Association between residential air pollution and preterm birth in Finland in 1990-2016

Abstract Background Decreases in air pollution levels in Europe during the past decades call for studies to examine the association between low levels of air pollution and preterm birth. Methods We included all spontaneous singleton livebirths in Finland 1990-2016. Annual mean concentrations of fine particulate matter (PM2.5), nitrogen dioxide (NO2), black carbon, and ozone (O3) at residential addresses at the time of birth were estimated using standardised Europe-wide land use regression models. We used logistic regression to estimate odds ratios (OR) for preterm birth (<37 completed weeks of gestation) adjusting for birth year, maternal age, maternal country of birth and parental education. A second model was additionally adjusted for area-level socioeconomic situation (area-SES) and for PM2.5 when studying O3 and for O3 when studying the other pollutants. Results We included 1,152,796 livebirths in the analysis (96% of the total population). Mean exposure levels during the study period were 9.3 µg/m3 (SD 3.2) for PM2.5, 19.7 µg/m3 (SD 10.1) for NO2, 0.42 *10-5/m (SD 0.29) for black carbon, and 64.2 (SD 3.2) µg/m3 for O3. A 10 µg/m3 increase in PM2.5 (OR 0.96, 95% CI 0.94-0.97), NO2 (OR 0.99, 95% CI 0.98-1.00) and a 0.1*10-5/m increase in black carbon (OR 1.00, 95% CI 0.99-1.00) concentrations were negatively and a 10 µg/m3 increase in O3 concentration positively (OR 1.06, 95% CI 1.03-1.09) associated with preterm birth. However, when additionally adjusting the models for area-SES and O3, the associations of PM2.5, NO2 and black carbon attenuated to the null, while O3 remained associated with increased risk of preterm birth (OR 1.05, 95% 1.02-1.08). Conclusions In low air pollution areas, O3 is more consistently associated with increased risk of preterm birth than PM2.5 or NO2. Public health actions should focus on enhancing air quality monitoring and implementing policies to reduce O3 levels, such as stricter emission limits for vehicles and power plants. Key messages • Our register-based study including all spontaneous livebirths in Finland 1990-2016 shows that ozone is a risk factor for preterm birth while no increase in risk was found for other pollutants. • Although air quality has improved, air pollution still reminds as a risk factor for preterm birth and should be targeted by interventions focusing on reducing ozone levels.