OBJECTIVE: To determine factors associated with early formula feeding in infants whose mothers intend to breastfeed. DESIGN: Demographic and infant feeding data for all inborns were collected prospectively at a military tertiary care facility. Our sample was limited to liveborns who received exclusively Level I nursery care. Feeding intention was noted prior to delivery and feeding practice was documented at discharge. Intentions and practices were ranked on a 1 to 5 scale defining a spectrum from exclusive formula feeding to exclusive breastfeeding. Breastfeeding-intended infants were defined as attritted if the practice at discharge was at least 50% formula feeding. Multiple logistic regression identified factors (demographic and medical) related to breastfeeding attrition. RESULTS: During a 10 month period 1038 infants were analyzed. 69.3% were intended to breastfeed; however, only 46.7% were exclusively breastfed and 12.5% were predominately breastfeeding at hospital discharge (p < 0.0001). 17.8% of infants were exposed to breastfeeding attrition. An initial regression analysis found two factors related to attrition: multiple gestation and maternal pregnancy weight gain. Factors found unrelated included mode of delivery, epidural anesthesia, and socioeconomic factors. A second analysis identified birth weight and pre-eclampsia as related to attrition; however, infant hypoglycemia and maternal diabetes were not. Hospitalization days also became significant.CONCLUSIONS: A substantial breastfeeding attrition occurs during the immediate post-partum hospitalization; this attrition contributes to a failure to achieve Healthy People 2000 goals for breastfeeding in our population (goal: 75% breastfeeding at hospital discharge). The low incidences of multiple gestation and pre-eclampsia in this healthy term population result in a negligible consequence for the larger population. Mothers with pre-eclampsia may have difficulty feeding their newborns. Extended hospital stay in this population is reflective of a maternal or neonatal complication. The identification of increased birth weight as a risk factor is not readily explained, but could be due to subtle mechanical or endocrine hindrances. These factors identify a sub-population of infants at higher risk for breastfeeding attrition.