BACKGROUND
Dog walking is an important but largely under-recognized contributor to adults’ total physical activity. It is known that the physical environments of neighborhoods (particularly connectedness of streets, residential density and proximity of destinations) influence adults’ walking habits. Dog walking has been integral to physical activity campaigns in several countries, but without objective evidence on actual levels, social variations or on the determinants of dog-walking behaviour within particular social groups.
OBJECTIVE
To better understand socioeconomic variations in dog walking at the population-level, and to assess the implications for future physical activity campaigns and for developing walking-friendly ordinances at the local community level.
METHODS
We assessed socioeconomic variations in dog ownership and dog walking, and their relationships with total leisure-time physical activity and BMI. In the PLACE (Physical Activity in Localities and Community Environments) study, 2,650 residents were recruited from 32 neighborhoods made up of clusters of contiguous Census Collectors’ Districts in Adelaide (the capital city of the state of South Australia), and selected so as to be strongly distinguished by their socioeconomic status and their ‘walkability’ attributes (residential density, street connectivity, intersection density), land use mix, and retail area ratio). Neighborhoods were high walkable (top 25%; HW) or low walkable (bottom 25%; LW); and, higher socio-economic status (top 25%; HSES) or lower SES (bottom 25%; LSES). Neighborhood-environment measures were from Geographic Information Systems databases; neighborhood socio-economic indices from 2001 Census data; behavior patterns, preferences and attitudes were assessed via mail-out surveys, and included items on individuals’ walking habits (IPAQ long form), other physical activities, dog ownership and dog walking.
RESULTS
Dog ownership varied by environmental and social attributes of the neighborhoods (LW/LSES = 55%; LW/HSES = 44%; HW/LSES = 45%; HW/HSES = 18%). Among dog owners, the percentage of people who ever walked their dog varied significantly by neighborhood attributes (LW/LSES = 62%; LW/HSES = 69%; HW/LSES = 68%; HW/HSES = 90%). Dog walking did not vary by gender or age group, but was more prevalent among those with tertiary education (75% of dog owners) compared to other educational categories (64%, p<0.05). Total daily mean minutes of leisure-time physical activity (LTPA) varied significantly between non dog-walkers and dog walkers, for all neighborhood attribute categories: LW/LSES (daily mean minutes of LTPA for non dog-walkers = 16, versus 35 minutes for dog-walkers); LW/HSES (16 versus 43 minutes); HW/LSES (17 versus 40 minutes); HW/HSES (16 versus 64 minutes). Main effects were significant for variation in minutes of LTPA for dog ownership F1,2490 = 15.9, p<0.01;, and for location F 3,2490=10.5, p<0.01. Clear location-level differences were apparent for those dog owners who walked their dogs for at least 10 minutes per day. This was reported by 65% of the HW/HSES group; and by 51% of the HW/LSES and LW/HSES groups; and, by 43% of the LW/LSES (p<0.01). The proportion of all leisure time walking minutes contributed to by dog walking varied between 50-60%, for all locations. Mean body mass index (BMI) varied significantly and in a consistent manner between non dog-walkers and dog-walkers, for all neighborhood attribute categories: LW/LSES (BMI for non dog-walkers 27.5; versus 27.1 for dog-walkers); LW/HSES (27.5 versus 26.1); HW/LSES (28.1 versus 26.2); HW/HSES (25.7 versus 24.1).
CONCLUSIONS
Dog walking is less prevalent in less-walkable, lower-SES neighborhoods than it is in more affluent and walkable neighborhoods. Non dog-walkers do significantly fewer daily minutes of leisure-time physical activity and have significantly higher mean BMI levels. Further, dog walking, among dog owners contributes to over half of their total leisure-time walking minutes, which is an important contribution to their total physical activity. We also use multi-level statistical methods to examine the interactions of more particular neighborhood-level social and environmental factors with individual-level factors, for explaining not walking one’s dog, and obesity, among dog owners. Such data are helpful elements of an evidence base for walking promotion strategies, targeting dog owners who insufficiently utilize their canine walking opportunities. The findings emphasize the need for promotional strategies and for community-environment initiatives that are evidence-based, targeted and of specific relevance to inactive and overweight dog owners, especially those residents in less-walkable, lower-SES neighborhoods.
