We examined the variation between 126 New York hospitals in formula supplementation among breastfed infants after adjusting for socioeconomic, maternal, and infant factors and stratifying by level of perinatal care.


We used 2014 birth certificate data for 160 911 breastfed infants to calculate hospital-specific formula supplementation percentages by using multivariable hierarchical logistic regression models.


Formula supplementation percentages varied widely among hospitals, from 2.3% to 98.3%, and was lower among level 1 hospitals (18.2%) than higher-level hospitals (50.6%–57.0%). Significant disparities in supplementation were noted for race and ethnicity (adjusted odds ratios [aORs] were 1.54–2.05 for African Americans, 1.85–2.74 for Asian Americans, and 1.25–2.16 for Hispanics, compared with whites), maternal education (aORs were 2.01–2.95 for ≤12th grade, 1.74–1.85 for high school or general education development, and 1.18–1.28 for some college or a college degree, compared with a Master’s degree), and insurance coverage (aOR was 1.27–1.60 for Medicaid insurance versus other). Formula supplementation was higher among mothers who smoked, had a cesarean delivery, or diabetes. At all 4 levels of perinatal care, there were exemplar hospitals that met the HealthyPeople 2020 supplementation goal of ≤14.2%. After adjusting for individual risk factors, the hospital-specific, risk-adjusted supplemental formula percentages still revealed a wide variation.


A better understanding of the exemplar hospitals could inform future efforts to improve maternity care practices and breastfeeding support to reduce unnecessary formula supplementation, reduce disparities, increase exclusive breastfeeding and breastfeeding duration, and improve maternal and child health outcomes.

You do not currently have access to this content.