Razi Journal of Medical Sciences

  Background : Body mass index (BMI) is one of the most popular anthropometric techniques to identify obesity. The purpose of this study was to evaluate the sensitivity and specificity of BMI percentiles of Center of Disease Control (CDC2000) in determining obesity in preschool children in comparison to real obesity ( based on body fat index FMI) and compare the status of overweight and obesity based on this two index.

  Methods : Height , weight and triceps skinfold thickness was measured in 603 preschool children in Tehran. BMI calculated by dividing weight ( in kilograms) to squared height ( in meters) and FMI calculated by dividing the amount of fat mass ( in kilograms) to squared height (in meters). BMI≥95th percentile of CDC2000 standards were used as the criteria of obesity and the 85th

  Result : Based on BMI prevalence of obesity and overweight among preschool children were 5.1 and 9.8 , respectively moreover the prevalence of overweight and obesity was higher in boys than girls. The difference between two sexes was not significant (p=0.211, p=0.834). About 46% of really obese children and 4% of the non-obese children, who were not obese according to FMI , were diagnosed as obese by BMI. Moreover, using FMI as a real measure of obesity, sensitivity and specificity of BMI 90 percentile appeared to be 73% and 97.4%, respectively.

  Conclusion : Despite having high specificity, BMI had low sensitivity and may have low efficiency in diagnosing childhood obesity. It seems that comparing to BMI FMI is a better indicator to determine obesity in children. However , the lack of precise standard references for FMI makes further studies in this area necessary .