OBJECTIVE: To investigate determinants of early weaning in Northwest of Campinas, Sao Paulo state, Brazil. METHODS: During the National Immunization Day on June 8th, 2001, a survey was conducted among 385 children less than 2 years old (10% of attendance). The responsible for the child was interviewed about social, demographic, and health assistance characteristics, age, and reasons for the introduction of other foods to the child and its weaning. The complete interruption of breast-feeding before the age of 6 months was considered as early weaning. The associations between weaning and cause and/or risk factors were investigated by bivariate correlation. To check if the association was maintained, logistic regression was run to identify the best predictors for early weaning. The proportions' difference test was performed to identify alleged reasons for early weaning. RESULTS: Early weaning affected 63.6% of the surveyed children. The education level of the mother was associated to early weaning, according to t-test (p = 0.04). There was no association between early weaning and mother's age, marital status or occupation, family density, income, kind of health assistance, and gender of the child. Logistic regression indicated time of artificial milk or formula introduction as predictor for early weaning (p<0.0001). The alleged reasons for early weaning were: decrease of maternal milk production, baby's refusal to breast-feed, mother's occupation, pain during breast-feeding, breast injuries, and sickness of either mother or baby. CONCLUSION: The results indicated the early introduction of milk or formulas as a predictor for early weaning.
breast feeding; child welfire; early weaning; milk human
