Interaction Between Black Smoke and Neighbourhood Socio-Economic Status and Their Association with Birth Weight

Abstract

ISEE-0660 Background and Objective: An accumulating body of evidence suggests that exposure to ambient air pollutants can adversely affect birth weight. However, the effect of socio-economic status (SES) on this relationship is less well described. We investigated whether the association between black smoke (BS) and birth weight is modified by neighbourhood SES. Methods: We used the Particulate Matter and Perinatal Events Research (PAMPER) dataset of singleton births (N = 109,086) in Newcastle upon Tyne during 1961–1992, which contained information on birth weight, gestational age, infant sex, maternal age, parity and neighbourhood SES. Weekly BS exposures for each individual pregnancy were estimated using a two-stage statistical modelling strategy, incorporating temporally and spatially varying covariates. Weekly estimates were averaged over the whole pregnancy and the three trimesters. Neighbourhood SES was based on the Townsend deprivation score (TDS), which was calculated using four major UK census indicators – unemployment, car ownership, house ownership and overcrowding. TDS was grouped into quintiles. Linear regression with fractional polynomials was used to model the relationship between BS and birth weight. Results: Birth weight was significantly associated with BS and TDS individually, and an interaction term was also significant for whole pregnancy (P = 0.003), second trimester (P < 0.001) and third trimester (P = 0.001) exposures. BS and birth weight were significantly associated for each of the quintiles of TDS. For the whole pregnancy exposure, at the 25th percentile (17.2 μg/m3) compared to the 1st percentile (7.4 μg/m3) the estimated reduction in birth weight was 15g, 23g, 41g, 27g and 9g for the 1st (most affluent) to the 5th quintile (most deprived) of TDS respectively. The reduction in birth weight increased with higher BS exposures. Conclusion: We found a significant interaction between neighbourhood SES and BS in their association with birth weight suggesting a modifying effect of SES on the relationship between ambient air pollution and birth weight.