The relationship between the timing of maternal postpartum hospital discharge and breastfeeding.

This study examines the association between discharge timing and breastfeeding, controlling for demographic, economic, and health factors that influence both. Using the 1988 National Maternal and Infant Health database, maximum-likelihood modeling was used to generate 2-equation models to estimate both outcomes simultaneously. The joint estimation of breast-feeding and discharge timing demonstrates that mothers who spent 1 night in the hospital were more likely to breastfeed than mothers who spent 2 or 3 nights in the hospital (OR = 1.96, 95% C.I. 1.86-2.03). Furthermore, the model altered the association between breastfeeding and age, breastfeeding advice, and prenatal classes. In 1988, when decisions were a function of clinical judgments about readiness for discharge, early discharge was associated with an increased likelihood of breastfeeding. Instead of basing discharge solely on predetermined rules, derived largely from financial criteria, attention to the decision-making process should be an important part of newborn discharge policies.