The impact of residential relocation on childhood BMI: an exposome-wide natural experiment
Résumé
BACKGROUND AND AIM: Residential relocation can cause sudden and simultaneous changes in multiple environmental exposures and provides a unique opportunity to estimate the potential health impacts of interventions on the external exposome. This study uses residential relocation as a natural experiment to assesses the impact of exposome relocation trajectories on BMI during childhood in the EXPANSE project. METHOD: We collected 27,019 repeated age-and-sex standardized BMI z-scores (z-BMI) observations from 2,915 children and adolescents in the PIAMA (Netherlands) and BAMSE (Sweden) studies who changed address once between 2 and 24 years. Considering various domains of the urban exposome (traffic-related pollution, built surface, green variability, blue space, socioeconomic deprivation), we used a k-means clustering approach to characterize their living environment before and after moving as low, medium and high hazard environments for each exposome domain. The impact of changes in the urban exposome following residential relocation on z-BMI was estimated using fixed-effects models for panel data. We fitted single and multiple exposome domains models and included non-linear time-trends in z-BMI. RESULTS: We found consistent associations between relocating to areas with a worse living environment and increasing z-BMI in the single domain models. The multiple domains models revealed that the association for air pollution and built surface was partially confounded by the other exposome domains. In PIAMA, moving to areas with increased socioeconomic deprivation was significantly associated with increased z-BMI after relocation in all models. CONCLUSIONS: This is one of the first studies to use residential relocation as a natural experiment to assess the impact of changes in various aspects of the urban exposome on health. Results suggest that relocating to new environments with different levels of pollution, built surface and socioeconomic deprivation can have direct impacts on childhood bodyweight. Results inform planning of urban environments to promote healthy weight in children.
