Purpose of this study was to cross-validate SKF equations and bioimpedance equations to determine their accuracy for measuring body composition in American Indian men. For the sum 3SKF equation (chest + abdomen = thigh SKFs) the correlation coefficient ® was .83 and the prediction errors (SEE = 4.26 % BF or 0.0093 g/ml: E = 4.55 % BF or 0.0099 g/ml) were unacceptable. Similar results were with the " 7 SKF.Therefore, a race specific SKF equation was developed and cross-validated for this sample. The physical characteristics of the validation and cross-validation samples were similar, both groups were combined to develop a pooled, race-specific SKF equation for the entire sample (N = 97) Db (g/ml) = 1.138117 – 0.000887057(Waist Circumference.) – 0.000415103(Subscapular SKF; r = .85, SEE = .0089 g/ml. The pooled equation accurately estimated the %BF four component within ± 3.5% BF, for 91% of the individuals in this sample. In conclusion the Jackson and Pollock SKF (sum 3SKF and sum7SKF) equations should not be used to estimate the %BF or Db of American Indian Men.
The modified Segal fatness-specific equation yielded the most accurate estimate of FFM in this sample: 108 men age 18-62. We used the Valhalla Bioimpedance Analyzer (Model 1990B; Valhalla Scientific, SanDiego. The Modified Segal Fatness specific equation yielded the most accurate estimate of FFM in this sample. (r˛ ranged .94 to .95, and the prediction errors (SEE and E) were good (<2.8 kg), ranging from 2.5 to 2.65 kg. Using the Segal equation with the modified Segal fatness-specific method, the average predicted FFM of the sample was very close to reference values of FFM. Based on these results, we recommend using the Segal-Bioelectrical Impedance Analysis equations, in conjunction with the modified Segal fatness-specific method to assess the body composition of American Indian men. If the participant being assessed for body composition is obviously lean ( < 20% BF) or obviously obese at or > 20 %BF, use the appropriate Segal Fatness –Specific equation listed. If it is not obvious then use both equations and determine the average score. Segal Fatness-specific (1988)
( < 20 %BF) FFM (kg) = 0.00066366 (HT˛) – 0.02117® + 0.62854(WT) – 0.1238(AGE) + 9.33285 (R˛ = .89 and SEE (kg) = 2.5)
( at or > 20 %BF) FFM (kg) = 0.0008858(HT˛) – 0.02999® + 0.42688(WT) – 0.07002(Age) + 14.52435 (R˛ = .88 and SEE (kg) = 3.0)