Longitudinal associations between neighbourhood walkability and incident childhood asthma

Methods We evaluated the association between neighbourhood walkability and incident childhood asthma using prospectively-collected administrative data at the Institute for Clinical Evaluative Sciences. Among the 1997-2003 birth cohorts of children living in Toronto, neighbourhood quintile of walkability was reported using a validated walkability index with 4 dimensions: population density, dwelling density, access to services and street connectivity. Incident asthma was defined by the time of entry into the Ontario Asthma Surveillance Information System (OASIS) database, requiring 2 outpatient visits for asthma within 2 consecutive years or any hospitalization for asthma. Sex and neighbourhood income quintile were obtained from the Registered Person’s Database. Histories of preterm delivery, obesity and other atopic conditions were obtained from Ontario Health Insurance Plan records. We calculated the associations between incident childhood asthma and the two lowest versus the two highest neighbourhood walkability quintiles using Cox proportional and discretetime hazard models.