Impedance vector distribution by sex, race, body mass index, and age in the United States: standard reference intervals as bivariate Z scores.

Bioelectrical impedance measurements were collected in the Third National Health and Nutrition Examination Survey (NHANES III), but their results have not been published. In the NHANES III population, resistance (R) and reactance (Xc) values at 50-kHz frequency were obtained with a Valhalla Scientific meter (model 1990B; San Diego, CA, USA). The RXc graph method was used to identify bivariate pattern distributions of mean vectors (95% confidence ellipses by sex, race, age, and body mass index [BMI]), and individual impedance vectors (50%, 75%, and 95% tolerance ellipses). Data from 10 222 adults (5261 men and 4961 women) formed 90 four-way classification groups, with two sexes, three races or ethnicities (non-Hispanic white, non-Hispanic black, Mexican American), five age classes (20-29, 30-39, 40-49, 50-59, and 60-69 y), and three BMI classes (19-24.9, 25-29.9, and 30-34.9 kg/m(2)). Sex, race or ethnicity, BMI and age, in decreasing order, influenced the vector distribution pattern. Mean vectors in women were significantly longer than those in men. Within each sex, the mean vector of non-Hispanic white subjects was shorter and with a smaller phase angle than that of corresponding BMIs from the two other race/ethnic populations. Tolerance ellipses were calculated from sex- and race-specific reference populations 20 to 69 y old and 19 < or = BMI < 30 kg/m(2) (8022 subjects, 4226 men and 3796 women). After transformation of impedance vector components into bivariate Z scores (standardized deviates, as differences from the mean divided by the standard deviation of the reference population), we constructed one standard, reference, RXc-score graph (50%, 75%, and 95% tolerance ellipses) that can be used with any analyzer in any population. The pattern of impedance vector distribution and reference bivariate intervals for the individual impedance vector are presented for comparative studies (free software at E-mail: apiccoli@unipd.it).