The interplay of home and work neighbourhood environment characteristics and associations with active commuting

Introduction Despite its health benefits, the use of active commuting is rarely examined in relation to the co-occurring combinations of built and social environment features at home and work environments that reflect the home-to-work commute. Methods Self-reported mode of commute data from participants of the 2016 Canadian National Census (n = 2,077,405) was linked to environmental features (48,624 dissemination areas) (e.g., walkability, area marginalization). Hierarchical cluster analysis identified typologies/clusters of co-occurring features in urban areas of Canada, and these typologies were assigned to respondents’ home and work neighbourhood locations. Associations between co-occurring environmental clusters and commuting modes (walking or cycling, public transit, and motorized vehicle) were examined independently and combined for both home and work locations using multinomial regression. Results Four environmental clusters were identified, ordered by increasingly supportive walking and cycling infrastructure. Cluster 1 had fewer active commuting resources (bottom 30% mean scores for walkability, public transit, cycling), high greenness, and air quality (NO2, PM2.5) (top 33%) with moderate scores (between top/bottom 33%) for proportion living alone or unowned dwellings (residential instability) and material deprivation (low income, reduced basic needs access). Clusters 2 and 3 had moderate active commuting support and air quality but differed in residential instability (lower in cluster 2, higher in cluster 3). Cluster 4 had supportive active commuting environments, higher material deprivation, more older adults and non-working respondents, higher immigrant and visible minority concentrations, and residential instability. Multinomial regression models showed that compared to a cluster 1 home and work location, those in a cluster 3 environment had the most active commuters (adjusted Risk Difference, RD: 10.3 per 1000; 95% CI: 9.6, 11.0) and public transit users (RD: 185.6 per 1000; 95% CI: 178.1, 193.2). Combinations of cluster 3 and cluster 4, when examined at combined home and work locations, also resulted in more active commuters and public transit users. All other combinations resulted in fewer active commuters, and more public transit and motorized vehicle users. Conclusions Combinations of clusters 3 and 4 (environments with mid-high active commuting supports) at both home and work locations contributed to the highest levels of active commuting and public transit use and fewest motorized vehicle users relative to the least supportive typology. Findings highlight the importance of built environments at both home and work environments, even in deprived or unstable social contexts, in promoting active commuting.