Preliminary Findings From A Stealth Physical Activity Intervention Targeting Inactive Dog Owners

Abstract

Given that 60 million American households own at least one dog, there is growing interest in promoting dog walking to increase physical activity at the population level. An estimated 40% of dog owners do not walk the dog regularly, providing a large target population for intervention. Dog obedience training could plausibly serve as a stealth physical activity intervention as it aims to strengthen the dog-owner bond, a construct strongly associated with dog walking behavior. PURPOSE: To examine changes in dog owners’ self-reported dog walking behavior and device-measured moderate-to-vigorous physical activity (MVPA) after completing basic obedience training. METHODS: Forty-one healthy but inactive individuals (85% female; mean age=40) who reported walking their dog ≤3 d/wk were randomized to a 6-week basic obedience training class (INT; n=21) or wait list control group (CON; n=20). Participants recorded all dog walking bouts and wore an Actigraph GT3X+ on their right hip for 7d at baseline and 6-weeks. T-tests assessed group differences in self-reported dog walking and device-measured MVPA change scores. RESULTS: At baseline, participants reported 48.5±62.7 min/wk of dog walking and averaged 22.0±14.0 min/d of MVPA. Intervention participants that completed post-program assessments (n=17) attended an average 5.6 of 6 training classes. Preliminary analyses of n=31 participants (n=13 INT) with valid Actigraph data (≥4d with ≥8hrs weartime at both time points) found a differential change in self-reported dog walking behavior between groups (+36.1±58.4min/wk versus -26.7±90.3min/wk in INT and CON participants, respectively; p=0.04), but no differential change in MVPA (+7.4±22.2min/d versus +1.6±11.3min/d in INT and CON participants, respectively; p=0.35). CONCLUSIONS: In this small pilot study, dog obedience training led to increases in self-reported dog walking behavior but no change in device-measured MVPA as compared to a waitlist control group. There was large variability in both physical activity outcomes measures, so results should be interpreted with caution. This approach should be tested in a larger sample and should specifically target inactive dog owners that intend to or would like to walk the dog regularly but report barriers (rather than owners that do not intend to walk the dog).