Development and Testing of the Observational System for Recording Physical Activity in Children-Toddlers.

Background: Direct observation (DO) tools developed for preschoolers have been used to describe toddler physical activity (PA). No DO system created to assess PA levels and the childcare environment of toddlers exists. The purpose of this study was to develop the Observational System for Recording Physical Activity in Children-Toddlers (OSRAC-T) and assess the inter- and intra-rater reliability. Methods: This tool is an extension of the Observational System for Recording Physical Activity in Children-Preschool and uses the same focal child, time sampling system (5-second observation and 25-second recording). Tool content was established through identifying similar research, consulting with experts, and conducting informal observations. A sample of toddlers (12-36 months) was observed and video recorded during childcare. In-person observations were compared to video observations from one coder (intrarater reliability). Video observations (39% of intervals) were coded by two raters to determine interrater reliability. Results: The final instrument included nine categories that described PA level and type, social and environmental contexts, and transition support relevant to toddlers. Observers completed 124 observation sessions (n = 31; 25.5 ± 6.0 months) resulting in 7,757 30-second observation intervals. Interval-by-interval agreement was moderate to high (58.90%-95.30%) for all categories, and interrater reliability was low to moderate (k = 0.28-0.69). Conclusion: The OSRAC-T is a reliable observation system to assess several PA-related behaviors of toddlers. It may be used to better inform early childcare center design, future intervention studies, or to assess correlates or relationships between PA behavior and health outcomes in toddlers.