Validity, practical utility and reliability of Actigraph accelerometry for the measurement of habitual physical activity in dogs

In the validation study, 30 dogs wore the accelerometer for 1 day while being filmed. Accelerometer and film were synchronised and 10-minute periods of the filmed records were extracted with dogs in continuous periods of sedentary behaviour, light intensity physical activity indoors, light to moderate intensity physical activity outdoors and vigorous physical activity outdoors. For the -practical utility and reliability studies, 20 dogs wore the GT3-X accelerometers for 1 week: practical utility was quantified as data loss and was also assessed by owner questionnaire; reliability was determined by 2 to 7 days of monitoring using the Spearman-Brown prophecy formula.
In the validation study, accelerometry output differed significantly between activity intensities (Friedman test, P < 0 center dot 01). In the practical utility study, no data were lost from any dogs and dog-owners reported that accelerometry was well tolerated. Reliability of accelerometry output was high: for 3 days of wear, it was 91% [95% confidence interval (CI) 82 to 96] and for 7 days of wear, it was 94% (CI 88 to 97).
Clinical Significance: The GT3-X accelerometer is valid, practical and reliable for the measurement of habitual physical activity in dogs.