Associations between short-term and long-term exposure to particulate matter and preterm birth

Despite the longstanding evidence on the effect of air pollutants on preterm birth (PTB), few studies have focused on its subtypes, including spontaneous preterm birth (sPTB) and medically indicated preterm birth (miPTB). Most studies evaluated only the short-term or long-term effects of particulate matter (PM) on PTB. Thus, we designed this study, based on a cohort of 179,385 women, to evaluate both short-and long-term effects of PM with diameters <= 2.5 mu m and <= 10 mu m (PM2.5 and PM10) on PTB, sPTB and miPTB in Shanghai. Generalized additive models (GAMs) were applied to evaluate short-term effects. Lagged effects were identified using different lag structures. Exposure-response correlation curves were plotted using GAMs after adjustment for confounders. ORs and 95% CIs were calculated using logistic regression to estimate the long-term effect after adjustment for confounders. There was 5.67%, 3.70% and 1.98% daily incidence of PTB, sPTB, and miPTB on average. Every 10 mu g/m3 increase in PM2.5 and PM10 was positively associated with PTB and sPTB at lag 2 day. The exposure-response curves (lag 2 day) indicated a rapid increase in sPTB for PM2.5 and a linear increase for PM10, in PTB for PM2.5 and PM10 at concentrations over 100 mu g/m3. Regarding long-term exposure, positive associations were found between 10 mu g/m3 increases in PM2.5 and PM10 in 3rd trimester and greater odds of sPTB (aOR: 1.042, 95% CI: 1.018-1.065, and 1.018, 95% CI:1.002-1.034), and during the 3 months before conception and miPTB (aOR: 1.023, 95% CI: 1.003-1.042, and 1.017, 95% CI: 1.000-1.036). Acute exposure to PM2.5 and PM10 at lag 2 day and chronic exposure in 3rd trimester was significantly associated with sPTB, while miPTB was related to chronic exposure during the 3 months before pregnancy. These findings indicate that susceptibility windows of PM exposure can be influenced by different underlying etiologies of PTB.