Does dog-ownership influence seasonal patterns of neighbourhood-based walking among adults? A longitudinal study

Participation in regular physical activity is important for the prevention of many chronic diseases, including type 2 diabetes, hypertension, cardiovascular disease, obesity, osteoporosis, some cancers, irritable bowel syndrome, dementia and depression [1‐5]. Despite these health benefits, at least 60% of the global population does not achieve the minimum recommendation of 30 minutes of moderate-intensity physical activity daily [6]. In Canada, only 31% of adults walk more than one hour per week [7], and in the Province of Alberta, 25% of adults report not walking for leisure at least once in the previous three months [8]. The correlates of physical activity include demographic, biological, psychological, social, and physical environmental factors [9‐12]. Moreover, physical activity behaviour tends to fluctuate across seasons, especially in countries where extreme differences in seasonal weather patterns are experienced [13‐15]. For example, levels of physical activity are significantly lower during colder months particularly in countries that experience extreme weather such as the United States, Canada, Scotland, The Netherlands, and France [15]. During winter, 64% of Canadians are inactive as compared with 49% in the summer [13]. Extended periods of physical inactivity results in physiological deconditioning and, when accumulated over time, could increase the risk of adverse health outcomes [16‐18].

Several quantitative studies have found that dog-owners tend to be more physically active than non-owners [11, 19]; all of these cross-sectional studies, however, were conducted in geographical areas with mild or moderate climates, including one Canadian study [19] undertaken in a coastal city with a temperate climate. Focus group research, meanwhile, suggests that dog-owners view their dogs as a source of companionship and motivation for getting outdoors and walking [20, 21] - regardless of weather conditions [21]. As one participant put it, "But if you've got a dog, you've got to come out in all weathers" [21], p.443], while another said, "If it's absolutely foul weather sometimes you think, oh dear, I've got to go out, but once you're out there, providing you're dressed appropriately, it's just fabulous, so yes, so I walk every day" [21], p.444]. The available evidence thus suggests a potential role for dog-ownership in overcoming or mitigating the reduction in physical activity often observed during winter months, but this possibility has yet to be examined quantitatively. We investigated the extent to which dog-ownership influences seasonal patterns in neighbourhood-based walking among adults living in highly-variable climate.

The average age of the sample (n = 428) was 53 ± 15 years. The sample included 115 dog-owners (26.9%) and 313 non-owners (73.1%). While dog-owners were disproportionately white, higher-income, residing in a detached or semi-detached house, there was no difference between dog-owners and non-owners in education, marital status or sex (Table 1).

In the summer, dog-owners on average spent 213.6 min/week in NRW, compared with non-owners who spent on average 123.3 min/week (Table 2). Meanwhile, in the winter, dog-owners on average spent 253.2 min/week in NRW compared with non-owners who spent 107.1 min/week. Less time per week was spent in NTW than in NRW among dog-owners and non-owners (Table 2). Weekly minutes of NTW were not significantly different between dog-owners and non-owners, for either summer or winter. Adjusting for demographic characteristics, there was no statistically significant difference in the change in weekly NTW minutes from winter to summer between dog-owners and non-owners (Table 2). We did, however, find that dog-owners undertook on average 39.6 min/week more NWR in winter than in summer, while non-owners undertook 16.2 min/week less NWR in winter than in summer (Table. 2). This result suggests a difference between dog-owners and non-owners in weekly NWR minutes across seasons - even after adjusting for demographic characteristics.

Among all respondents, 65.1% participated in NRW in both winter and summer, 8.9% participated in NRW in the summer only, 10.5% participated in NRW in the winter only, and 15.4% did not report any participation in NRW in the summer and winter (Table 1). Respondents who participated in NRW in the summer only (i.e., no NWR reported in winter) were non-owners. Adjusting for demographic characteristics, we found that dog-owners were at least 3 times more likely than non-owners to be consistent participators (i.e., in winter and summer) in NRW (OR = 3.28; 95%CI = 1.86, 5.79) (Table 3). Dog-owners were, in addition, less likely than non-owners to be inconsistent participators in NWR (OR = 0.35; 95%CI = 0.16, 0.79) (Table 3). Among all respondents, 43.7% participated in NTW in both winter and summer, 11.9% participated in NTW in the summer only, 12.8% participated in NTW in the winter only, and 24.8% reported no participation in NTW in summer and winter. Moreover, dog-owners were no more or less likely than non-owners to participate in NTW, in summer, in winter, or in both seasons (Tables 1 and 3).

Of the dog-owners in our sample, 63% reported walking their dog in both summer and winter; similarly, Australian research found that the majority of dog owners regularly walked their dogs [28]. Overall, dog-owners reported greater participation in NRW than did non-owners, supporting previous studies showing dog-owners to be more physically active than non-owners in recreational walking [11, 19, 29, 30]. Yet our finding that dog-owner's participation in physical activity remained relatively stable from winter to summer is novel, as is the finding that dog-owners' NRW increased in winter, while non-owners' NRW decreased.

Overall, our study's longitudinal design allowed us to pose questions and draw conclusions that are not possible with cross-sectional data. To the best of our knowledge, research investigating dog-ownership as a factor in longitudinal patterns in neighbourhood walking has not previously been undertaken. Compared to non-owners, dog-owners walked more for recreation in both the winter and summer, which corresponds with previous studies that found dog-owners spend more time in mild to moderate physical activity, including walking [11, 19, 31] and in recreational physical activity [30]. Moreover, dog-owners' NWR, on average, was equivalent of 30-minutes/day regardless of season - the recommended minimum amount of physical activity to accrue health benefits [32]. In contrast, non-owners, on average, were not achieving this level of NWR, and in fact reported slightly less NWR in winter than in summer.

A significant difference in NTW between dog-owners and non-owners was not found in our study, but previous work has found that dog-owners are less likely to walk for transportation compared to non-owners [30]. The limited difference in NTW among dog-owners compared with non-owners could be attributable to a lack of dog-friendly parks, shops, and other commercial destinations within walking distance of home [33]. Allowing people to bring their dogs to work could be beneficial to owners as well as to non-owners [34], but dogs are not generally allowed in workplaces unless they are designated as service animals, such as seeing-eye dogs. Consequently, adults who walk to work generally leave their dogs at home. In addition, even though dogs are allowed on public transportation in the City of Calgary, the dog-owners in our sample who commute by public transit would have few options for dog care at or near businesses, workplaces, or educational institutions, thus impeding them from walking with their dog to bus stops or commuter train stations located within their neighbourhood.

This study contributes to the literature on determinants of physical activity by showing that dog-ownership contributes to walking for recreation in residential neighbourhoods despite seasonal fluctuations, yet has several limitations. While we used previously-tested items for capturing walking [26, 27], self-reported physical activity is nevertheless affected by recall and memory bias [35]. The use of objective measures of physical activity (i.e., pedometers and accelerometers) in future research could overcome this limitation. About 45% of our original sample was lost to follow-up or was missing data, which may have introduced a bias to our sample [7]. The reasons why we found higher levels of NRW among dog-owners in the winter compared with the summer, furthermore, are not evident. One possibility is that during summer months owners and their dogs may be more likely to undertake recreational activities at destinations or in open spaces located outside their immediate neighbourhood (i.e., mountains, provincial parks, lakes) rather than close to home. Another possibility, is that in the summer months, dog-owners living in detached or semi-detached housing may be inclined to unleash their dogs in the backyard, rather than taking them out for walks in the neighbourhood. Future research using qualitative and quantitative methods is needed to better understand the interrelationships between dog-ownership, seasonality and physical activity behaviour.

Health professionals should discuss but not prescribe dog-ownership as a way to promote physical and mental health via regular physical activity. Dog-ownership may provide a social support that encourages walking, therefore, initiatives to help shape a similar type of social support, such as dog-sharing programs, may help increase physical activity [36, 37]. Furthermore, as not all dog-owners regularly walk their dog [28], destinations such as cafes, shops, and parks with dog-friendly attributes could encourage more walking among this group [38]. Dog-ownership status should be considered when designing and assessing interventions to increase year-round physical activity participation.