Waist circumference measurement by site, posture, respiratory phase, and meal time: implications for methodology.

Waist circumference (WC) has been advocated as a simple, reliable, and cost-effective measure to understand an individual's cardio-metabolic risk. Although several protocols exist for measuring WC, the variation induced by a few factors has not been investigated. We compared several established and experimental WC measurement protocols to identify factors that may cause variations in WC measurement. In this cross-sectional study, we examined the variations in the measurement of waist circumference (WC) measures carried out in 11 ways differing by anatomical site, posture, respiratory phase, and time since last meal, using repeated measure analysis of variance (using mixed models) after Tukey-Kramer adjustment. We estimated the proportion of variance in percentage of body fat (%BF) and fat-free mass (FFM) explained by each of the WC measures. We studied 123 apparently healthy Asian Indians (75 females), with mean (s.d.) age of 34 (8.7) years and BMI of 23.9 (4.8) kg/m(2). Overall, the mean of WCs measured using the 11 protocols were statistically different. Further, post hoc analysis showed statistically significant, yet mostly small, differences between most of the pairs. No single WC measure explained highest variance in %BF or FFM for both genders. Although, the National Institute of Health (NIH), USA, protocol was convenient and may be less prone to errors, at present it does not control for many variables tested in this study. Measures of WC measured using different protocols were statistically different. We suggest that the site of measurement, posture, phase of respiration, and time since last meal should be standardized for the development of a protocol for measurement of WC for worldwide use.