Prediction of obesity in pubertal school girls using various anthropometric measurements

ABSTRACT

The aim of the work was to choose a sensitive indicator to diagnose overweight and obese girls other than the currently used body mass index [BMI]. The study has covered 123 school girls at age range 11-16 years. Anthropometric measurements were done including weight, height, sitting-height and span. Circumferences that included head, chest, mid-upper-arm, thigh, waist and hip were done. Six sites of fat thickness triceps, biceps, subscapluaris, thigh, abdominal and suprailiac were all measured. BMI of all girls ranged between 15.60-44.0. A significant increase in weight, sittingeight, trceps, biceps, subscapulair, thigh, abdominal and suprailiac skinfolds thickness and in head, mid upper arm, chest, waist, thigh and hip circumference with the increase in BMI was show. Stepwise regression analysis was applied to test for the extent of contribution of the variables measured in the study to the level of BMI. The results showed that only mid upper arm circumference, suprailiac skinfold, chst circumference and thigh circumference showed correlation with BML. Logistric regression analysis was accordingly performed to identify ignificant factors for predicting obesity. Of all the measurements entered into the logistic regression model only mid arm-circumference and suprailiac skin fold thickness were selected by the model as the most important factors to predict obesity with sensitivity=85.2% and specificity=87.6%