NEIGHBORHOOD RISK-FACTORS FOR LOW-BIRTH-WEIGHT IN BALTIMORE - A MULTILEVEL ANALYSIS

Objectives. Past research on low birthweight has focused on individual -level risk factors. We sought to assess the contribution of macroleve l social factors by using census tract-level data on social stratifica tion, community empowerment, and environmental stressors. Methods. Cen sus tract-level information on social risk was linked to blah certific ate records from Baltimore, Md, for the period 1985 through 1989, Indi vidual-level factors included maternal education, maternal age, medica l assistance health insurance (Medicaid), and trimester of prenatal ca re initiation. Methods of multilevel modeling using two-stage regressi on analyses were employed. Results. Macrolevel factors had both direct associations and interactions with low birthweight. All individual ri sk factors showed interaction with macroleveI variables; that is, indi vidual-level risk factors for low birthweight behaved differently depe nding upon the characteristics of the neighborhood of residence, For e xample, women living in high-risk neighborhoods benefited less from pr enatal care than did women living in lower-risk neighborhoods. Conclus ions. Multilevel modeling is an important tool that allows simultaneou s study of macro- and individual-level risk factors. Multilevel analys es should play a larger role in the formulation of public health polic ies.