The Relationships between Maternity Practices and Exclusive Breastfeeding Rates in the Hospital

Background While breastfeeding overall is beneficial, exclusive breastfeeding (EBF) has been associated with an even greater reduction in infant morbidity and mortality. Research has shown breastfeeding rates in the hospital are influenced by maternity practices. The Centers for Disease Control and Prevention (CDC) launched the Maternity Practices in Infant Nutrition and Care survey (mPINC) in 2007. To date there is limited research on the association between the variables measured in the mPINC survey and hospital EBF rates. Study Objective The purpose of this study was to examine relationships between maternity practices measured in mPINC and EBF rates, and build a model to predict EBF rates in the hospital. We also explored whether the Baby Friendly (BF) designation would improve our ability to predict EBF rates over and above a selected group of maternity practices measured by mPINC. Methods Data on EBF rates was obtained from the Joint Commission, a leading agency in hospital accreditation, and merged with the 2013 mPINC survey from the CDC. Census data for hospitals was applied using block groups from the American Community Survey conducted by the United States Census Bureau. The data set contains 983 hospitals nationwide, 82 are BF. Relationships between continuous variables measured in mPINC and EBF were explored with linear regression. ANOVA with post‐hoc Tukey corrected pair‐wise comparisons of means was used to compare EBF rates between levels of categorical variables. Models with collections of variables from mPINC were fit in an attempt to find a model that explained a large proportion of the variability in EBF. It was not possible to fit a model with all variables included because of the large amount of missing data. Adjusted R‐squared was used to rank models. Results Most of the maternity practices we examined were significantly associated with EBF rates, p <0.001. Some of the strongest associations were with the following variables (p<0.001): skin‐to‐skin contact and routine newborn procedures conducted while mother is holding the infant following an uncomplicated cesarean birth; breastfed infants not routinely taken to the nursery or separate area for transition; and nurses/birth attendants with at least 4 hours of breastfeeding education as a new employee. Each of these factors independently accounted for approximately 11–19% higher rates of EBF. A model including BF and a collection of demographic variables explained 25% of the variability in EBF rates. A second model using a collection of mPINC and demographic variables increased the variance explained to 42%. When BF designation was added into the second model, it did not improve our ability to predict EBF rates. Conclusions While most maternity practices measured through the mPINC survey are significantly associated with EBF rates, no one variable was able to explain a majority of the variation in EBF rates. However, we were able to rank variables with respect to their associations with EBF rates and build a model with a collection of variables that strongly predicts EBF rates in the hospital setting. The maternity practice variables that were included in the model were found to be important regardless of the BF designation. Support or Funding Information This research is supported by the University of Wisconsin‐ Madison College of Agriculture and the Wisconsin Alumni Research Foundation.