Philippa Dall and Malcolm GranatGlasgow Caledonian University James Y. PatonUniversity of Glasgow John J. Reilly University of StrathclydeThe Actigraph is well established for measurement of both physical activity and sedentary behavior in children. The activPAL is being used increasingly in children, though with no published evidence on its use in free-living children to date. The present study compared the two monitors in preschool children. Children (n 23) wore both monitors simultaneously during waking hours for 5.6d and 10h/d. Daily mean percentage of time sedentary (nontranslocation of the trunk) was 74.6 (SD 6.8) for the Actigraph and 78.9 (SD 4.3) for activPAL. Daily mean percentage of time physically active (light intensity physical activity plus MVPA) was 25.4 (SD 6.8) for the Actigraph and 21.1 (SD 4.3) for the activPAL. Bland-Altman tests and paired t tests suggested small but statistically significant differences between the two monitors. Actigraph and activPAL estimates of sedentary behavior and physical activity in young children are similar at a group level.Accelerometers can provide important insights into both physical activity and sedentary behavior of children (1,8). The Actigraph (MTI, Florida) for example provides objective measurement of both physical activity (with particular emphasis Martin, McNeill, Penpraze, and Paton are with the College of Medical, Veterinary, and Life Sciences,

Sedentary behaviour is associated with increased energy intake and obesity in young children (1) . The Actigraph is well established for measurement of both free-living sedentary behaviour and physical activity (2) . Estimates of sedentary behaviour and physical activity by the novel activPAL have not yet been published in children but the validity for measurement of both constructs in young children is established (3) . The present study aimed at comparing measures of sedentary behaviour and physical activity between the two monitors in a convenience sample of free-living pre-school children.Twenty-three pre-school children (mean age: 4.5 years) wore the activPAL and the Actigraph accelerometer simultaneously during waking hours for a mean of 5.6 d and 10 h/d. Measures of daily mean percentage time sedentary (lying/sitting and standing with no translocation) and physically active were compared between the two monitors at a group and individual level.Daily mean percentage of time sedentary was 74.6 (SD 6.8) for the Actigraph and 78.9 (SD 4.3) for the activPAL. Daily mean percentage of time physically active (light intensity plus MVPA) was 25.4 (SD 6.8) for the Actigraph and 21.1 (SD 4.3) for the activPAL. Rank order correlations between the two devices were statistically significant (r = 0.676, P < 0.001). Differences in percentage of time spent sedentary and physically active between the monitors were significant (paired t-test, P < 0.001). The Bland-Altman analysis (Fig. 1) showed that for estimates of percentage of the day spent sedentary there was a difference of -4.3 % (SD 4.8) of daily time (limits of agreement -14.0 to 5.4%) for the Actigraph relative to the activPAL. Estimates of percentage of daily time spent sedentary and in total physical activity were broadly similar between the monitors at the group level but differed at the individual level. An apparently small difference may have an important impact for some applications, such as in obesity treatment and assessment of compliance with physical activity and sedentary behaviour guidelines.