Document Type



In order to improve assessment of risk of poor postnatal health and growth and developmental outcomes, groups of infants were defined using low birth weight status (LBW, 2,500 g) and an index of body proportions at birth (Rohrer’s Index, RI = weight X 100/length3). The sample, derived from a prospective study of infant-feeding practices, health and survival in the Philippines, consisted of 1971 mother-infant pairs from randomly selected urban and rural communities of metro Cebu. Multinomial logistic regressions assessed how the probability of belonging to a particular risk group was influenced by a set of maternal and infant biological and behavioral variables. Increased risk of falling into a LBW risk group was associated with low gestational age (GA), low maternal stature, and primiparity. Proportionate and disproportionately grown LBW infants could be differentiated by the magnitude of the effects of gestational age, mother’s height, and primiparity (all stronger determinants of proportionate small size) and rural residence, low maternal arm fat area, and smoking during pregnancy (all of which increased risk of being proportionately small but had no effect on risk of being disproportionate). The analyses show that the determinants of weight and proportionality at birth differ markedly. RI, by capturing information about patterns of prenatal growth that contribute to proportionality, should prove to be an important predictor of postnatal outcomes when used along with birth weight.