Regression coefficients for the association of county-level socioeconomic, demographic, environmental and health-related variables with county-level risk of preterm delivery.*

*See Table 1 for variable definitions and scaling. Variables with papb = pc = p

Figshare2015-12-02 更新00

County-level study variables possibly associated with county-level prevalence of preterm delivery: Definitions and weighted means and standard deviations. *

*Each study variable was derived separately for blacks and whites and restricted to women when possible; means are for the 468 counties that had at least 20 preterm deliveries at 20–31 weeks gestation

NIAID Data Ecosystem30

Percentage of variability in county-level risk of preterm delivery explained by regression models that include varied combinations of variables or sets of counties.*

*Percentages are based on R-squared values from the regression models. All models include 468 common counties unless indicated as otherwise.**Stepwise models were created using forward stepwise select

Figshare2015-12-02 更新00

Regression coefficients for the association of county-level socioeconomic, demographic, environmental and health-related variables with county-level risk of preterm delivery.<sup>*</sup>