IMPORTANCE Airborne particulate matter pollution has been associated with preterm birth (PTB) in some studies. However, most of these studies assessed only populations living near monitoring stations, and the association of airborne particulate matter having a median diameter of 1 μmor less (PM1) with PTB has not been studied. OBJECTIVE To evaluate whether PM1 concentrations are associated with the risk of PTB. DESIGN, SETTING, AND PARTICIPANTS This national cohort study used National Free Preconception Health Examination Project data collected in 324 of 344 prefecture-level cities from 30 provinces of mainland China. In total, 1 300 342 healthy singleton pregnancies were included from women who were in labor from December 1, 2013, through November 30, 2014. Data analysis was conducted between December 1, 2016, and April 1, 2017. EXPOSURES Predicted weekly PM1 concentration data collected using satellite remote sensing, meteorologic, and land use information matched with the home addresses of pregnant women. MAIN OUTCOMES AND MEASURES Preterm birth (<37 gestational weeks). Gestational agewas assessed using the time since the first day of the last menstrual period. Cox proportional hazards regression analysis was used to examine the associations between trimester-specific PM1 concentrations and PTB after controlling for temperature, seasonality, spatial variation, and individual covariates. RESULTS Of the 1 300 342 singleton live births at the gestational age of 20 to 45 weeks included in this study, 104 585 (8.0%) were preterm. In fully adjusted models, a PM1 concentration increase of 10 μg/m3 over the entire pregnancy was significantly associated with increased risk of PTB (hazard ratio [HR], 1.09; 95%CI, 1.09-1.10), very PTB as defined as gestational age from 28 through 31 weeks (HR, 1.20; 95%CI, 1.18-1.23), and extremely PTB as defined as 20 through 27 weeks' gestation (HR, 1.29; 95%CI, 1.25-1.34). Pregnant women who were older (30-50 years) at conception (HR, 1.13; 95%CI, 1.11-1.14), were overweight before pregnancy (HR, 1.13; 95%CI, 1.11-1.15), had a rural household registration (HR, 1.09; 95%CI, 1.09-1.10), worked as farmers (HR, 1.10; 95%CI, 1.09-1.11), and conceived in autumn (HR, 1.48; 95%CI, 1.46-1.50) appeared to be more sensitive to PM1 exposure than their counterparts. CONCLUSIONS AND RELEVANCE Results from this national cohort study examining more than 1.3 million births indicated that exposure to PM1 air pollution was associated with an increased risk of PTB in China. These findings will provide evidence to inform future research studies, public health interventions, and environmental policies.