Concordance between national and international bodymass index as a predictor of body composition in premenarcheal and menarcheal adolescents.

Use of national and international references for the diagnosis of nutritional status is controversial. Concordance between national and international body mass index as predictors of body composition in 364 premenarcheal and menarcheal female adolescents (ages 10-15), classified according to occurrence of menarche, that were part of the database of the bioanthropology, physical activity and health unit, were evaluated. This study compares the capacity of body mass index (BMI) to predict body composition, diagnosed by upper arm fat area (UFA) and/or upper arm muscle area (UMA), using national reference (ENCDH) vs International (WHO, 2007). The weighted Kappa coefficient was applied to evaluate the concordance between BMI by national and international references, as well as to assess the precision of BMI by means of sensibility and specificity. Additionally, diagnostic verisimilitude ratio was estimated to measure the efficiency of both references in the classification of body composition. The weighted Kappa showed greater concordance in UFA (0.64) versus UMA (0.51). The presence of menarche increased the concordances: UFA (0.63) and UMA (0.59) with respect to premenarcheal girls: UFA (0.46) and UMA (0.35). The positive and negative diagnostic likelihood ratios were consistently greater than one, for fat and muscle area, especially when using ENCDH reference. The findings suggest that prevalence of deficit or excess in UMA was more sensitive with the BMI_ ENCDH than with the BMI_ WHO, in premenarcheal girls. On the other hand, WHO was more sensitive to predict UFA excess in the same group. Key words: Menarche, body composition, weighted Kappa coefficient, diagnostic likelihood ratios, international and national growth references, Venezuela.