作者: Jay S. Kaufman , Nancy Dole , David A. Savitz , Amy H. Herring
DOI: 10.1016/S1047-2797(02)00480-5
关键词:
摘要: PURPOSE: We demonstrate modeling of community-level socioeconomic influences on risk preterm birth (< 37 weeks gestation) in the Pregnancy, Infection, and Nutrition (PIN) Study. METHODS: Community-level information from US Census was linked to 930 White 817 African-American (Black) participants a prospective cohort central North Carolina through geocoded addresses, providing 123 census tracts with individual-level data for multi-level statistical analyses. RESULTS: Preterm delivery experienced by 12.1% Black 10.4% participants. No appreciable aggregation community discernable women. For women, random-coefficient logistic regression tract-specific prevalence estimates ranged 10.1% 14.5%, “shrunk” observed prevalences 0% 100%. Adding tract-level variables model representing median splits household income percent single women heads households dependents, adjusting maternal age income, accounted much remaining between-tracts variation. CONCLUSIONS: Residing wealthier tract (> $30,000/year income) associated reduced adjusted OR = 0.59 (95% CI: 0.36, 0.96). The estimated conditional effect lower female headed 0.71 0.43, 1.17).