Associations between the physical activity levels of fathers and their children at 20 months, 3.5 and five years of age

BACKGROUND Early childhood (under five years of age) is a critical developmental period when children's physical activity behaviours are shaped and when physical activity patterns begin to emerge. Physical activity levels track from early childhood through to adolescence with low levels of physical activity associated with poorer health. The aims of this study were to examine cross-sectional and longitudinal associations between the physical activity levels of fathers and their children at the ages of 20 months, 3.5 and 5 years, and to investigate whether these associations differed based on paternal body mass index (BMI) and education. METHODS The Melbourne Infant Feeding Activity and Nutrition Trial (InFANT) Program was a cluster randomized-controlled trial delivered to pre-existing first-time parent groups. Physical activity levels of fathers and their first-born children were assessed using the Active Australia Survey and ActiGraph accelerometers respectively. Cross-sectional associations between father and child physical activity behaviours were assessed at each time point. Longitudinal associations between father and child physical activity were also investigated from child age 20 months to both 3.5 and 5 years. Additional stratified analyses were conducted based on paternal BMI and paternal education as a proxy for socioeconomic position (SEP). Data from the control and interventions groups were pooled and all analyses adjusted for intervention status, clustering by first-time parent group and accelerometer wear time. RESULTS Physical activity levels of fathers and their children at child age 20 months were not associated cross-sectionally or longitudinally at child age 3.5 and 5 years. Positive associations were observed between light physical activity of healthy weight fathers and children at age 3.5 years. Inverse associations were observed for moderate/vigorous physical activity between fathers and children at age 5 years, including between overweight/obese fathers and their children at this age in stratified analyses. CONCLUSIONS There were no clear associations between the physical activity of fathers and children. Future research should include the use of more robust measures of physical activity among fathers to allow in-depth assessment of their physical activity behaviours. Investigation of well-defined correlates of physical activity in young children is warranted to confirm these findings and further progress research in this field.