Measuring Physical Behavior after Stroke : Sedentary behavior, body postures & movements, and arm use

The primary aim of this thesis was to investigate two methodological aspects of measuring physical behavior from the perspective of stroke rehabilitation. The methodological aspects were:
1) the effect of applying different operationalizations of the construct to be measured, and
2) the validity of a measurement device.
These aspects were investigated with respect to three components of physical behavior: sedentary behavior, body postures & movements, and arm use. Another aim was to apply physical behavior monitoring to describe daily-life arm use in people after stroke.

It was found that different operationalizations of sedentary behavior had a clear effect on the outcomes related to the total amount of sedentary time and the way sedentary time accumulates in bouts, in healthy people and in people after stroke. In both groups, the differences were not only significant but also large enough to acknowledge differences between the different operationalizations. Next, we found that the Activ8 Physical Activity Monitor was sufficiently valid to detect body postures & movements in people after stroke. The Activ8 Arm Use was developed and proved to be sufficiently valid to measure arm use during lying/sitting and standing in people after stroke. Therefore, both these activity monitors can be used to measure components of physical behavior in stroke rehabilitation. The results of using the Activ8-AUM in people after stroke showed that, 3 weeks after the stroke, the arm use ratio was low, i.e. the arms were used in a non-symmetrical way and with low use of the affected arm. During the first 26 weeks after the stroke, although the arm use ratio increased it remained significantly lower than the ratio in healthy people, as reported by others. Moreover, both the arm use ratio and its increase showed considerable variability between participants. The arm use ratio seems to be non-linearly related with arm function, because the positive relation between arm use and arm function was more clearly observed at higher levels of arm function.