Ravi K. Sharma, PhD, Behavioral and Community Health Sciences, University of Pittsburgh, 130 DeSoto St., Room 228, Pittsburgh, PA 15261, (412) 624-3615, rks1946@pitt.edu and Jeanine Buchanich, MPH, Biostatistics, Uinversity of Pittsburgh, 130 Desoto Street, Pittsburgh, PA 15261.

A number of studies have shown that ecological settings are related to the variety of health outcomes including health related behaviors. Social epidemiologists have become increasing interested in “neighborhoods or place effects” and their relationships to health disparities among social groups. This paper examines the social processes underlying the spatial variations in birthweight across US counties. We focus on three social mechanisms: (a) concentrated disadvantage (2) health services and (3) social stress. We specifically account for the spatial nature of the data by employing spatial statistical analysis tools. We used Area Resources File that is a US county database of vital and health statistics and census data. The spatial autocorrelation was highly significant. This shows the null hypothesis of spatial randomness can be rejected and there is evidence of significant clustering of LBW rates. A spatial modeling approach is therefore used. Spatial regression model is used to incorporate spatial effects into the models. The overall model fitness is reasonable with adjusted r-square of .59 respectively. Of the two measures of structural disadvantage, index of concentration at the extremes (ICE) is by far the most important. The next variable in order of importance is percentage of African Americans in the county which is positively related to LBW rates. Violent crime rate is positively and significantly related to LBW rate. Health resources are positively and significantly relatively to LBW rates but coefficient value is relatively small.

Learning Objectives:

1.Recognize what spatial effects are and how they influence low birth weight
• Describe social processed that account for spatial variations in low birth weight
• Identify appropriate statistical model for analyzing spatially correlated data

Keywords: Low Birthweight, Measuring Social Inequality

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.