0390 MEASURING ACTIVITY BY ACTIGRAPHY - ARE WE USING THE RIGHT METHODS?

Actigraphy is commonly used in research to provide an objective measure of sleep. Many attempts to evaluate activity using this measure have been undertaken. These tend to involve day/night activity averages which disregards the variability of the hour-by-hour data and variability. To our knowledge, there have been no attempts to aggregate minute-by-minute activity data into hour-by-hour bins which may provide for a better description of any activity changes across the diurnal day. The present analysis focuses on this data aggregation approach. 15 patients (Mean Age=42.9 ± 10.4; f=12) with chronic pain (non-malignant neck or back pain) who received 8-session CBT-I were evaluated by 7-day actigraphy at baseline and post-treatment (only 10 subjects completed actigraphy at both time-points). Daytime activity averages were computed for baseline and post-tx. Daytime activity averages were evaluated using paired sample t-test. Hour-by-hour activity averages were computed for an average 24-hr period using activity count from minute-by-minute epochs over the 7-days. Mixed models were utilized to evaluate the difference between baseline to post-tx (phase) over the 24-hr period (time). Model included main effect for phase and time along with a phase*time interaction. Weekly activity averages showed no difference in average activity throughout the day between baseline and post-tx (Baseline Activity=376, Post-tx Activity=408.6, p=0.26). However, mixed models assessing the daily activity hour-by-hour revealed a significant main effect for time-of-day (p<0.001) and for phase (p=0.001). While the phase*time-of-day was not significant (p=0.25), descriptive data suggest marked improvement post CBT-I especially in the morning (6-11AM). Utilizing daily activity averages may result in a failure to resolve findings that are evident when assessing activity on an hour-by-hour basis. Further, arraying data in this manner may allow for an appreciation of how activity patterns across the diurnal phase of the 24-hr day. Supported by: NINR NR5R21NR009080-02.