Predictors of maternal residential mobility in a sibling-matched birth cohort in Massachusetts

Residential mobility during pregnancy and between births is common and can introduce exposure misclassification. It may also reflect broader sociodemographic and environmental inequalities that influence maternal and child health. The objective of this study is to evaluate associations between maternal sociodemographic characteristics, PM_2.5 exposure, and residential mobility. Among mothers who moved, additional analyses evaluated predictors of relocation to lower-income census tracts.

Data were obtained from 155,270 mothers with matched sibling birth records from the Massachusetts Pregnancy to Early Life Longitudinal data system (2001-2009). We define residential mobility as a change in geocoded birth addresses and identified relocation to a lower-income census tract among movers. PM_2.5 exposure estimates at each birth address were assigned using annual averages from a previously validated spatiotemporal model. Logistic regression was used to examine associations between residential mobility and maternal age, race/ethnicity, education, parity, and residential PM_2.5 exposure.

Among mothers with linked births, 49.3% moved between births, and 19.9% relocated to a census tract with a lower median income. Mothers that moved between births had significantly lower PM_2.5 at the subsequent birth address compared with mothers that did not move. Compared with mothers living at low PM_2.5 exposure levels (5th percentile, 1 µg/m³), mothers living at high PM_2.5 exposure levels (95th percentile, 11 µg/m³) had a nearly three-fold higher odds of moving. Relocation to a lower-income tract was less likely among older mothers, non-Hispanic mothers, and those with more than a high school education.

Environmental and sociodemographic factors shape residential mobility patterns between births. It is important to account for residential mobility to reduce exposure misclassification and improve accuracy in perinatal epidemiology.