Day and Night, Night and Day: Associations of Sleep and Activity Variability and Mortality

Self-reported sleep duration has been repeatedly found to predict mortality. Actigraphy has recently been added to population-based studies to provide more accurate sleep measures. Actigraphy sleep duration has not consistently predicted mortality, but actigraphy measures of sleep disruption measures are generally found to be predictive of mortality for older adults. A few studies have more fully used actigraphy data and constructed variables to summarize 24-hour activity patterns, which have also predicted mortality. In this study, we use a nationally representative study of Americans aged 61 – 91 to examine the associations between mortality and actigraphy-derived measures of variability, for both sleep and 24-hour activity patterns. We use 72-hour wrist actigraphy data from a substudy of the 2010/11 round of the National Social Life, Health and Aging Project (NSHAP) linked to the National Death Index (NDI) to establish 5-year mortality. Sleep variability was represented by sleep fragmentation and the standard deviation of wake and bed times. Intraday variability and between day (interday) variability described the 24-hour activity patterns. Cox proportional hazards models were adjusted for sociodemographic confounders and average daytime activity. In general, more variability was associated with increased death hazard for all measures. Fragmentation (HR: 1.04, 95% CI: [1.01, 1.07], p = 0.01), standard deviation of bedtimes (HR: 1.16, 95% CI: [1.02, 1.31], p = 0.02), and intraday variability (HR: 1.19, 95% CI: [0.98, 1.43], p = 0.07) showed the strongest associations. This study suggests that both consistent sleep and 24-hour activity patterns are associated with better prospective health.

Relationship between the Daily Rhythm of Distal Skin Temperature and Brown Adipose Tissue 18F-FDG Uptake in Young Sedentary Adults

The present study examines whether the daily rhythm of distal skin temperature (DST) is associated with brown adipose tissue (BAT) metabolism as determined by 18F-fluorodeoxyglucose (18F-FDG) uptake in young adults. Using a wireless thermometer (iButton) worn on the nondominant wrist, DST was measured in 77 subjects (26% male; age 22 ± 2 years; body mass index 25.2 ± 4.8 kg/m2) for 7 consecutive days. The temperatures to which they were habitually exposed over the day were also recorded. The interday stability of DST was calculated from the collected data, along with the intraday variability and relative amplitude; the mean temperature of the 5 and 10 consecutive hours with the maximum and minimum DST values, respectively; and when these hours occurred. Following exposure to cold, BAT volume and mean and peak standardized 18F-FDG uptake (SUVmean and SUVpeak) were determined for each subject via static 18F-FDG positron emission tomography/computed tomography scanning. Relative amplitude and the time at which the 10 consecutive hours of minimum DST values occurred were positively associated with BAT volume, SUVmean, and SUVpeak ( p ≤ 0.02), whereas the mean DST of that period was inversely associated with the latter BAT variables ( p ≤ 0.01). The interday stability and intraday variability of the DST were also associated (directly and inversely, respectively) with BAT SUVpeak ( p ≤ 0.02 for both). All of these associations disappeared, however, when the analyses were adjusted for the ambient temperature to which the subjects were habitually exposed. Thus, the relationship between the daily rhythm of DST and BAT activity estimated by 18F-FDG uptake is masked by environmental and likely behavioral factors. Of note is that those participants exposed to the lowest ambient temperature showed 3 to 5 times more BAT volume and activity compared with subjects who were exposed to a warmer ambient temperature.