0281 An Assessment of the Agreement Between Sleep Diary and Actigraphy in Firefighters

Abstract Introduction Sleep disruption is a common complaint for firefighters, largely thought to emanate from work and personal demands that prohibit recovery sleep. These disruptions increase the risk of sleep disorders, cardiovascular disease, cancer, and other disorders. Given the significance of sleep on health and the high frequency of sleep disruption in the fire service, we sought to examine the concordance between actigraphic and daily diary sleep assessments. The purpose of this analysis was to contribute information on the use of actigraphy and sleep diaries in shift workers individuals prone to insufficient sleep. Methods The sleep of 60 firefighters (n = 329 observations) working 24-hour shifts was assessed on a 6-day recovery period. Objective sleep was assessed via the Actiwatch-2, a research-grade, wrist-worn actigraph that measures motor activity in 30 sec intervals. Subjective sleep was assessed via the research consensus daily sleep diary, a self-reported measure collected upon awakening. Major sleep indices compared were: sleep onset latency (SOL), total sleep time (TST), sleep efficiency (SE), and wake time after sleep onset (WASO). Results Repeated effects Bland-Altman analysis using a mixed effects technique (measurement = fixed effect; participant = random effect) found that firefighters underestimated WASO (M = 32 min) and overestimated SE (M = 10%) at a level greater than the a priori clinical significance thresholds set by the American Academy of Sleep Medicine for insomnia. The limits of agreement for all sleep indices were very broad. For instance, 95% of the differences between self-report and actigraphic TST fell within a 4.7h range. The majority of the variability could be attributed to within-subject sources of variability versus between-subject sources of variability. Conclusion In firefighters, actigraphy and sleep diaries showed substantial disagreement in major sleep indices, with a systematic underestimation of WASO and overestimation of SE. The wide range of differences suggest that daily assessments of SOL, WASO, SE, or TST in on-call shift workers should be compared against a feasible gold-standard. Further research is needed to understand within-subject factors (e.g., daily differences) that predict a discrepancy between actigraphic and self-reported sleep in different working populations. Support (if any) #1R01HL117995-01A1