Background: Body Mass Index (BMI) is commonly used to determine overweightness and obesity in epidemiological studies regardless of the sex and age of the subjects. Golden Standards were presented by the World Health Organization to estimate obesity by measuring body fat percentiles. The primary purpose of this study was to estimate the accuracy of the BMI by Cut-points of gold standard.

 Materials and Methods: This is a cross-sectional and descriptive study. Measuring body composition was done by Bioelectrical Impedance Analysis (BIA) method. Accuracy of the BIA method is documented in comparison with more complicated methods Diagnosis performance to estimate the accuracy of Body Mass Index was based on measuring specificity, sensitivity, percentile of the power of positive anticipation, and percentile of the power of negative anticipation to diagnose obesity, with due attention to the cut-points of world health organization gold standards and according to sex and age ranking. Data analysis was performed by T-test, Chi-Square and Roc curve. Two Roc curves were compared by Honely formule and regression analysis.

Results: According to the cut-point of gold standard, statistical findings showed that the amounts of sensitivity and specifity were 66% and 90.5% for all persons, respectively. With due attention to data analysis, the level of sensitivity was 50.2% to 73.3% in males and 28.9% to 69.7% in females. Also, with respect to age ranking, percentile of the power of negatire anticipation was reported between 26.7% to 63.6% in males and between 28.9% to 69.7% in females.

Conclusion: Body Mass Index could not accurately classify people in obesity and overweightness groups. The power of BMI for classifying obese and overweight people decreases with increasing age. Overall, comparing BMI and the cut-points of gold standard showed the medium effect of this index in classifying obese and overweight persons. The appropriate cut-point to diagnose obesity was 27.8.