A study was made to verify which of five variables, age, parity, spacing of pregnancies, family income per month "per capita" and schooling had the greatest effect on the average length of the period of breast-feeding and, for each variable, which was the critical point. Statistical analysis according to Morgan and Sonquist statistical method permitted the recognition of the characteristic of women with greatest risk of premature weaning. The use of the concepts "premature-weaning risk" and "accumulated child risk" was proposed.
Breast feeding; Weaning; Health education
