A prospective study was undertaken to identify possible factors related to the duration of breast feeding. Two hundred and thirty-eight mothers who had delivered normal single babies with birth weights greater than 2.5 kg and had initiated breast feeding were randomly selected at the maternity hospital, Hospital de Clinicas de Porto Alegre, Brazil, and followed by mail questionnaires until termination of breast feeding, or until the end of the first year. If no reply was received, telephone contact or home visits were made. The group of mothers who stopped breast feeding prior to the end of the third month was compared with those who extended breast feeding beyond three months with respect to socioeconomic, biological, environmental, medical and psychological factors. The variables with a significant coefficient of association with early termination of breast feeding were maternal education, past experience with breast feeding, help of a maid, help with housework provided by a relative, breast feeding orientation during prenatal care and encouragement from the husband. These factors act simultaneously, with interactions among them.

An individual educational session in the maternity ward about infant sleep position significantly increased the prevalence of supine position for sleeping in the infant's third month. Nevertheless, the intervention was not sufficient to guarantee that most mothers would put their infants to sleep in the recommended position.

Resumo O objetivo foi avaliar a implantação da Rede Amamenta Brasil e seu impacto sobre indicadores de aleitamento materno (AM

To test the hypothesis that the low socioeconomic population living is shanty towns in Porto Alegre presents different levels of poverty which are reflected on its health status, a cross-sectional study was designed involving 477 families living in Vila Grande Cruzeiro, Porto Alegre, Brazil. The poverty level of the families was measured by using an instrument specifically designed for poor urban populations. Children from families living in extreme poverty (poorest quartile) were found to have higher infant mortality rate, lower birth weights, more hospitalizations, and higher malnutrition rates, in addition to belonging to more numerous families. Thus, the shanty town population of Porto Alegre is not homogeneous, and priority should be given to the more vulnerable subgroups. Diagnosis