OBJECTIVE: This study compared the classification of the nutritional status of preschoolers according to three different growth curves. METHODS: This cross-sectional study included 350 children aged 2 to 5 years from public and private institutions in the county of Viçosa (MG). The following anthropometric indices were obtained from weight and height measurements: weight-for-age, weight-for-height, body mass index-for-age and height-for-age. The classification of underweight (below -1Z-score), well nourished and overweight (above +1Z-score) was done according to National Center for Health Statistics 1977, Centers for Disease Control and Prevention 2000 and World Health Organization 2006 curves. Kappa statistics was used to evaluate conformity among curves. The National Center for Health Statistics was considered, which is already recommended by the Ministry of Health, and the World Health Organization, which is currently recommended as the reference standard, to compare the sensitivity and specificity values, positive predictive value and negative predictive value. RESULTS: The prevalence of underweight was higher when the Centers for Disease Control and Prevention curves were used. On the other hand, regarding overweight and stunting, the World Health Organization curves resulted in the highest prevalence. Both the World Health Organization and the National Center for Health Statistics curves showed closest sensitivity and positive predictive values, as well as better correlation (k>0.61), when compared with the Centers for Disease Control and Prevention curves. CONCLUSION: There are differences in the prevalence of nutritional deficits, and the World Health Organization curves were the ones which showed the highest sensitivities. The use of these curves by health professionals enables a more adequate evaluation of growth during childhood, since it regards a current growing pattern which is also recommended for international use.
Anthropometry; Nutrition assessment; Nutritional status; Child; preschool
