This study suggests that nearly half of adults in this sample who walked a dog accumulated 30 minutes or more of walking in bouts of at least 10 minutes each in 1 day and that neither sex nor income level made a significant difference in whether this amount of dog walking was accomplished. Although for some people the pace of dog walking may be slower than is recommended for physical activity, dog walking may contribute to participation in regular physical activity.
These findings are similar to findings from other studies, in which dog owners who walked their dogs accumulated physical activity that contributed to meeting recommendations. In an Australian cross-sectional analysis, most dog owners who walked their dogs for more than 1 hour per week were more likely than non-dog owners to meet physical activity guidelines of greater than or equal to 150 minutes of total activity per week (11
). In a U.K. prospective study, dog owners significantly increased the number and duration of walks after the first month of dog ownership, and this increase was sustained for the 10 months of the study (12
Our analysis indicates that among dog walkers, sex and family income make little difference in whether a person walks a dog two or more times in a day or accumulates 30 minutes or more of walking in bouts of at least 10 minutes each. These findings do not, however, suggest whether sex or income makes a difference in participating in dog walking generally. Studies on physical activity participation rates have found that overall walking (15
), walking for transportation (16
), and leisure-time physical activities (1
) vary by age, sex, income, and other demographic factors. Because we found that accumulating 30 minutes or more of walking a dog in 1 day is quite common, we urge researchers to find out more about the characteristics of dog walking.
Long-term adherence rates to physical activity programs tend to average only about 50% (7
). One reason proposed for low long-term adherence is that typical programs to promote physical activity do not promote purposeful activity (7
). The significance of purposeful physical activity in long-term adherence was demonstrated in a report of 10 case studies of individuals who regularly participated in physical activity for 5 to 79 years (7
). In seven of the studies, people walked either for transportation or to walk a dog. In three cases, the individuals walked dogs, and these individuals reported walking 3 to 6 miles per day at least 5 days per week for periods ranging from 5 to 15 years. People reported that their regular adherence was due to the need to provide the dog with exercise. The levels of physical activity reported as sustained for years by these people meet or exceed current recommendations.
In addition to providing purposeful activity, dog walking may create a form of social support that has been identified as an effective behavioral strategy for increasing physical activity. The U.S. Task Force on Community Preventive Services strongly recommends social-support interventions in community settings to increase physical activity (17
). Included are behavioral strategies such as developing a buddy system and establishing walking groups or other groups. These strategies provide support and motivation for being physically active. Dog walking may support and motivate physical activity by providing companionship and creating expectations in similar ways to human buddy systems. Walking the dog, in contrast to many other forms of physical activity, is relatively easy and convenient to do, because it can generally be done in one's own neighborhood. This is an important consideration, because having convenient exercise settings is a consistent predictor of regular participation in physical activity for many people (18
Several limitations of this study limit its generalizability. First, although households were randomly selected, the sample was obtained through the responses of participants to open-ended questions on the travel diary; thus, it was not a population-based sample, even though weights were used. Prevalence estimates for dog walking have not been published; however, about 39% of U.S. households have at least one dog (8
), and one study suggests that about two thirds of dog owners walk their dogs, and about 90% of them spend more than 10 minutes on their walks (19
). We must consider whether certain subgroups of the dog-walking population were underrepresented. People who walk their dog frequently could have disproportionately participated, thereby introducing selection bias that could have resulted in overestimating the time spent in dog walking and reduced the variability in the sample. Second, when the trip purpose was pet care and the travel mode was walking, it was presumed to be dog walking. These trips could have included misclassified trips to the veterinarian or walking other species of pets. The results of this study did not change, however, when we excluded individuals (n = 32) who did not own cars and would therefore likely walk their pets to the veterinarian. Third, the intensity or speed of walking could not be assessed because walking distance and time are subject to recall bias and digit preference. Dog walking of moderate intensity may be less prevalent than this study suggests.
Fourth, for many dog walkers (especially those with certain breeds of dogs or those who often walk in inclement weather), dog walking may be always of short duration (less than 10 minutes) and thus not count at all toward accumulating 30 minutes or more of dog walking in bouts of at least 10 minutes as recommended by the Centers for Disease Control and Prevention and the American College of Sports Medicine (5
). Although we can understand the rationale for requiring bouts of at least 10 minutes, we might argue that for dog walkers, short bouts of physical activity are better than sedentary lifestyles. Fifth, proxy interviews for youths aged younger than 16 years may introduce underreporting or overreporting of data.
Finally, the sample design of the NHTS, the low response rate, and the low reporting of dog walking in the diary (n = 1282) might have biased the sample. The 2001 NHTS included a national sample of 25,000 households and oversampled to obtain additional households (N = 37,260) in nine states and transportation regions. Because of the low reporting of dog walking in the open-ended travel diaries, these data may not accurately estimate the characteristics of dog walking and should be interpreted with caution. No differences in characteristics were seen when conducting the analyses with and without sample weights.
This is the first national study to estimate the contribution of dog walking to physically active lifestyles among people in the United States who walk dogs. The data suggest that dog walking can contribute to a physically active lifestyle. Given the limitations of the study, however, more research is needed to determine the national prevalence of dog walking and its contribution to physically active lifestyles. Because it is purposeful, is convenient for most dog owners, and can regularly motivate and support physical activity, dog walking may address several important barriers to physical activity in humans.