Sleep disruption is common in neurodegenerative diseases and other CNS disorders and may be a contributor to brain pathology as well as its consequence. In addition to providing important outcome measures in studies where the medicinal product is expected to affect sleep or wakefulness, sleep quantity and quality might confound outcome measurement where changes in brain protein deposits are evaluated. The collection of sleep data during a clinical trial could thus provide important additional measures to aid the interpretation of results. Common sleep endpoints include time in bed, sleep onset latency, sleep duration, duration of wake after sleep onset, number of wake bouts and sleep efficiency.
IXICO collaborated with the Centre Hospitalier Universitaire, Montpellier, France on a study of sleep difficulties, lifestyle factors and general health. We analyzed actigraphy data from 26 participants (aged 83.9±7.2, 65% female) using the ESS, Cole-Kripke (CK) and McVeigh algorithms and our in-house developed Deep Learning Sleep (DLS) algorithm. We also analyzed sleep diary data from 22 participants (aged 80.6±9.9, 68% female) and compared similar measures from actigraphy and the diaries. All participants wore an Axivity 3-axis accelerometry device on their non-dominant wrist for 14 nights and were asked to complete sleep diaries for the same 2-week period. The DLS algorithm better correlated with polysomnography-derived sleep efficiency than the ESS and CK algorithms (r=0.82, r=0.52, r=0.64, respectively). Actigraphy-derived sleep efficiency correlated with self-reported sleep quality at r =0.16. McVeigh-estimated times in bed at night and self-reported times from into-bed to out-of-bed correlated at r =0.32. CK-estimated in-bed sleep times and self-reported times from lights-out to awakening (minus sleep onset latency) correlated at r =0.15. The relatively low correlations between sleep diaries and actigraphy measures may reflect poor reliability and inaccuracy of self-reported sleep measures.
Actigraphy also provides objective data on parameters relevant to assessing circadian rhythm sleep-wake disorders. Abnormalities of the sleep-wake rhythm are often seen in CNS disorders and include delayed sleep-phase, advanced sleep-phase, non-24-hour, and irregular sleep-wake rhythmicity. Endpoints of interest include the average (mesor), maximal point (acrophase) and daily length (period) of the circadian rhythm, the timing of peak average hourly activity, average activity and its variability during the peak hour, and the amount of movement during active periods (amplitude). As more stable daily rhythm is linked with favourable health outcomes, the intra-day variability and inter-day stability are amongst important endpoints.