Résumé
Introduction: Autism spectrum disorder (ASD) is influenced by environmental, genetic, and socio-economic factors. While air pollution exposure during development has been linked to ASD outcomes, the roles of green spaces, grey spaces, and healthcare accessibility in this relationship remain understudied. This research examines how these factors during the first 1000 days (from conception to the first two years of life), moderate the association between air pollution and ASD severity in children from the ELENA cohort ("Etude Longitudinale chez l'Enfant avec Autisme").Methods: Data from 237 children with ASD were analyzed. Perinatal exposure to particulate matter (PM 10 and PM 2.5 ) was estimated using a validated model, with green spaces quantified using the normalized difference vegetation index (NDVI) and grey spaces through impervious surface coverage. Healthcare accessibility was measured as the distance to autism resource centers (CRA) from residential addresses. ASD severity was assessed using the total score of the Social Responsiveness Scale. Linear regression models within structural equation modeling estimated the effects of air pollution, green spaces, grey spaces, and healthcare accessibility on ASD severity.Results: Higher PM exposures were paradoxically associated with lower ASD severity. However, grey spaces significantly moderated this association, with greater impervious surface coverage attenuating the negative association. Healthcare accessibility was crucial: when considering center proximity, pollution was no longer associated with ASD severity.Conclusions: These findings highlight the critical role of healthcare accessibility and environmental factors in shaping the relationship between air pollution and ASD severity. Integrating geographic and environmental contexts is crucial when evaluating the environmental determinants of ASD outcomes