The validation of a novel activity monitor in the measurement of posture and motion during everyday activities

BACKGROUND: Accurate measurement of physical activity patterns can be used to identify sedentary behaviour and may facilitate interventions aimed at reducing inactivity. OBJECTIVE: To evaluate the activPAL physical activity monitor as a measure of posture and motion in everyday activities using observational analysis as the criterion standard. METHODS: Wearing three activPAL monitors, 10 healthy participants performed a range of randomly assigned everyday tasks incorporating walking, standing and sitting. Each trial was captured on a digital camera and the recordings were synchronised with the activPAL. The time spent in different postures was visually classified and this was compared with the activPAL output. RESULTS: Intraclass correlation coefficients (ICC 2,1) for interdevice reliability ranged from 0.79 to 0.99. Using the Bland and Altman method, the mean percentage difference between the activPAL monitor and observation for total time spent sitting was 0.19% (limits of agreement -0.68% to 1.06%) and for total time spent upright was -0.27% (limits of agreement -1.38% to 0.84%). The mean difference for total time spent standing was 1.4% (limits of agreement -6.2% to 9.1%) and for total time spent walking was -2.0% (limits of agreement -16.1% to 12.1%). A second-by-second analysis between observer and monitor found an overall agreement of 95.9%. CONCLUSION: The activPAL activity monitor is a valid and reliable measure of posture and motion during everyday physical activities.