Back to Annual Meeting

Back to Annual Meeting

APHA Scientific Session and Event Listing

Recorded presentation

Michael D. Kogan, PhD, Office of Data and Program Development, HRSA/ Maternal and Child Health Bureau, 5600 Fishers Lane, Room 18-41, Rockville, MD 20857, 301-443-3145, mkogan@hrsa.gov, Gopal K. Singh, PhD, Office of Data and Program Development, HRSA/Maternal and Child Health Bureau, 5600 Fishers Lane, Room 18-41, Rockville, MD 20857, Deborah L. Dee, MPH, School of Public Health / Dept. of Maternal and Child Heatlh, University of North Carolina, Rosenau Hall, CB 7400, Chapel Hill, NC 27599, and Laurence Grummer-Strawn, PhD, National Center for Chronic Disease Prevention and Health Promotion, Centers for Disease Control and Prevention, Koger Bldg., Atlanta, GA 30341.

Research indicates that breastfeeding rates vary by demographic and socioeconomic factors, and state of residence. However, no study has yet examined whether state variations are diminished after multivariate adjustment. We used the 2003-2004 National Survey of Children's Health, a nationally representative study of 102,000 children that can produce state estimates, to examine this question. This study was limited to children 0-5 years old. We calculated both the unadjusted and adjusted prevalences (from the predicted marginals using logistic regressions) both for ever breastfed, and breastfed at least six months. Logistic regression was used to control for poverty level, race/ethnicity, gender, family structure, language/immigration status, maternal health status, maternal mental health, maternal exercise, and household smoking. Unadjusted prevalence rates among states for ever breastfed ranged from 45 - 88%, while the range was 17 – 55% at 6 months. After covariate adjustments, the state prevalence rates range diminished to 54 – 85% for ever breastfed, and 22 – 54% for breastfeeding at least 6 months. Multivariate analysis indicated that after controlling for individual explanatory factors, where a child lived still had a strong association with breastfeeding behavior. The odds of not being breastfed were 2 – 4 times greater for children in most southern states compared to Washington State, which had the highest rate. Smaller disparities existed for most New England, and mid-Atlantic states, compared to Washington. Findings were similar for breastfeeding at six months. Demographic factors do not fully explain breastfeeding rates among states, indicating that geographically targeted programs would be appropriate.

Learning Objectives:

• Identify states in the US with low breastfeeding rates after adjusting for other factors.
• Describe other risk factors for not breastfeeding.
• Understand how to obtain adjusted prevalence rates.

Presenting author's disclosure statement:

Any relevant financial relationships? No

Recorded presentation