Physical activity assessment with accelerometers.

OBJECTIVE: Evaluation of motion sensors, specifically accelerometers, as an objective tool for the assessment of physical activity in large populations, over periods long enough to be representative of normal daily life and with minimal discomfort to the subjects.
METHOD: Review of validation studies of accelerometers with indirect calorimetry as a reference method. Accelerometers were commercially available one-axial accelerometers: Caltrac, Computer Science Application (CSA) accelerometer, Mini Motionlogger Actigraph; the tri-axial accelerometer Tritrac-R3 D; and an tri-axial accelerometer for movement registration (Tracmor) from our laboratory.
RESULTS: There is no clear difference for correspondence between indirect calorimetry and accelerometer counts during level walking, whether one-axial or tri-axial and placed at the wrist, hip or low back. Sedentary activities are better reflected with a tri-axial accelerometer than with a one-axial accelerometer. Two accelerometers were validated in free living conditions with doubly labeled water. The highest correlation between accelerometer output and activity induced energy expenditure was found for Tracmor.
CONCLUSIONS: From all accelerometers tested, the tri-axial accelerometer for movement registration is an objective method that can be used to distinguish differences in activity levels between individuals and assess the effect of interventions on physical activity within individuals.