Validity of the Actigraph GT3x and influence of the sensor positioning for the assessment of active energy expenditure during four activities of daily living in stroke subjects

Objective: To explore the validity of the Actigraph and the influence of the placement of the sensor when estimating the active energy expenditure in four common daily activities (transfers, manual task, walking in overground and walking up and down stairs) in stroke survivors. Design: Cross-sectional study. Setting: Department of Physical Medicine and Rehabilitation. Subjects: A total of 46 subjects affected with stroke sequelae were wearing the Actigraph on three different locations-ankle, hip and wrist on the non-affected side-and performed four tasks: transfers, manual task, walking on flat ground and walking up and down stairs. Main measures: The values of active energy expenditure estimated by the Actigraph were compared to those measured by a portable breathing gas exchange analyzer Metamax3B. The accuracy and agreement between Actigraph and Metamax values were analyzed with mean bias, root mean square error, correlation coefficient and Bland-Altman plots for each task and each sensor location. Results: The mean bias between the Metamax and Actigraph placed on wrist, hip and ankle were, respectively, MD = 1.16 kcal (%MD = 3%), MD =-20.44 kcal (%MD =-58%) and MD = 17.64 kcal (%MD = 50%). The agreement with the Metamax was poor in general regardless of the sensor location and type of task (r = 0.12-0.58). Conclusion: This study found large differences and a poor agreement between the active energy expenditure as measured by the Actigraph and the Metamax according to the location of the sensor and the type of task performed by the subject.