RT Journal Article SR Electronic T1 Breast feeding in Israel: maternal factors associated with choice and duration. JF Journal of Epidemiology and Community Health JO J Epidemiol Community Health FD BMJ Publishing Group Ltd SP 281 OP 285 DO 10.1136/jech.48.3.281 VO 48 IS 3 A1 Ever-Hadani, P A1 Seidman, D S A1 Manor, O A1 Harlap, S YR 1994 UL http://jech.bmj.com/content/48/3/281.abstract AB STUDY OBJECTIVES--To determine the influence of maternal characteristics on the incidence and duration of breast feeding. DESIGN--All the women who delivered in three obstetric wards within a two year period were surveyed. These three wards cover 93% of all births in the Jerusalem district. Women were interviewed on breast feeding of the previous child on the first or second day post partum by a research nurse. PARTICIPANTS--Altogether 8486 women whose previous pregnancy had resulted in a live born singleton who survived for at least one year. MEASUREMENTS AND MAIN RESULTS--Breast feeding information was linked to demographic and health information from hospital records. Using logistic regression analysis, failure to start breast feeding was best predicted (p < 0.001) by caesarean delivery, infant's birth weight, maternal smoking habits, and mother being non-immigrant. Maternal age (< 24 or > 40 years) and father being an ultraorthodox Jew were also positively (p < 0.05) associated with the decision to breast feed. Long term breast feeding (three months or more) was strongly affected (p < 0.001) by maternal education level, with both women with the fewest and the greatest number of years of schooling more likely to breast feed. A similar association was observed in all ethnic groups. Primipara and grandmultipara (parity > 4), new immigrants, ultraorthodox Jews, and non-smokers breast fed their babies for longer. CONCLUSIONS--The importance of maternal characteristics in relation to breast feeding was shown. Caesarean delivery and the infant's birth weight were strongly related to the decision to breast feed as were the demographic characteristics of mother's age and her country of birth. Education was not related to this decision but was strongly associated with the duration of breast feeding, as was parity. The behavioural characteristics of smoking and being ultraorthodox were related to both the decision to start and the duration of breast feeding. Efforts to encourage breast feeding ought to be targeted during the hospital stay and post partum period towards women identified as being at increased risk.