Revisiting the MotionWatch8©: Calibrating Cut-Points for Measuring Physical Activity and Sedentary Behavior Among...

Reduced moderate-to-vigorous physical activity (MVPA) and increased sedentary behavior (SB) are common following stroke, which can limit stroke recovery and contribute to greater cognitive decline. Hence, the MVPA and SB of adults with stroke should be measured concurrently using objective methods. One currently available method for objectively measuring MVPA and SB is the MotionWatch8© (MW8). However, adults with stroke can have significant mobility restrictions (depending on stroke severity) and thus it is important to determine separate MVPA and SB cut-points for adults with stroke, as well as validate separate cut-points: (1) when the MW8 is worn on the stroke affected side compared to the non-affected side; and (2) for adults with mild stroke versus adults with moderate-to-severe stroke. In the current study, we concurrently measured MW8 actigraphy (worn on both the stroke affected side and the non-affected side) and indirect calorimetry during 10 different activities of daily living for 43 adults with stroke (aged 55-87 years). Using intra-class correlations (ICC), we first investigated the agreement of the MW8 when placed on the affected side as compared to the non-affected side for: (1) all participants irrespective of stroke severity; (2) participants with mild stroke, classified as a Fugl Meyer motor score of ≥79/100; and (3) participants with moderate-to-severe stroke (i.e., Fugl Meyer < 79/100). We then determined cut-points for all participants-as well as separate cut-points based on stroke severity-on both the stroke affected side and non-affected side for SB and MVPA using receiver operating characteristic curves. The results of our analyses indicate that the agreement in MW8 output between the stroke affected and non-affected sides was moderate across all participants (ICC = 0.67), as well as for each sub-group (mild stroke: ICC = 0.64; moderate-to-severe stroke: ICC = 0.77). Additionally, the results of our cut-point analyses support using different cut-points for different levels of stroke severity and also for the stroke affected side. We determined the following cut-points: (1) for the affected side, adults with mild stroke have cut-points of SB ≤134 counts per minute (CPM) and MVPA ≥704 CPM, while adults with moderate-to-severe stroke have cut-points of SB ≤281 CPM and MVPA ≥468 CPM; and (2) the non-affected side, adults with mild stroke have cut-points of SB ≤162 CPM and MVPA ≥661 CPM, while adults with moderate-to-severe stroke have cut-points of SB ≤281 CPM and MVPA ≥738 CPM. Hence, these data provide a new measure for concurrently examining the dynamic relationships between MVPA and SB among adults with stroke.