Measurement of Sleep in Chronic Traumatic Brain Injury: Relationship Between Self-report and Actigraphy

To examine the relationship between self-report and actigraphy measurement of sleep in people with and without traumatic brain injury (TBI) by addressing 2 aims: (1) to assess the relationship between self-report and actigraphy for sleep quantity in people with and without TBI; and (2) to explore how self-report and actigraphy capture sleep quality in TBI.

Participants completed the study over 2 weeks in their own homes. They wore activity monitors, day and night, throughout the experiment and completed morning sleep diaries while interacting with an experimenter on videoconference.

This project was embedded in a larger study on sleep and word learning in 100 adults: 50 with chronic, moderate-severe TBI and 50 demographically matched noninjured peers. Of the 100 participants who completed the larger study, 92 participants (45 with TBI and 47 noninjured peers) had sufficient actigraphy data for inclusion in the current study.

We used multilevel linear regression models and correlation analyses to assess how well participants' self-report corresponded to actigraphy measurement of sleep.

Actigraphy measures included nightly sleep duration and nighttime wakeups. Sleep diary measures included self-reported nightly sleep duration, nighttime wakeups, sleep quality, and morning fatigue.

People with and without TBI did not differ in the relationship between self-reported and actigraphy measurement of sleep quantity. Performance on a neuropsychological memory assessment did not correlate with the difference in self-reported and actigraphy-measured sleep in the TBI group. Sleep characteristics that were measured by actigraphy did not predict subjective experiences of sleep quality or fatigue.

Short-term self-report diaries capture accurate information about sleep quantity in individuals with TBI and may support self-report of other daily habits. Future research is needed to identify reliable metrics of sleep quality, and how they relate to other domains such as memory and mood, in the chronic phase of TBI.