The 131st Annual Meeting (November 15-19, 2003) of APHA

Abstract #73186

Jessica Burke, PhD, Department of Population and Family Health Sciences, Johns Hopkins School of Public Health, 615 North Wolfe Street, E4010, Baltimore, MD 21205, 410-614=6127, jburke@jhsph.edu, Patricia O'Campo, PhD, Department of Population and Family Health Sciences, Johns Hopkins University, 615 N. Wolfe Street, Baltmore, MD 21205, and Isabelle Horon, DrPh, Vital Statistics Administration, Maryland Department of Health and Mental Hygiene, 201 W. Preston St., Baltimore, MD 21201.

Multilevel modeling is a tool for studying the social and environmental determinants of health. These models have increasingly been used in the field of maternal and child health. These tools can be used to gain a better understanding of how multiple levels of influence are associated with racial inequalities and adverse pregnancy outcomes. As part of a larger project concerned with identification of neighborhood level factors contributing to pregnancy outcomes as well as the policy implications of findings from this type of research, four states (Maryland, Michigan, North Carolina and Pennsylvania) have analyzed data on neighborhoods and preterm birth. This presentation will focus on the theoretical considerations, data needs and analytic issues involved in conducting multilevel modeling on state level data on perinatal health. We will specifically describe issues related to geocoding individual-level data, sources of neighborhood-level information, issues of confidentiality, the software necessary to carry out such analyses and the development of models that best inform policy making for reducing inequalities in health. This discussion will complement presentations on results from several of the state multilevel analyses.

Learning Objectives:

• The objective is to increase the knowledge of statisticians, epidemiologist and public health analysts in the utility of multi-level analysis of individual and contextual characteristics in identifying determinants of preterm delivery and other areas of health.

Keywords: Pregnancy Outcomes, Epidemiology

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.