Few studies assess dog ownership and walking with both self-reported or perceived and audited or objective walkability and physical activity measures. Across two years, we examined both types of walkability and activity measures for residents living within 2km of a "complete street"-one renovated with light rails, bike lanes, and sidewalks. Audited walkability (Irvine-Minnesota Inventory) was more consistently related to dog ownership and walking groups than perceived walkability (Neighborhood Environment Walkability Scale-Abbreviated). Self-reported leisure walking was much higher (289-383 min per week) among dog walkers than among other groups (100-270 min per week), despite no difference in accelerometer-measured light or moderate-to-vigorous physical activity (MVPA). Furthermore, the most powerful difference between groups involved single-family detached home residence, which was much lower among non-dog-owners (44%) than among non-dog-walkers or dog walkers (81% and 70%, respectively). Given discrepancies across walkability and activity measures, we recommend future use of walkability audits and objectively measured physical activity over the current emphasis on self-report measures. We also urge greater attention to increased densities of housing, which may negatively affect dog ownership levels unless compensating supports for dog ownership and walking are created by public health messaging, dog-friendly policies, and dog-friendly housing and community design.
Keywords: accelerometer; audited walkability; dog ownership; dog walking; perceived walkability; physical activity.