0022 Characterization and Evaluation of Digital Dim Light Melatonin Onset in a Population-Based Sample

Abstract

Abstract Introduction Misalignment between circadian phase (i.e., timing) and the desired sleep window is associated with sleep disturbances including insomnia and circadian rhythm sleep-wake disorders. Identifying circadian-driven sleep disruption requires assessment of circadian phase, the gold-standard of which is dim light melatonin onset (DLMO). While DLMO is traditionally measured in the lab using saliva or blood sampled over multiple hours, novel methods of estimating DLMO using activity and light data obtained from wearable sensors (i.e., digital DLMO) have been recently validated in samples with both healthy and disordered sleep. Such methods could potentially provide pragmatic, low-burden ways of assessing circadian phase, which in turn could be used to aid diagnostic decisions and to adjust circadian-targeted interventions in real time. However, digital DLMO has yet to be characterized within the broader adult population, which is needed to differentiate between normative and clinically salient values. Therefore, this study will examine digital DLMO in a large, population-based sample, as well as explore the potential clinical utility of digital biomarkers related to circadian phase. Methods This study will be a secondary analysis of data obtained during the ancillary sleep study of the Multi-Ethnic Study of Atherosclerosis (i.e., MESA Sleep). Seven days of activity and light data, measured via wrist-worn actigraphy, were obtained from 2,237 participants. Digital DLMO will be estimated using the extended Kronauer limit-cycle model of the human circadian pacemaker. Phase angles, or the differences in clock time, between digital DLMO and the following will be calculated: self-reported in-bed time; and sleep onset, midpoint, and offset measured by actigraphy. The associations between these phase angles and sleep outcomes (i.e., sleep onset latency, sleep efficiency, and sleep quality) will be explored using linear regression. Covariates will include the age, gender, and race/ethnicity of the participant. Weekend and weekday averages will be evaluated separately. Results Results will include the distribution of digital DLMO in a population-based sample, as well as the associations between digital DLMO phase angles and sleep outcomes. Conclusion This study will inform future research into the clinical potential of digital circadian biomarkers. Support (if any)