BackgroundMany environmental factors have been independently associated with preterm birth (PTB). However, exposure is not isolated to a single environmental factor, but rather to many positive and negative factors that co-occur. The environmental quality index (EQI), a measure of cumulative environmental exposure across all US counties from 2000—2005, was used to investigate associations between ambient environment and PTB.MethodsWith 2000–2005 birth data from the National Center for Health Statistics for the United States (n = 24,483,348), we estimated the association between increasing quintiles of the EQI and county-level and individual-level PTB; we also considered environmental domain-specific (air, water, land, sociodemographic and built environment) and urban–rural stratifications.ResultsEffect estimates for the relationship between environmental quality and PTB varied by domain and by urban–rural strata but were consistent across county- and individual-level analyses. The county-level prevalence difference (PD (95 % confidence interval) for the non-stratified EQI comparing the highest quintile (poorest environmental quality) to the lowest quintile (best environmental quality) was −0.0166 (−0.0198, −0.0134). The air and sociodemographic domains had the strongest associations with PTB; PDs were 0.0196 (0.0162, 0.0229) and −0.0262 (−0.0300, −0.0224) for the air and sociodemographic domain indices, respectively. Within the most urban strata, the PD for the sociodemographic domain index was 0.0256 (0.0205, 0.0307). Odds ratios (OR) for the individual-level analysis were congruent with PDs.ConclusionWe observed both strong positive and negative associations between measures of broad environmental quality and preterm birth. Associations differed by rural–urban stratum and by the five environmental domains. Our study demonstrates the use of a large scale composite environment exposure metric with preterm birth, an important indicator of population health and shows potential for future research.