Unstable Sleep Patterns Linked With Worsening Cognition

CHARLOTTE, N.C. -- People with greater fluctuations in their sleep duration were at greater risk of cognitive decline as they aged, a study found.

Based on data from over 1,100 participants of the Seattle Longitudinal Study (SLS), longitudinal sleep variability -- not sleep duration -- turned out to be a predictor of cognitive impairment, according to Samantha Keil, PhD, of the University of Washington School of Medicine in Seattle.

Indeed, the same average 7 hours of sleep on paper can be reported by sleepers who slept more hours later in life, those who slept less, and those who maintained a constant number of hours over time, among other patterns, she reported at SLEEP 2022, a joint meeting of the American Academy of Sleep Medicine and the Sleep Research Society.

Other predictors of cognitive impairment were education, APOE4 status, depression, and sleep restlessness. Sleep variability improved the concordance of a model of cognitive impairment, already accounting for depression, from 0.68 to 0.73, Keil said.

Ultimately, sleep variability (log[HR] 0.41, P=0.018) and depression (log[HR] 0.029, P=0.014) were predictors of cognitive performance 10 years later, she reported, adding that her group observed consistently worse survival at older age with increasing sleep variability and increasing depression scores on the Center for Epidemiologic Studies Depression Scale.

It's "surprising" how strong of an effect sleep variability seemed to have in this study, noted Yue Leng, PhD, of the University of California San Francisco. She added that the effect seemed "even larger than some of the other really important factors such as age and APOE4 status."

Leng, who was not involved in the study, cited a previous analysis of the Whitehall II cohort that showed that sleep duration remained relatively stable over 28 years without any evidence of differences in cognition between sleep trajectory groups.

For now, more evidence of sleep variability's effects is needed before findings can be translated into guideline recommendations, according to Leng. "It would be interesting to understand potential mechanisms, and particularly the characteristics of the people who have large sleep variability over the years."

The study by Keil and colleagues relied on data from the SLS, a study initiated in the 1950s. Participants have undergone regular cognitive testing since 1984, and answered health behavioral surveys since 1993.

Gender and ethnicity were not predictors of cognitive decline.

Andrew Varga, MD, of Mount Sinai Hospital in New York City, asked at the meeting about the potential impact that direction of sleep variability -- from more sleep to less sleep, or vice versa -- might have on the development of cognitive decline.

Keil acknowledged that the present study's statistical methods did not allow authors to analyze specific sleep trajectories, and this is being explored in follow-up analyses.

Her group also has some longitudinal brain imaging and EEG data that will be used in future studies investigating the link between sleep and cognitive function, she noted.

"Prior studies have rarely measured sleep duration longitudinally or examined the association between the change in sleep duration over the years and cognitive impairment. This study highlights the importance of longitudinal assessment of sleep duration and close monitoring of [its] change or variability in older adults," Leng commented.

Comment