This variable is only available in MAP.

kAR is a measure of transitioning from a state of activity to a state of rest and is a metric of sustained activity fragmentation. Transition probabilities are estimated for both activity-to-rest and rest-to-activity states; this is the activity-to-rest variable. kAR is measured using a portable actigraph: AX3 (2020-present), GENEActiv (2018-2020), Actical (2005-2018). See the Actigraphy category description for detailed information on each device.

Participants are asked to wear the actigraph 24 hours per day for up to 10 days. The device records activity counts for 1-second samples and sums the counts into 15-second intervals called epochs. Each epoch is classified as resting (0 counts per epoch) or active (greater than 0 counts per epoch).

Runs of activity begin with at least one epoch of activity and end at the epoch before the first epoch of rest. That is, each run of activity begins with a rest-to-activity transition and ends with an activity-to-rest transition. For each individual record, the total numbers and duration of runs of activity are used to calculate the probability of transitioning from sustained activity to rest. kAR is then obtained by calculating the weighted average value of the probabilities. See reference for details on the calculation of kAR.

The higher the kAR, the more poorly sustained are runs of activity, and the lower the kAR, the more consolidated they are.

Note: Periods during which the actigraph was not worn is removed from analysis by flagging periods with greater than 4 consecutive hours of no movement and excluding the entire 24 hour period surrounding the flagged period.