Assessment of overweight and underweight of Bangladeshi adults using waist-to-height ratio: A cross sectional analytic study

Waist-to-Height-Ratio (WHtR) may be an alternate anthropometric index that can overcome the constraint of Body-Mass-Index (BMI) cut-off values for assessing health-risks. To get suitable cut-off values of WHtR, this study evaluated its use as a substitute to BMI cut-off <18.5 and >25 to identify adult undernutrition and overweight respectively. During 2012 a cross-sectional-study was conducted among 650 adult-attendants of the Dhaka-Hospital of icddr,b in Bangladesh. Waist-circumference, height and weight of 260 male and 390 female aged 19-60 years were measured. Sensitivity and specificity of WHtR against BMI<18.5 and BMI ≥ 25 were determined. Significant positive linear correlation was found between WHtR and BMI in males (r = 0.807, p<0.001) and females (r = 0.823, p<0.001). Based on BMI <18.5, area under receiver-operatingcharacteristic- curve (AUC) for WHtR among males was 0.888; 95% CI 0.850-0.927 and females was 0.889; 95% CI 0.855-0.922, and on BMI ≥25 among males it was 0.936; 95% CI 0.890-0.982 and among females it was 0.955; 95% CI 0.933-0.976. WHtR cut-off <0.44 for male and <0.48 for female to identify undernutrition, and ≥0.51 for male and ≥0.53 for female to identify overweight were chosen separately based on highest corresponding Youden-index. These cut-offs of WHtR intimately make out adult nutritional status as defined by BMI.