Theresa L. Osypuk, MS, Dolores Acevedo-Garcia, PhD, S.V. Subramanian, PhD, Mah-Jabeen Soobader, PhD, and Lisa F. Berkman, PhD. School of Public Health, Dept of Society, Human Development & Health, Harvard University, Kresge Building, 7th Floor, 677 Huntington Avenue, Boston, MA 02115, 617-432-1135, tosypuk@hsph.harvard.edu

Black infants are twice as likely as white infants to be born low birthweight (LBW, <2500g). Since only a small proportion of the variance in LBW risk is accounted for by individual-level risk factors, the LBW literature has been criticized for individualizing the problem and ignoring social context factors that influence population rates of LBW. U.S. metropolitan areas exhibit considerable variation in low birthweight rates, yet few analyses have included data beyond one geographic area to explain this metropolitan-level variation. We explore whether and to what extent infant LBW risk is influenced by contextual features of racial inequality across U.S. metropolitan areas (central cities and their suburbs). Using a two-level multilevel statistical model, and after controlling for individual-level demographic, behavioral, and medical risk factors, we test how a racialized, unequal neighborhood and metropolitan context may affect individual infant birthweight among non-Hispanic whites and non-Hispanic blacks. We investigate the average effects of racial disparities on birth outcomes, as well as quantify the variability in these associations across metro areas. We merge individual-level data from the 2000 Detail Natality Dataset, with metropolitan-level racial disparity indicators calculated from Census 2000 SF1 and SF3 file tract level data, as well as with metro-level indicators from other public-use datasets (U.S. Economic Census data, Home Mortgage Discrimination Act data). Metro-level indicators include racial residential segregation (e.g. central city-suburban dissimilarity), and racial disparities in small business ownership, access to home loans, neighborhood homeownership rates, neighborhood median home values, and neighborhood median incomes.

Learning Objectives:

• Evaluate how neighborhood social context and racial inequality affect birth outcome disparities, Appreciate the importance of multilevel statistical modeling to test multilevel hypotheses with multilevel data from different data sources, Describe the patterns of geographic variability in birth disparities throughout the US, Link persistent racial health disparities to racial disparities in other domains of life, including disparities in neighborhood environment, and socioeconomic opportunities

Keywords: Social Inequalities, Health Disparities

Related Web page: www.celebratingdiversity.org

Presenting author's disclosure statement:
I do not have any significant financial interest/arrangement or affiliation with any organization/institution whose products or services are being discussed in this session.