Prediction of body fat by skinfold caliper: assumptions and cadaver evidence.

A series of assumptions required for use of skinfold calipers for the estimation of human body composition is examined in terms of direct evidence from the measurement and dissection of 6 male and 7 female cadavers. Skinfold compressibility showed significant inter-subject and inter-site differences with the highest values (64.7 percent and 63.8 percent) at the spinale and biceps sites and the lowest (33.6 percent and 34.4 percent) at the front thigh and medial calf. Skin thickness as a percentage of skinfold thickness was highest at the subscapular site (28.1 percent). Adipose tissue patterning showed great variability, indicating the importance of using skinfold caliper readings from a variety of different sites including upper limb, lower limb and trunk. Further sources of error include variability in fat content of adipose tissue (estimated at 20 percent in our sample) and in the internal and subcutaneous fat deposition. Recommendations for the use of skinfold calipers to assess adiposity and monitor change are made in the light of these findings.