A fresh look at the use of nonparametric analysis in actimetry

Abstract

Actimetry has been used to estimate the sleep-wake cycle instead of the rest-activity rhythm. Although algorithms for assessing sleep from actimetry data exist, it is useful to analyze the rest-activity rhythm using nonparametric methods. This would then allow rest-activity rhythm stability, fragmentation and amplitude to be quantified. In addition, sleep and wakefulness efficiency can be quantified separately. These variables have been used in studies analyzing the effect of age, diseases and their respective treatments on human circadian rhythmicity. In this study, we carried out a comprehensive analysis of the main results from published articles and devised a functional model of interaction among the several components involved in generating the sleep-wake cycle. The nonparametric variables render it possible to infer the main characteristics of circadian rhythms, such as synchronization with a zeitgeber, and its amplitude and robustness.

Adolescents at clinical-high risk for psychosis: Circadian rhythm disturbances predict worsened prognosis at 1-year follow-up

“…Circadian rhythms variables were derived from the actigraphy data using the method described by Castro and colleagues (2015), including: autocorrelation function (Ac; slope of the time correlation line (log-transformed); indicative of rhythm fragmentation; lower values represent less fragmented rhythms), interdaily stability (IS; value for 1440 min provided by Sokolove–Bushell periodogram analysis; denotes synchronization of circadian rhythms with the light/dark cycle), intradaily variability (IV; mean of the first derivative of the actigraphy data normalized by the total variance; a measure of rest–activity rhythm fragmentation), M10 (mean activity level during the most active 10 hours of the day; higher values are indicative of a more active lifestyle), F10 (onset of M10), L5 (sum activity during the least active 5 hours of the day), F5 (onset of the least active 5 hour period/nocturnal activity), and relative amplitude (RA = (M10-L5)/(M10+L5); measures the relationship between diurnal amplitude and night amplitude, the maximum value of 1 occurs when there is no movement during the night) (Chiesa et al, 2010; Eke et al, 2002; Goncalves et al, 2015; Goncalves et al, 2014; Richman and Moorman, 2000; van Beek et al, 1989). …”

Blue-Enriched White Light Enhances Physiological Arousal But Not Behavioral Performance during Simulated Driving at Early Night

“…The rhythms of motor activity, body position, wrist temperature, TAP and environmental light were then subject to non-parametric analyses (Gonçalves et al, 2015) providing several indicators referring to sleep and circadian rhythms characteristics:…”

Circadian rest–activity rhythm in individuals at risk for psychosis and bipolar disorder

“…Interdaily stability (IS) and intradaily variability (IV) were measured using the new protocol described in Goncalves et al (2015). The average activity during the most active 10 h period (daily activity) is represented by M10.…”

“…The beginning of M10 is F10. The beginning of the least active 5 h period (nocturnal activity) is F5 (Goncalves et al, 2015;Gonçalves et al, 2014). The Ac function calculated as the slope of the line obtained in logarithmic scale is the given autocorrelation and Ac is related to rhythm fragmentation and values closer to zero indicate a stronger correlation (less fragmented rhythm).…”