The reliability and validity of using clothing size as a proxy for waist circumference measurement in adults.

Waist circumference (WC) is a useful tool for predicting health risk, but its use at the population level remains uncertain. This study examined (i) the ability of participants to report their current WC without actual measurement, (ii) the accuracy of reporting WC using self-reported or self-measured WC, and (iii) the reliability and validity of using clothing size (CS) as a proxy to predict WC. Men and women (n = 293), aged 18-80 years, were randomized to either self-report WC or self-measure WC groups. Both completed 2 telephone surveys and their WC was professionally measured. Predictive equations were then developed to determine whether CS could be used as a proxy for WC. Only 66% of participants reported their current WC, although this was underreported (p < 0.05) compared with professionally measured WC. Professionally measured WC correlated strongly with CS for men (r = 0.8; p < 0.01) and women (r = 0.78; p < 0.05), respectively. While predicted WC demonstrated good agreement for men (κ = 0.82) with respect to classifying individuals at increased health risk, this was attenuated in women (κ = 0.6). Due to the fact that only 66% of participants know their current WC and that both self-report WC and self-measure WC groups underreport actual WC, a reliable and valid proxy for WC is needed. CS presents a reliable and feasible means of obtaining an estimate of WC at the population level in adults and predicting the percentage of the population at increased health risk.