Measuring Daily Diurnal Rhythms Using Passively Collected Smartphone Data (Preprint)

BACKGROUND Irregularity in circadian, diurnal, and social rhythms have been associated with adverse health outcomes. Regularity of rhythms can be quantified using passively collected smartphone data to provide clinically relevant biomarkers of routine. OBJECTIVE To develop a metric to quantify the regularity of diurnal rhythms and evaluate the relationship between routine and mood as well as demographic covariates. METHODS Passively sensed smartphone data from a cohort of 38 individuals from Penn/CHOP Lifespan Brain Institute and Outpatient Psychiatry Clinic at the University of Pennsylvania was fitted with two-state continuous-time hidden Markov models (CT-HMMs), representing active and rest states. Regularity of routine was modeled as the hour of the day random effects on probability of state transition, i.e. the association between hour-of-day and state membership. A regularity score, Diurnal Rhythm Metric (DRM), was calculated from the CT-HMMs and regressed on clinical and demographic covariates. RESULTS Regular diurnal rhythms were associated with longer sleep durations (P=.0088), older individuals (P=.001) and less-severe depression (P=.0496). CONCLUSIONS Passively sensed DRMs are comparable to the existing Social Rhythm Metrics but do not require burdensome survey based assessments. Low-burden, passively sensed metrics based on smartphone data are a promising and scalable alternative to traditional measurements.