Local environments as determinants of walking in Melbourne, Australia.

Structural changes to neighbourhoods that promote walking are likely to benefit everyone in a community and result in long-term improvements in the population's physical activity and health. We consider time spent walking in relation to objectively measured features of people's local environments (functionality, safety, destinations and aesthetics) in Melbourne, Australia. We used multilevel ordered logistic regression analysis to examine variations in time spent walking amongst 2334 residents of 49 small areas (Census Collection Districts - CCDs). Features on each street segment within a 400 m radius of a randomly selected household within each CCD were measured, supplemented by geographic information system data. Models were adjusted for socio-demographic and socio-economic characteristics of individual and socio-economic characteristics of areas. We found that increasing track length, having paths located closer to roads, fewer buildings with driveways, a greater presence and variety of destinations and views of shops, light industry, offices and/or schools (commercial views) were associated with more time spent walking in local environments in adjusted models. As such, each of the four features of local environments considered (design, safety, destinations and aesthetics) were associated on at least one measure with walking. Characteristics of areas reflecting urbanisation (destinations, commercial views, path location and driveways) were associated with increased time spent walking. This study provides important evidence on how urban design can be utilised to improve physical activity.