OBJECTIVE: A cohort prospective study of newborns in the city of Pelotas, Southern Brazil, between September 2002 and May 2003, which aims at evaluating the perception of mothers as to fatherly support and its influence in breastfeeding duration. METHODS: The initial population included 2,741 babies, and a random and representative sample of 30% was followed-up on the first, third, and sixth months, with a significance level of 95% and statistical power of 80% to detect a relative risk of 2.0. Univariate and multivariate analyses were applied. Variables with p<0.05 were considered as significantly associated with the outcome. RESULTS: In the first month, approximately 10% of infants were not breastfed. Low paternal schooling and lack of support during breastfeeding were associated with weaning in the first month. In the third month, a strong association between weaning and lack of paternal support was verified. The fact that the mother no longer lived with her partner and the number of years in school were also associated with the outcome. In the sixth month, no correlation was found between paternal variables and breastfeeding. CONCLUSIONS: The present study could be useful as a reference to future public health policies as well as an incentive to insert the paternal figure in the prenatal, labor, and postdelivery care.
father; breast feeding; epidemiology; weaning; infants