Dog ownership, accelerometer-measured physical activity, and mental health in Chinese adults: a 12-month cohort study

Dog ownership has been associated with higher physical activity levels in Western populations, yet longitudinal evidence from Asian countries using objective measurement remains scarce. This study investigated the associations between dog ownership, accelerometer-measured physical activity, and mental health among adults in Hefei, China. A 12-month prospective cohort study enrolled 258 participants (132 dog owners, 126 non-owners) from June 2024 to May 2025. Physical activity was measured using ActiGraph wGT3X-BT accelerometers at baseline, month 6, and month 12. Mental health was assessed using the Patient Health Questionnaire-9, Generalized Anxiety Disorder-7, and Satisfaction with Life Scale. Linear mixed-effects models examined group differences adjusting for age, sex, body mass index, education, income, and season. Age-stratified sensitivity analyses and exploratory mediation analyses were conducted. Dog owners accumulated more daily steps than non-owners at baseline (8,707 vs. 7,129; difference = 1,579, 95% CI: 1,264 to 1,894) and more moderate-to-vigorous physical activity (39.6 vs. 27.4 min/day; difference = 12.2, 95% CI: 9.8 to 14.6). Adjusted models confirmed that the 95% CI for the dog ownership effect excluded the null for all physical activity outcomes. Effects were consistent across age subgroups. By month 12, dog owners reported lower depression (standardized mean difference = 0.52), lower anxiety (0.40), and higher life satisfaction (0.69). Exploratory mediation analyses indicated that moderate-to-vigorous physical activity did not mediate the dog ownership–mental health association (indirect effect: 0.22, proportion mediated < 1%), although the exploratory and cross-sectional nature of the mediation analysis precludes mechanistic conclusions. Dog ownership was associated with substantially higher objectively measured physical activity and progressively better mental health over 12 months. Exploratory mediation analyses did not provide evidence that physical activity mediated the mental health associations, although these analyses were cross-sectional and cannot support causal inference. These findings extend the predominantly Western evidence base to an East Asian context and highlight the need for quasi-experimental designs to establish causality.