Automatic-Scoring Actigraph Compares Favourably to a Manually-Scored Actigraph for Sleep Measurement in Healthy Adults

Introduction Actigraphy has been used widely in sleep research due to its noninvasive, cost-effective ability to monitor sleep. Traditionally, manually-scored actigraphy has been deemed the most appropriate in the research setting; however, technological advances have seen the emergence of automatic-scoring wearable devices and software. Methods A total of 60-nights of sleep data from 20-healthy adult participants (10 male, 10 female, age: 26 +/- 10 years) were collected while wearing two devices concomitantly. The objective was to compare an automatic- scoring device ( Fatigue Science Readiband(TM) [AUTO]) and a manually-scored device (Micro Motionlogger (R) [ MAN]) based on the Cole-Kripke method. Manual-scoring involved trained technicians scoring all 60-nights of sleep data. Sleep indices including total sleep time (TST), total time in bed (TIB), sleep onset latency (SOL), sleep efficiency (SE), wake after sleep onset ( WASO), wake episodes per night (WE), sleep onset time (SOT) and wake time ( WT) were assessed between the two devices using mean differences, 95% levels of agreement, Pearson-correlation coefficients (r), and typical error of measurement ( TEM) analysis. Results There were no significant differences between devices for any of the measured sleep variables (p >= 0.05). All sleep indices resulted in very-strong correlations ( all r >= 0.84) between devices. A mean difference between devices of <1minutes for TST was associated with a TEM of 15.5minute (95% CI = 12.3 to 17.7 minutes). Conclusion Given there were no significant differences between devices in the current study, automatic- scoring actigraphy devices may provide a more practical and cost-effective alternative to manually- scored actigraphy in healthy populations.