April 4, 2024 | A wearable device measuring movement at the wrist can detect clear differences in people who do and don’t have beta-amyloid deposition in their brain, according to Adam Spira, Ph.D., a longtime sleep researcher with the Johns Hopkins Bloomberg School of Public Health. The discovery was enabled by a relatively new statistical technique revealing time-of-day changes in activity levels not captured by standard methods focused on average levels of activity across the day.
In a pair of studies, each comparing an amyloid-positive cohort to an amyloid-negative group, Spira and his colleagues found differences in mean activity in certain afternoon periods and in variability of activity across days in a broader range of time windows. The first study, published in 2021 in SLEEP Advances (DOI: 10.1093/sleepadvances/zpab007), compared sleep and circadian rest/activity rhythms (RARs) between controls and participants with mild cognitive impairment who were part of the Anti-Amyloid Treatment in Asymptomatic Alzheimer’s and the Longitudinal Evaluation of Amyloid Risk and Neurodegeneration studies.
The second study, recently published in SLEEP (DOI: 10.1093/sleep/zsae037), intentionally set out to replicate findings from the initial exploratory analysis with a new group of 82 community-dwelling individuals whose average age was about 76. Notably, higher afternoon activity and lower afternoon variability echoed the results of the prior study.
No differences were detected between the amyloid-positive and amyloid-negative groups in the studies using standard sleep metrics looking at averages, points out Spira. The time-of-day variances between the two cohorts were found by examining activity count data over the course of a 24-hour day using a function-on-scalar regression model, described in a 2015 article in Biometrics (DOI: 10.1111/biom.12278), which doesn’t impose the same assumptions on the data.
Actigraphs are small motion sensor detectors (accelerometers) that measure activity based on movement. The signals generated get translated into an activity count every 30 to 60 seconds, with the higher counts representing greater activity, he explains.
Spira’s interest in circadian RARs reflects mounting evidence in the scientific literature over the last 15 years linking sleep disturbances to a greater risk of Alzheimer’s disease. His research often employs research-grade wrist actigraph devices validated against polysomnography (sleep study)—in this case, the Actiwatch 2 from Philips Respironics that has been popular with sleep researchers but was recently discontinued.
Actigraph devices provide “a sense of when people are asleep and when they are awake based on their patterns of movement,” says Spira. And sleep is governed in part by the brain’s circadian clock, which generates the approximately 24-hour rhythms of physiological systems such as core body temperature, gene expression, and hormone release.
The central pacemaker of the circadian timing system is the suprachiasmatic nucleus in the hypothalamus, he continues. For processes to qualify as truly circadian, it needs to persist “even in the absence of environmental cues that would influence them.”
One of the biggest environmental stimuli for synchronizing the circadian clock is bright light, which serves as a “zeitgeber” (time giver), as scientists like to say. The light/dark cycle is what helps keep people’s circadian rhythms in sync with the natural environment, optimizing their health, says Spira.
Wrist actigraphs provide a 24-hour signal that is distant from the internal circadian rhythm but offer useful information about the timing of behavior that may be “circadian-adjacent,” he says.
It is well appreciated that the 24-hour rhythms of people with more advanced Alzheimer’s disease often are different than those of individuals without it. In mouse models of Alzheimer’s disease, researchers have shown changes in sleep and wake patterns as amyloid begins to aggregate in the brain, says Spira. It has also been the clinical experience of clinicians caring for Alzheimer’s patients, and their caregivers, that sleep/wake disturbances are a common problem—as well as the primary reason for placing a loved one in a care facility.
Amyloid likely plays a role in disturbing normal sleep rhythms, he says, although it could also be related to neurological changes accompanying that plaque buildup that wasn’t the focus of measurement in the actigraphy studies. “Eventually, it would be great if we could develop a digital signature that is more sensitive and specific.
“This is just one signal that... on its own is probably of limited utility,” he continues. “However, it points to the potential promise of sensors, perhaps multiple sensors at some point, to provide early warnings.”
People are already sending data from their smartwatches to their physician, says Spira. In the future, they might also be able to transmit digital signals that could be compiled to form a risk index for Alzheimer’s disease so treatments—once they arrive—can be delivered at an earlier stage when they may well be more effective.
Spira says he is hopeful that others will replicate the findings of the actigraphy studies and determine whether time-of-day-specific RAR features might serve as markers of preclinical beta-amyloid deposition and if they predict clinical dementia and agitation starting in the late afternoon—what is commonly referred to as “sundowning.”
The sundowning phenomenon is not well understood, although some people think it may be related to older adults having sensory loss as well as cognitive impairment that makes them more vulnerable to confusion when the light gets low, Spira says. Others hypothesize that it has something to do with anxiety around changing shifts at nursing facilities.
But the fact that Alzheimer’s patients frequently experience afternoon agitation supports the idea that activity level changes seen in the same period might be signaling preclinical sundowning, he says. “Part of it could be just the result of... restlessness” that with greater neurodegeneration devolves into clinically recognized sundowning syndrome.
Next steps for Spira and his team could include examining the extent to which differences in the patterns differentiating the amyloid-positive and amyloid-negative folks are predictive of future changes in cognition and brain health, and if those exhibiting more pronounced differences have poorer outcomes or faster disease progression. Another possibility, he adds, is to follow individuals exhibiting time-specific activity features to see if those who become cognitively impaired are more likely to develop agitation during the day.
In the interim, Spira advises anyone who will listen to prioritize their sleep. “I am a big advocate for good, healthy sleep. We think it plays an important role not only with respect to Alzheimer’s disease, but in health more broadly.”