Associations of hospital staff training and policies with early breastfeeding practices.

BACKGROUND: In 2009, the Centers for Disease Control and Prevention implemented the Maternity Practices in Infant Nutrition and Care (mPINC) survey in all US birth facilities to assess breastfeeding-related maternity practices. Maternity practices and hospital policies are known to influence breastfeeding, and Alabama breastfeeding rates are very low. OBJECTIVE: Our objective was to assess whether staff training and structural-organizational aspects of care, such as policies, were associated with infants' breastfeeding behaviors 24 to 48 hours postpartum. METHODS: We linked 2009 mPINC data from 48 Alabama hospitals with birth certificate and newborn screening databases. We used data collected 24 to 48 hours postpartum to classify 41 536 healthy, term, singleton infants as breastfed (any breast milk) or completely formula fed and examined associations with hospitals' mPINC scores in comparison with the state mean. We conducted multilevel analyses to assess infants' likelihood of being breastfed if their birth hospital scores were lower versus at least equal to the Alabama mean, accounting for hospital clustering, demographics, payment method, and prenatal care. RESULTS: The odds of breastfeeding were greater in hospitals with a higher-than-state-mean score on the following: new employees' breastfeeding education, nurses' receipt of breastfeeding education in the past year, prenatal breastfeeding classes offered, having a lactation coordinator, and having a written breastfeeding policy. The number of recommended elements included in hospitals' written breastfeeding policies was positively associated with newborn breastfeeding rates. CONCLUSION: Educating hospital staff to improve breastfeeding-related knowledge, attitudes, and skills; implementing a written hospital breastfeeding policy; and ensuring continuity of prenatal and postnatal breastfeeding education and support may improve newborn breastfeeding rates.