Distribution of fat is important when considering health risk; however, the value added from skinfold measurements (SKF) when using body mass index (BMI) refined by waist circumference (WC) is not well understood. The purpose of this study was to assess the utility of SKF compared with WC in determination of health risk in the general population. Data from the Canadian Health Measures Survey (cycles 1 and 2; N = 5217) were used. Health outcomes included directly measured blood pressure, cholesterol, glycated haemoglobin, lung function, self-reported health, and chronic conditions. Technical errors of measurements (TEM), sensitivity, and specificity analysis and linear regressions were conducted. Data indicated that TEM for SKF was above the acceptable 5% in most age and sex categories. Sensitivity and specificity of chronic conditions was not improved with the inclusion of SKF in models containing WC (in those aged 45-69 years) and SKF did not explain any additional variance in regression models containing WC. Health outcomes for those in the normal weight and overweight BMI category were significantly worse in those classified as high risk based on WC, whereas SKF did not consistently discriminate risk. In conclusion, evidence-based WC cut-points were shown to identify health risk, particularly in normal weight and overweight individuals. Thus, BMI refined by WC appears to be more appropriate than SKF for assessment of body composition when determining health risk in the general population.
Keywords: body composition; cardiovascular disease; composition corporelle; maladie cardiovasculaire; metabolic syndrome; obesity; obésité; syndrome métabolique.