The relationship between air pollution and low birth weight: effects by mother's age, infant sex, co-pollutants, and pre-term births

Abstract

Previously we identified associations between the mother's air pollution exposure and birth weight for births in Connecticut and Massachusetts from 1999-2002. Other studies also found effects, though results are inconsistent. We explored potential uncertainties in earlier work and further explored associations between air pollution and birth weight for PM₁₀, PM_2.5, CO, NO₂, and SO₂. Specifically we investigated: (1) whether infants of younger (≤24 years) and older (≥40 years) mothers are particularly susceptible to air pollution's effects on birth weight; (2) whether the relationship between air pollution and birth weight differed by infant sex; (3) confounding by co-pollutants and differences in pollutants' measurement frequencies; and (4) whether observed associations were influenced by inclusion of pre-term births. Findings did not indicate higher susceptibility to the relationship between air pollution and birth weight based on the mother's age or the infant's sex. Results were robust to exclusion of pre-term infants and co-pollutant adjustment, although sample size decreased for some pollutant pairs. These findings provide additional evidence for the relationship between air pollution and birth weight, and do not identify susceptible sub-populations based on infant sex or mother's age. We conclude with discussion of key challenges in research on air pollution and pregnancy outcomes.

Keywords: PM10; PM2.5; air pollution; birth weight; carbon monoxide; nitrogen dioxide; particulate matter; pregnancy; sensitivity analysis; sulfur dioxide.

Figure 1

Difference in the change in birth weight (gm) for an interquartile increase in pollutant exposure over the gestational period for infants of the youngest (<25 years) and oldest (>39) mothers, compared to the change in birth weight for the same increment in pollution for infants of mothers ages 30–39 years. (Note: the point reflects the central estimate; the vertical line represents the 95% confidence interval.)

Figure 2

Number of observations available for analysis in single and two pollutant models. (Note: each bar provides the per cent of births from the entire dataset with data for the specified pollutant combination.)

Figure 3

Change in birth weight per IQR increase in gestational exposure to pollutant, for single and two pollutant linear models. (Note: the point reflects the central estimate; the vertical line represents the 95% confidence interval. Models without co-pollutant adjustment are shown with solid lines; models adjusted by co-pollutants are shown with dashed lines.)

Figure 4

Change in birth weight per IQR increase in trimester exposure to pollutant, with and without inclusion of pre-term births. (Note: each line represents the range of results across the trimester models. The horizontal line reflects results based on births with gestational length 32–44 weeks; the vertical line reflects results based on births with gestational length of 37–44 weeks. Dotted lines indicate that results were not statistically significant across all trimester models. Results for a pollutant and trimester without statistically significant results are not shown.)