Background: In Italy recent information about breast-feeding (BF) is lacking

Aim: identify factors influential in duration of BF in a community of north-east of Italy.

Subjects: Consecutively delivered, single newborns who were or were not exclusively breast-fed beyond 3 months and their mothers. Interventions: data, acquired by a 6 month postpartum questionnaire and by neonatal hospital records, related to the 2 groups were compared. A multivariate analysis model by stepwise logistic regression identified the variables that significantly reduced the odds ratio(or) of exclusive BF. Results: Among 284 mothers, 51.4% completed 3 months of BF. The best fitted model which included low (<7 times/day) frequency of nursing during the first month of life(or 0.2 CI 0.08-0.5), early formula supplementation in the nursery(or 0.3 CI 0.2-0.5), ≥ 15 of maternal education(or 0.1 CI 0.03-0.5), regular use of perfumer(or 0.5 CI 0.2-0.9) and breast-oinments (or 0.5 CI 0.2-0.9), blue collar father profession(or 0.4 CI0.2-1.0) and lack of BF orientation during prenatal care(or 0.6 CI 0.3-1.1) had a sensitivity of 79%, specificity of 67%, PPV and NPV of 76% and 70% respectively; ROC area=0.83;(log likelihood=-111.8, goodness of fit test p=0.57)

Conclusions: In our community, the logistic model identified specific risk variables for early weaning that are important to understand in order to extend BF period.