The "sécurité naissance - naître ensemble" network in the Pays de la Loire region of France, with 24 maternity units, conducted a survey of nursing women to determine the duration of breastfeeding and to study the factors influencing this practice. Phone calls were made to mothers at 1 month, 2 months, 3 months, and 6 months or until weaning. The statistical methods used were the Kaplan-Meier survival curve method with the calculation of the median duration of breastfeeding (exclusive or partial) in weeks, and the Cox multivariate model. The hazard ratio or relative risk (RR) was used in univariate analysis, and the adjusted odds ratio (aOR) in multivariate analysis (>1 related to short duration, and<1 related to long duration). Fifteen maternity units participated, and 239 women were followed. The median duration of breastfeeding was 15 weeks (3.5 months) with 10% discontinuing the first month and 25% continuing breastfeeding at 6 months. The final model of risk of duration contains the following variables: maternal age less than 30 years (aOR=1.54; 1.08-2.19), the time the decision was made before pregnancy (aOR=0.56; 0.38-0.83), BMI greater or equal to 30 (aOR=2.12 (1.18-3.79), and for the first month, the administration of bottle feeding or complements (aOR=2.51; 1.62-3.89), the baby's need to be stimulated to feed (aOR=5.07; 3.24-7.90), breastfeeding on demand (aOR=0.56; 0.39-0.82), and problems breastfeeding (aOR=1.45; 1.005-2.09). This survey to determine the duration of breastfeeding in western France, a region with low initiation rates, duration appears to be relatively long compared to other studies. The factors influencing breastfeeding duration were for some non-modifiable (age, time of the decision, etc.), and for others assistance by a professional (supplements, need for stimulation, difficulty sucking, breastfeeding on demand, etc.). Criteria for identifying difficulties are presented.
Copyright © 2012. Published by Elsevier SAS.