[Evaluation of the use of the 10th percentile of weight for age as a cut point for detection of children under nutritional risk]

OBJECTIVE: To evaluate the use of the 10th percentile of weight for age as a cut off point for detection of children under nutritional risk, especially for programs of alimentary supplementation. METHODS: 841 children with age between 10 days and 60 months were studied in a primary health care center located in the periphery of Ribeirão Preto, SP, Brazil. It was a cross-sectional study that included age, sex, weight and height, later computing the z scores of weight for age, height for age and weight for height, on the basis of the data of the NCHS. Through 2x2 type tables, where the results in relation with the 10th percentile (above or below this cut off point) and other indices were compared (above or below - 2 z scores), the capacity of 10th percentile of weight for age to detect children with various anthropometric deficits was estimated. RESULTS: The false negative results were always very low, between 0 and 1%, in opposition to the false positive results, that varied from 76.3 to 90.5%. The values of sensitivity/specificity have been 100/85.9%; 93.3/83.8%; 82.9/85.7%, respectively for deficits of weight for age, weight for height and height for age. CONCLUSIONS: The 10th percentile of weight for age was found adequate for population screening of children with weight for the age and weight for the height deficits (high sensitivity), but these children must be better evaluated later on (too many false positive results). For height for age deficits, the 10th percentile resulted inadequate.