Submitted to: Clinical Nutrition
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 9/12/2009
Publication Date: 4/1/2010
Publication URL: http://handle.nal.usda.gov/10113/46985
Citation: Shafer, K.J., Siders, W.A., Johnson, L.K., Lukaski, H.C. 2010. Body Density Estimates from Upper-Body Skinfold Thicknesses Compared to Air-Displacement Plethysmography. Clinical Nutrition. 29(2): 249-254. Interpretive Summary: Interpretive Summary Assessment of human body composition rests upon determination of body density that is related to body fatness. Although a number of techniques are available to measure body density, many are costly and limited to use in a laboratory and require skilled personnel to perform the tests. A simple and cost-effective technique is the measurement of skin-fold thicknesses which are used in prediction equations to estimate body density. Preliminary information suggested that body size, as indicated by body mass index (BMI), can affect the validity of skinfold method to predict body composition. We tested the hypothesis that BMI would not impact the accuracy of the skinfold method compared to a reference method that directly measured body density in healthy adults with a wide range of BMI. Skinfold method underestimated body density in people with a normal BMI and overestimated it in women and the adults with a BMI classified as obese. These difference translated into errors in predicting body fatness of 2-3%, which is consistent with variability associated with the skinfold and density measurement methods except in the obese adults in whom the error was 6-7%. These findings suggest that skinfold measurements yield acceptable estimates of body fatness for normal and overweight people but not obese adults. This information will be useful to health professionals who assess body composition in wellness activities.
Technical Abstract: Technical Summary Objectives: Determine the effect of body mass index (BMI) on the accuracy of body density (Db) estimated with skinfold thickness (SFT) measurements compared to air displacement plethysmography (ADP) in adults. Subjects/Methods: We estimated Db with SFT and ADP in 131 healthy men and women classified by normal (N; 18.5-24.9 kg/m2), overweight (OW; 25-29.9 kg/m2), and obese (OB; 30-39.9 kg/m2) BMI. Results: Compared to ADP, SFT underestimated Db in the N BMI group (0.0058 g/cc; p<0.05) and overestimated Db in the total sample (0.0030 g/cc; p<0.001), women (0.0044 g/cc; p<0.05) and the OB BMI group (0.0141 g/cc; p<0.05). Db was not different between methods in the men or the OW BMI group. In the total group, Db estimates by method were highly correlated (r=0.823; p<0.0001); the error in Db determined by ADP and SFT increased with increasing Db (r=0.495; p<0.001). Within each BMI group, Db values estimated with SFT and ADP were related (r=0.650-0.888; p<0.0001). The difference in Db (ADP - SFT) increased with increasing Db values only in the OW BMI group (r=0.365; p<0.009). SFT overestimated body fatness in the N (2.61%, p<0.05) and underestimated it in the OB (6.75%, p<0.05) BMI groups. Conclusions: SFT measurements predict Db similarly as ADP in men and adults with BMI classified as OW. The error in estimation of Db and %fat with SFT in women and adults classified as N and OW BMI was within the precision of the SFT and ADP methods, and supports the use of SFT in these groups.