LONGITUDINAL-STUDY OF THE MOTHER AND CHIL D POPULATION IN AN URBAN REGION OF SOUTHERN BRAZIL, 1993 - METHODOLOGICAL ASPECTS AND PRELIMINARY-RESULTS

All babies born in the hospitals of the city of Pelotas, Brazil, in 19 82 were studied soon after delivery and followed up prospectively duri ng the first years of their fives. in 1993, this study was repeated wi th a similar methodology, with the aim of assessing eventual changes i n the level of maternal and child health. All five maternity hospitals in the city were visited daily and the 5,304 babies born included in the study. They were weighed and measured, and their gestational age w as assessed using the Dubowitz method. Their mothers were examined and interviewed regarding a large number of risk factors. The mortality o f these children was studied through the surveillance of all hospitals , cemeteries and death registries, and all hospital admissions were al so recorded. Two nested case-control studies were carried out to asses s risk factors for mortality and hospital morbidity. A systematic samp le of 655 children were examined at home at one and three months of ag e, and these infants, as well as another sample of 805 children includ ing all low-birthweight babies were also examined at the ages of six a nd twelve months. Their psychomotor development was also assessed. Los ses to follow-up were only 6.6% at twelve months. Relative to the 1982 indicators, perinatal mortality fell by about 30% and infant mortalit y by almost 50%. The median duration of breastfeeding increased from 3 .1 to 4.0 months. On the other hand, there was little change in the pr evalences of low birthweight or of length for age at twelve months. Th e article that refers this abstract describes the methodology of the s tudy and forthcoming publications will present detailed results.