Effect of PM10 and PM2.5 Exposure on Birthweight in North Carolina

Studies suggest that air pollution exposure may be associated with adverse birth outcomes. We use innovative geospatial methods to link NC Detailed Birth Records for 2000–2003 to USEPA PM₁₀ and PM_2.5 monitoring data for pregnant women residing within 20km of a monitor. Analysis was restricted to 44,960 (55,246) singleton, non-anomalous, first births without maternal medical conditions or alcohol use for PM₁₀ (PM_2.5). Multivariate linear modeling was used to estimate birthweight (BWT) as a function of PM exposure, gestation, race, education, marital status, and infant sex.

Maternal exposure to PM₁₀ and PM_2.5 in the second trimester and during the entire pregnancy was negatively associated with BWT (p<.02 for both). Compared to women living in areas with average PM levels, women living in areas with 10^th percentile levels of PM₁₀ (PM_2.5) showed an 8 (7) g increase in BWT and those living in areas with 90^th percentile levels of PM₁₀ (PM_2.5) showed a 9 (8) g decrease in BWT. Tobacco use was associated with 150g lower BWT in both PM models (p<.00001 for both). The combination of residing in areas with 90^th percentile PM_2.5 levels and tobacco use was associated with an additional 28g decrease in BWT (p<.005).

NC has relatively low PM levels. If the NC results are generalizable to areas with much higher PM levels, the effects on BWT would be very significant. The federal standard, intended to provide an “adequate margin of safety,” may need adjustment to account for impacts on pregnant women.