Too Little, Too Much Sleep Tied to Impaired Cognition

Erik Greb

September 15, 2021

Short and long sleep duration in older individuals are both associated with impaired cognition, new research suggests. However, imaging data reveal only short sleep is linked to greater amyloid beta (Aβ) burden. 

In a large, cross-sectional study of patients older than 64 years, Mini-Mental State Examination (MMSE) scores were .08 points lower among those with short sleep duration vs individuals with normal sleep duration, while Digit Symbol Substitution Test (DSST) scores were 1.17 points lower in the long sleep group vs the normal sleep group.

Short sleep duration was defined as 6 hours or less and long sleep duration was defined as 9 hours or more.

Dr Joseph Winer

In addition, "we confirmed that short sleep was associated with amyloid — but we found that wasn't true in the long sleepers," lead investigator Joseph R. Winer, PhD, a postdoctoral research fellow in neurology at Stanford University, Stanford, California, told Medscape Medical News. "It's thought that people who are sleeping more than 8 hours probably have something else going on, too, like cardiovascular health issues."

The findings were published online August 30 in JAMA Neurology.

Multicenter Study

Aging is often accompanied by sleep changes that vary widely between individuals. Previous research shows a link between sleep disruption and cognitive decline and Alzheimer's disease (AD).

Sleep duration also influences cognitive health. Short sleep has been linked to greater amyloid beta burden, an outcome viewed as a preclinical stage of AD. However, small sample sizes and the low prevalence of long sleep duration data have hampered understanding of the potentially distinct factors between the two sleep types.

The researchers examined data from the Anti-Amyloid Treatment in Asymptomatic Alzheimer's Disease (A4) study, a multisite investigation with centers in Australia, Canada, Japan, and the United States.

Eligible participants were between ages 65 and 85 years and had normal cognition. In all, 6763 individuals were screened, and 4486 underwent a fluorine 18–labeled-florbetapir PET scan to measure amyloid beta.

Because amyloid beta and sleep duration vary between ethnic groups, the researchers used self-reported race or ethnicity as a covariate in the analyses.

To assess various cognitive domains, data were included from participants' performance on the MMSE, DSST, Logical Memory Delayed Recall test, and Free and Cued Selective Reminding Test (FCSRT).

Subjective reports of cognition were also measured using the Cognitive Function Index (CFI), which was administered separately to participants and their partners. Depressive symptoms were measured with the Geriatric Depression Scale (GDS).

U-Shaped Association  

A total of 4417 A4 participants were included in the analysis. In this population, 59% were women and the mean age was 71.3 years.

Higher Aβ burden was significantly associated with fewer hours of nightly sleep. Each additional hour of sleep was associated with .01 lower amyloid PET standard uptake value ratio (P = .005). Only short sleep duration was associated with increased Aβ burden.

There were no significant linear associations between sleep duration and cognitive test performance. However, when groups of participants with short, normal, or long sleep were examined, differences between groups were found that suggest nonlinear associations.

Among participants with short sleep, MMSE score was 0.08 points lower (P = .04) and Logical Memory Delayed Recall test score was 0.23 points lower (P = .03) than in participants with normal sleep. Those with short sleep also were less likely to achieve the maximum FCSRT total recall score compared with participants with normal sleep (P = .01).

In the long sleep duration group, DSST score was 1.17 points lower than in the normal sleep duration group (P = .02). In addition, participants with short sleep duration performed 0.73 points better on the FCSRT free recall than those with long sleep duration (P = .04).

Sleep duration was also associated with subjective cognitive function. Each additional hour of sleep duration was linked to a 0.06-point reduction in CFI, which indicated better self-reported cognitive function (P = .03). Short and long sleep durations were each associated with worse self-reported cognitive function (P < .001 for each association).

In addition, depressive symptoms had a linear association with shorter sleep duration. Each additional hour of sleep was associated with a reduction of 0.06 points on the GDS, indicating fewer depressive symptoms (P = .005). Participants with short sleep and those with long sleep reported more depressive symptoms than participants with normal sleep.

"Distinct Underlying Processes"  

"The short-sleeping group had test performance that was pretty consistent with Alzheimer's disease," said Winer.

"The long-sleeping group did worse on tests of executive function, which might suggest that they have a distinct underlying disease process from the short-sleeping group," he added.

It would have been desirable to find that long sleep duration was linked to a particular biomarker, like short sleep was associated with amyloid beta, Winer noted.

However, because there is only a small percentage of long sleepers in the general population, "it's really hard to get at this question in these studies with biomarkers," he said.

Previous study findings have shown an association between long sleep and dementia, and the current findings indicate that the form of dementia is not AD, Winer added.

Overall, the findings underscore the importance of monitoring sleep in healthy older adults.

"If a patient is having problems with their sleep or feels like they are sleeping less than they need, that can be an indicator of their disease risk," Winer said.

Interventions such as cognitive-behavioral therapy can improve sleep, he added.

"Certainly, sleep is something that people can improve, and I think that's an important conversation for providers to have with their patients," he noted.

In future studies, the researchers plan to examine objective measures of sleep, such as those gained from smart watches. They also intend to examine changes in the brain that are important for brain health, such as inflammation, which could have an association with sleep.

Finding the Tipping Point  

Dr Prashanthi Vemuri

Commenting on the study for Medscape Medical News, Prashanthi Vemuri, PhD, professor of radiology at Mayo Clinic in Rochester, Minnesota, noted the results conclusively show that a normal sleep duration is important for reducing amyloidosis in a stepwise fashion.

Among the study's strengths were its good representation of various races and ethnicities, its good sampling of the population, and the examination of cognition, BMI, and amyloidosis together, said Vemuri, who was not involved with the research.

The association shown between longer sleep duration and depressive symptoms is understandable, she noted.

"More sleep attributable to depression is bad because that inversely impacts systemic vascular health, so BMI would be high and executive function would be impaired, related to depression," said Vemuri.

A future analysis that examines participants according to the quality of their sleep would be interesting, she said. Other goals for future research could be to identify the tipping point after which dramatic amyloid deposition occurs and to examine how it relates to sleep quality, she added.

"I think it's important to catch people without symptoms and see when amyloid deposition really takes off," said Vemuri. "Longitudinal studies would be fantastic, with more granular measurement of sleep quality."

More broadly, additional information is needed about the interactions and longitudinal trajectories of sleep, cognitive health, and brain health, she concluded.

The study was funded by the National Institutes on Aging. Winer and Vemuri have disclosed no relevant financial relationships.

JAMA Neurol. Published online August 30, 2021. Abstract

Follow Erik Greb on Twitter: @MedscapeErik .

For more Medscape Neurology news, join us on Facebook and Twitter


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.