation data.

Comparison with the Jackson-Pollock method

The Durnin-Womersley and Jackson-Pollock methods are the two most commonly cited skinfold equations, and each has its strengths. Durnin-Womersley uses four sites that include the biceps and triceps (arm sites) and the subscapular and suprailiac (trunk sites), providing a balanced sampling of body regions. Jackson-Pollock uses different sites for men (chest, abdomen, thigh) and women (triceps, suprailiac, thigh) and includes age as a continuous variable.

In validation studies, the two methods give similar accuracy (standard error approximately 3.5-4.0 percentage points for both), but Durnin-Womersley is sometimes preferred for older adults because its age-specific equations better capture the nonlinear age effect. Jackson-Pollock's continuous age term is more convenient computationally but may be less accurate at the extremes of the age range. Durnin-Womersley is also often preferred in clinical and research settings in Europe, while Jackson-Pollock is more popular in the United States.

Limitations

The Durnin-Womersley method shares the general limitations of skinfold-based body composition assessment. The equations were derived from a sample of approximately 4,800 adults in Glasgow, Scotland, in the 1960s and 1970s, a predominantly white, European population. Subsequent validation studies in other populations have produced mixed results, with some showing good agreement and others finding systematic overestimation or underestimation of body fat. The equations may be less accurate in athletes, in obese individuals (where subcutaneous fat thickness may exceed caliper capacity), and in populations with body proportions that differ from the original sample.

The two-component model used to convert body density to body fat also has limitations. It assumes constant densities of fat and fat-free mass, which may not hold in athletes (denser fat-free mass), older adults (less dense fat-free mass due to bone loss), or children. Multi-component models that measure bone mineral density and total body water can correct for these variations but are more expensive and less accessible.

the Durnin-Womersley equations have a known tendency to underpredict body fat in very lean individuals (especially athletes) and overpredict it in very obese individuals. For these populations, alternative equations such as Jackson-Pollock (for lean athletes) or bioelectrical impedance (for obese individuals) may be more appropriate.

Reference ranges for body fat percentage

The American Council on Exercise (ACE) classification is commonly used to interpret the resulting body fat percentage:

• Men: 2-5% essential, 6-13% athletic, 14-17% fitness, 18-24% acceptable, >25% obese
• Women: 10-13% essential, 14-20% athletic, 21-24% fitness, 25-31% acceptable, >32% obese

These ranges are based on the general adult population. Athletes in specific sports may have different ranges, for example, endurance runners often sit at the lower end of the athletic range, while strength athletes may be at the upper end. Older adults have somewhat higher acceptable ranges due to age-related changes in body composition.

How to take accurate skinfold measurements

For accurate Durnin-Womersley measurements:

1. Use a high-quality caliper with constant spring tension (Lange, Harpenden, or Slimguide).
2. Identify and mark the four sites with a water-soluble pen before measuring.
3. Pinch the skinfold firmly between thumb and index finger, lifting it away from the muscle below.
4. Place the caliper jaws 1 cm away from your fingers, perpendicular to the skinfold.
5. Release the caliper pressure lever fully and read the dial after 2-3 seconds.
6. Take each measurement twice; if readings differ by more than 1-2 mm, take a third measurement and average the closest two.
7. Measure all sites in sequence, then repeat the entire sequence rather than measuring each site twice in a row.

Consistency in technique is more important than absolute precision. The same technician, caliper, and landmarks should be used for each session to track changes accurately over time.

Clinical considerations and nuances

The Durnin-Womersley method has been validated in many populations, with varying results. The original validation was in a Scottish sample, and subsequent studies have tested the equations in other European, North American, South Asian, East Asian, and African populations. Results have generally been good in European and North American samples, but mixed in other populations. In some Asian populations, the equations tend to overestimate body fat, possibly because Asian body proportions differ from the original Scottish sample. Some researchers have proposed population-specific correction factors, but none has been widely adopted.

In older adults (above age 60), the Durnin-Womersley equations may be less accurate because of age-related changes in body composition. Older adults have less dense fat-free mass (due to bone loss and reduced muscle glycogen) at any given body density, so the Siri equation's two-component model overestimates body fat. The age-specific equations partially address this by using different coefficients for older age groups, but the underlying two-component assumption still introduces error. For older adults, DXA or multi-component models are preferred when available.

In athletes, the Durnin-Womersley equations tend to underestimate body fat, particularly in strength athletes with high muscle mass. The four sites (biceps, triceps, subscapular, suprailiac) include two arm sites and two trunk sites but no leg sites, so athletes with disproportionate leg development may receive inaccurate results. The Jackson-Pollock method, which includes a thigh measurement, may be more appropriate for athletes. Alternatively, sport-specific equations or DXA can be used.

The Durnin-Womersley method remains a reference standard for skinfold-based body composition assessment in research settings. It is widely cited in the scientific literature and is often used as a comparison method when new body composition techniques are validated. The method's longevity (over 50 years in continuous use) reflects the quality of the original work and the lack of clearly superior alternatives in field settings. For most non-athlete adults aged 20-60, the Durnin-Womersley method provides accurate and useful body composition information when performed by a trained technician using a quality caliper.

Practical takeaways

The Durnin-Womersley 4-site skinfold method is a classic, well-validated approach to estimating body fat percentage. Its strengths include the inclusion of trunk sites (subscapular and suprailiac) that capture metabolically important fat, age-specific equations that better model the nonlinear age effect, and a strong evidence base derived from decades of research use. Like all skinfold methods, its accuracy depends on measurement technique, caliper quality, and adherence to standardized anatomical landmarks. For tracking body composition changes over time, the Durnin-Womersley method is a reliable choice when performed by a trained technician using a quality caliper.