A Statewide Nested Case–Control Study of Preterm Birth and Air Pollution by Source and Composition: California, 2001–2008

Background: Preterm birth (PTB) has been associated with exposure to air pollution, but it is unclear whether effects might vary among air pollution sources and components. Objectives: We studied the relationships between PTB and exposure to different components of air pollution, including gases and particulate matter (PM) by size fraction, chemical composition, and sources. Methods: Fine and ultrafine PM (respectively, PM2.5 and PM0.1) by source and composition were modeled across California over 2000–2008. Measured PM2.5, nitrogen dioxide, and ozone concentrations were spatially interpolated using empirical Bayesian kriging. Primary traffic emissions at fine scale were modeled using CALINE4 and traffic indices. Data on maternal characteristics, pregnancies, and birth outcomes were obtained from birth certificates. Associations between PTB (n = 442,314) and air pollution exposures defined according to the maternal residence at birth were examined using a nested matched case–control approach. Analyses were adjusted for maternal age, race/ethnicity, education and neighborhood income. Results: Adjusted odds ratios for PTB in association with interquartile range (IQR) increases in average exposure during pregnancy were 1.133 (95% CI: 1.118, 1.148) for total PM2.5, 1.096 (95% CI: 1.085, 1.108) for ozone, and 1.079 (95% CI: 1.065, 1.093) for nitrogen dioxide. For primary PM, the strongest associations per IQR by source were estimated for onroad gasoline (9–11% increase), followed by onroad diesel (6–8%) and commercial meat cooking (4–7%). For PM2.5 composition, the strongest positive associations per IQR were estimated for nitrate, ammonium, and secondary organic aerosols (11–14%), followed by elemental and organic carbon (2–4%). Associations with local traffic emissions were positive only when analyses were restricted to births with residences geocoded at the tax parcel level. Conclusions: In our statewide nested case–control study population, exposures to both primary and secondary pollutants were associated with an increase in PTB. Citation: Laurent O, Hu J, Li L, Kleeman MJ, Bartell SM, Cockburn M, Escobedo L, Wu J. 2016. A statewide nested case–control study of preterm birth and air pollution by source and composition: California, 2001–2008. Environ Health Perspect 124:1479–1486; http://dx.doi.org/10.1289/ehp.1510133


Table of Contents
Description of the source constraint checks performed to directly evaluate the accuracy of simulated source contributions using the UCD_P model Figure S1. Directed acyclic graph of assumed relationships between preterm birth, air pollution and other risk factors, based on literature data. Table S1. Correlation matrix for pollutants Figure S2. Odds ratios of preterm birth by quartile of air pollution exposure Table S2. Sensitivity analysis of preterm birth and air pollution, by adjustment for smoking or body mass index, in addition to the covariates included in the primary models (years 2007-2008) Table S3. Sensitivity analysis of moderately preterm birth (MPTB, gestational age <35 weeks) or very preterm birth (VPTB, gestational age <30 weeks) and air pollution References Description of the source constraint checks performed to directly evaluate the accuracy of simulated source contributions using the UCD_P model.
The accuracy of estimates for PM contributions from major sources was evaluated by comparison to receptor-oriented source apportionment calculations based on measurements of PM molecular marker concentrations during specialized field campaigns. Although such campaigns do not occur frequently, good agreement with the available data does build confidence in predictions of PM source contributions from major categories including mobile sources, food cooking, and wood burning. The accuracy of contributions from other minor sources was evaluated by comparison to routine measurements of less-specific component concentrations (elemental carbon/organic carbon/metals). For each component that was predicted accurately at a measurement site by the model (correlation ≥ 0.8 and mean fractional bias within ±0.3), the top 95% of sources contributing to that component concentration within 100 km of the measurement site were identified. Sources identified through this procedure at 3 or more measurement sites were judged to be accurately predicted since their concentrations were consistent with available measurements. (Hu et al. 2014).

Figure S1. Directed acyclic graph of assumed relationships between preterm birth, air pollution and other risk factors, based on literature data.
Plain arrows represent established relationships. Dotted arrows represent relationships for which a greater degree of uncertainty exists (observations based on a few studies and that would call for confirmation by more studies, and/or associations for which mechanisms are not well understood). Square represent observed variables and circles unobserved variables.
The causal diagram theory (Greenland et al. 1999) warns against adjusting for potential colliders, which are factors determined by two or more other factors already included in the model (e.g., parity, which is determined in part by maternal age and socioeconomic factors) in order to avoid over-adjustment bias. Based on the causal diagram above, the statistical models we used for the primary analyses were adjusted for a minimal sufficient set of potential confounders, namely maternal age, race/ethnicity, education, and neighborhood socioeconomic level.   For each pollutant, the quartiles of exposure (averaged from the day of conception to the delivery date of the case in each case control set, see material and methods section) are ranked from left (first quartile) to right (fourth quartile). Dots with bars represent odds ratios for preterm birth and associated 95% confidence intervals in the second, third and fourth quartiles of exposure as compared to the first quartile of exposure (reference group, dot without bar).

Measured pollutant concentrations interpolated by empirical Bayesian kriging (years 2000-2008).
Concentrations of primary PM 2.5 and of species in PM 2.5, modeled at the 4 km*4 km resolution using the UCD_CIT chemical transport model (years 2000-2008).

Odds ratios
Odds ratios Concentrations of primary PM 0.1 and of species in PM 0.1, modeled at the 4 km*4 km resolution using the UCD_CIT chemical transport model (years 2000-2008).
Concentrations of species in PM 2.5, modeled at the 4 km*4 km resolution using the UCD_P chemical transport model (years 2000-2006).

Odds ratios
Odds ratios Concentrations of primary PM 2.5 and PM 0.1 mass by source, modeled at the 4 km*4 km resolution using the UCD_P chemical transport model (years 2000-2006).
Concentrations of primary pollutants from local traffic modelled with CALINE4, for infants with maternal addresses geocoded to the parcel level (years 2000-2008).

Odds ratios
Odds ratios  a) PM2.5; particulate matter less than 2.5 µm in aerodynamic diameter; PM0.1; particulate matter less than 0.1 µm in aerodynamic diameter; OC: organic carbon; EC: elemental carbon; SOA secondary organic aerosols b) Inter-quartile range in exposure. Units are micrograms per cubic meter for all particulate mass and elements, part per billion for gaseous pollutants. c) Odds ratios were estimated using conditional logistic regression models, adjusted for race/ethnicity, educational level and for maternal age using categorical variables and for median household income at Census block group level using polynomial functions. For estimated pollutant concentrations, odds ratios are expressed per interquartile range. For traffic density, they are expressed per 10,000 vehicles per day per meter. For distance to roadways, they compare births within the stated distance to those outside that distance.