Assessing the Effects of Residential Mobility on Air Pollution Exposure Misclassification in Longitudinal Birth Cohort Studies

摘要

To accurately assess environmental exposures in longitudinal cohort studies, a better understanding is needed about the effects of incomplete or inaccurate records of residential mobility on exposure misclassification. We developed an ALGorithm for Generating Address Exposures (ALGAE) that assigns air pollution exposure using residential address histories. We applied ALGAE to address records and historic daily air pollution estimates (particulate matter <10 μm [PM10]) for the Avon Longitudinal Study of Parents and Children (ALSPAC). ALSPAC recruited 14,541 pregnant women between 1990 and 1992 in the South West of England. For each pregnancy, we modelled pregnancy trimester- and first year of life-specific exposure estimates based on PM10 concentrations at residential address at birth and using reconstructed address histories to account for mobility during pregnancy and first year of life. We were able to assign trimester-specific PM10 exposures for 12,752 pregnancies and first year of life exposures for 12,252 pregnancies. Average PM10 exposures were 32.6 μg/m3 (StDev 3.0 μg/m3) during pregnancy and 31.4 μg/m3 (StDev 2.6 μg/m3) during first year of life. 6% of women moved address during pregnancy, and 13% moved during first year of life. Exposure differences ranged from -5.3 μg/m3 to 12.4 μg/m3 (up to 26%) during pregnancy and -7.22 μg/m3 to 7.64 μg/m3 (up to 27%) in first year of life between the exposure assessed using the address at birth compared to the exposure assessed using the complete cleaned address history. Accounting for residential mobility in the exposure assessment of birth cohort studies has the potential to substantially reduce exposure misclassification for some individuals. ALGAE provides an open-source solution to clean addresses stored in the cohort contact database and assigns life-stage specific exposures. Its generic code base means that it is adaptable for other cohort studies.