Sensitivity and specificity of waist circumference as a single screening tool for identification of overweight and obesity among Malaysian adults

Generalised obesity and central obesity are risk factors for Type II diabetes mellitus and cardiovascular diseases. Waist circumference (WC) has been suggested as a single screening tool for identification of overweight or obese subjects in lieu of the body mass index (BMI) for weight management in public health program. Currently, the recommended waist circumference cut-off points of ? 94cm for men and ?80cm for women (waist action level 1) and ? 102cm for men and ? 88cm for women (waist action level 2) used for identification of overweight and obesity are based on studies in Caucasian populations. The objective of this study was to assess the sensitivity and specificity of the recommended waist action levels, and to determine optimal WC cut-off points for identification of overweight or obesity with central fat distribution based on BMI for Malaysian adults. Data from 32,773 subjects (14,982 men and 17,791 women) aged 18 and above who participated in the Third National Health Morbidity Survey in 2006 were analysed. Sensitivity and specificity of WC at waist action level 1 were 48.3% and 97.5% for menand 84.2% and 80.6% for women when compared to the cut-off points based on BMI ?25kg/m2. At waist action level 2, sensitivity and specificity were 52.4% and 98.0% for men, and 79.2% and 85.4% for women when compared with the cut-off points based on BMI (?30 kg/m2). Receiver operating characteristic analyses showed that the appropriate screening cut-off points for WC to identify subjects with overweight (?25kg/m2) was 86.0cm (sensitivity=83.6%, specificity=82.5%) for men, and 79.1cm (sensitivity=85.0%, specificity=79.5%) for women. Waist circumference cut-off points to identify obese subjects (BMI?30 kg/m2) was 93.2cm (sensitivity=86.5%, specificity=85.7%) for men and 85.2cm (sensitivity=77.9%, specificity=78.0%) for women. Our findings demonstrated that the current recommended waist circumference cut-off points have low sensitivity for identification of overweight and obesity in men. We suggest that these newly identified cut-off points be considered