Skip to main content
ARS Home » Plains Area » Houston, Texas » Children's Nutrition Research Center » Research » Publications at this Location » Publication #382314

Research Project: Preventing the Development of Childhood Obesity

Location: Children's Nutrition Research Center

Title: Describing and exploring accelerometer use among toddlers

item CRUMBLEY, CHRISTINE - University Of Houston
item CEPNI, ALIYE - University Of Houston
item TAYLOR, ASHLEY - University Of Houston
item Thompson, Deborah - Debbe
item MORAN, NANCY - Children'S Nutrition Research Center (CNRC)
item OLVERA, NORMA - University Of Houston
item O'CONNOR, DANIEL - University Of Houston
item JOHNSTON, CRAIG - University Of Houston
item LEDOUX, TRACEY - University Of Houston

Submitted to: Meeting Abstract
Publication Type: Abstract Only
Publication Acceptance Date: 4/7/2021
Publication Date: 6/8/2021
Citation: Crumbley, C., Cepni, A.B., Taylor, A., Thompson, D.J., Moran, N.E., Olvera, N., O'Connor, D.P., Johnston, C.A., Ledoux, T.A. 2021. Describing and exploring accelerometer use among toddlers [abstract]. International Society of Behavioral Nutrition and Physical Activity (ISBNPA) Annual Conference (Virtual). June 8-10, 2021. Oral presentation.

Interpretive Summary:

Technical Abstract: Studying physical activity (PA) via accelerometers among toddlers is challenging due to non-compliance with wear time (WT) instructions and parental non-compliance with completing activity logs (ALs). Processing software categorizes low activity counts as "sedentary." Parent-completed ALs can help correctly categorize low activity counts for sleep and non-WT. The aims of this study are to ascertain 1) patterns of accelerometer and log compliance among participants, 2) whether compliance varies by parent qualities, and 3) whether sedentary time differs by compliance. Baseline accelerometer and demographic data from a pilot RCT of a community wellness program for parents with toddlers (12-35 months) was used for a secondary analysis. Parents were instructed to have toddlers wear an Actigraph wGT3x accelerometer (Pensacola, FL) on the hip for 8 consecutive days and to complete ALs. A valid day included >=600 minutes WT. The ActiLife6 program parsed raw activity counts into daily activity levels per standard protocol (Butte et al., 2014). Log accuracy scores were based on congruence between ALs and accelerometer counts. Non-parametric analyses examined relationships between variables. The sample (n=50) comprised toddlers (Mean age=27 months, 58% male) and parents (Mean age=31.7 years, 84% female), who were racially/ethnically diverse. Valid accelerometer data accompanied by ALs was provided by 28 families, whereas 8 returned only valid accelerometer data, 4 returned only ALs, and 10 provided neither. The mean for valid days was 5.02 (SD=3.78). The mean log accuracy score was 2.11 (SD=1.68) out of 4 possible points. On the ALs, 31 parents reported bedtimes, 23 reported naps, and 30 explained other non-WT. Parents with a spouse/partner were more likely to complete ALs (p<.05). Toddler sedentary time did not differ between those with ALs (356 min., SD=60.2) and those without ALs (341 min., SD=89.7). Toddler and parent compliance with WT instructions and AL completion varied. Returned AL quality was poor with many missing key information to correctly characterize low activity counts. Sedentary time did not differ between toddlers with or without ALs. Research is needed to identify strategies to improve compliance and data quality in studies of early childhood PA.