Individual determinants of residential mobility over a 20-year follow-up period (EGEA study)

Abstract Background Residential self-selection bias, which may result from demographic, socioeconomic or health-related characteristics predicting neighborhood “choice”, is a methodological issue in environmental health studies but has been rarely investigated. The objective is to identify individual determinants of residential moving among adults. Methods The study is based on the Epidemiological study on the Genetics and Environment of Asthma (EGEA), a French multi-center cohort recruiting in early 90's a group of asthma cases, their first-degree relatives and a group from general population. Overall, 753 EGEA participants aged ≥25 years old at inclusion and who provided their residential history since 1990, through a self-reported questionnaire completed about 20 years later (2011-2013) were included in the present study. We conducted multivariable logistic regression analyses to investigate the associations of baseline demographic characteristics (age categories, sex, marital and family status) and socioeconomic factors (educational level, occupational status), housing type (multiple/single dwelling), and asthma status, with the probability of residential moving (yes/no) between 1990 and response to the residential history questionnaire. Results Study population was 55% female, 44±10 years old in average; 51% moved during the study period(1.4±1.8 moves in average). Compared to non-movers, movers were significantly (p < 0,05) more likely to be younger, single, to live in multiple-unit dwellings and to have higher education level but lower occupational status. Sex, family and asthma status were equally distributed across movers and non-movers. Conclusions In our 20-year follow-up study, some demographic and socioeconomic characteristics were associated with residential moving behaviors. Further analyses will be conducted to evaluate if residential moving of EGEA participants result in differential environmental exposures, which may contribute to social inequalities in health. Key messages • Residential moving is not a random process and is associated with sociodemographic factors. • Studying determinants of residential mobility is essential to account for residential self-selection bias.