Validity of single-weight measurements to predict current malnutrition and mortality in children.

In this cross-sectional study of a random cluster sample of 4238 rural Zairian children aged 0-5 y, we assessed underweight and wasting, defined as weight-for-age < 75%, and weight-for-height < 80% of the U.S. National Center for Health Statistics reference median, respectively. We determined the diagnostic validity of underweight and wasting for protein-energy malnutrition, taking a low arm circumference and clinical signs of muscle loss as criteria. Both underweight and wasting had low sensitivity in recognizing low arm circumference, any clinical muscle loss and even severe marasmus, especially in the weaning period of 12-30 mos. Receiver operating characteristic (ROC) analysis showed that the diagnostic validity of weight-for-height can be improved by using a cutoff for wasting at Z-score -0.75 instead of Z-score -2 or 80% of reference median. ROC analysis of 30-mo mortality revealed a poor prognostic validity of weight-for-height and weight-for-age and better performances of arm circumference (cm) and of age. These data suggest that nutritional intervention programs targeted at wasted or underweight children can have only a limited effect on the prevalence of protein-energy malnutrition in the community or on the long-term mortality associated with it.