Measurements and Predictors of Physical Activity Among Toddlers

Objective Examine feasibility and predictors of physical activity (PA) among toddlers using accelerometry. Methods Low‐income toddlers (n=268) wore an Actical accelerometer on their ankle for an average of 5 days. Data recorded in 1‐min epochs summarized as total activity counts/day. Results Toddler mean age was 20 months (range 12–32), 67% African American, 12% overweight (WL>95th %tile). Maternal obesity was 51% (BMI>30 kg/m2). 72% toddlers had valid Actical data. Three PA categories were created from distribution of data <400,000 (low, 22%), 400,000–800,000 (moderate, 65%), >800,000 (high, 13%). In multinomial logistic regression, adjusting for WL, gender, and accelerometer days, toddler age and maternal BMI were associated with PA. Toddlers of obese mothers were 4.8 times (p=0.009) more likely in low PA and 2.9 times (p=0.035) more likely in moderate PA, compared to high PA. Younger toddlers were more likely in low (p<0.001) and moderate (p=0.037) compared to high PA (mean age/group: 17, 21, 23 months). Conclusions Accelerometry is a feasible method for assessment of toddler PA. Toddler PA increases with age and is associated with maternal BMI, but is not related to toddler overweight or gender.