Using Observation to Compare Different Accelerometer Cut Points for Sedentary Behavior in Children

Introduction Accurate objective methods for the assessment of sedentary behavior are crucial for population monitoring and evaluation of public health strategies for preventing childhood obesity. Actigraph accelerometers are a commonly used objective physical activity measurement tool. It is hypothesized that accelerometers may also provide a valid method for assessing children’s sedentary time. However, there is considerable variation in published accelerometer cut-off points for measuring sedentary time in children [1]. The objective of this study is therefore to compare different accelerometer sedentary cut points with observation of children performing specific sedentary behaviors in free-living conditions.

Methods A convenience sample of 29 apparently healthy children aged 5- to 11-years old (28% boys) were recruited. Direct observation and Actigraph uniaxial accelerometers were used to measure children’s activity intensity while playing computer games, non-electronic sedentary games, watching television and playing outdoors. The measurements took place in the children’s own houses. Direct observation was the criterion for assessing the validity of four different previously published sedentary cut-points: i.e., 100, 300, 800, and 1100 counts per minute (cpm). Data reduction from accelerometers (each of the four by 20 minute time periods) were performed manually with Actilife software, and then transferred to SPSS. Figures were created to depict how different cut-points compared to the observed 25th, 50th and 75th percentiles for the cpm derived from each activity.

Results In total, 25 children (six boys and 19 girls) aged 5-11 years were included in the analysis. The median cpm were lowest for computer games (30 cpm), followed by television viewing (109 cpm) and non-electronic sedentary games (172 cpm) and highest for outdoor play (1452 cpm). The median counts during all sedentary behaviors were below the lowest cut-point of 100 cpm. The 75th percentile values for the sedentary behaviors were always below the cut-point of 300 cpm.

Discussion and Conclusion The results suggest that estimated time spent in sedentary behavior would differ considerably using the various published cut-points available. It is important to be aware of the consequences of using the different cut-points when assessing sedentary behavior in children. Based on our results, the Actigraph accelerometry cut-point of <100 cpm is the most appropriate, if not too high, for quantifying the time children spend sedentary.

Reference [1] Reilly JJ, Penpraze V, Hislop J, Davies G, Grant S, Paton JY. Objective measurement of physical activity and sedentary behavior: review with new data. Arch Dis Child 2008; 93:614-619.