By Ellen Feldman, MD
Altru Health System, Grand Forks, ND
SYNOPSIS: This long-term study found that people who reported sleeping an average of seven hours or less nightly at age 50 years and 60 years were 30% more likely to develop dementia than their counterparts who reported sleeping seven hours or more.
SOURCE: Sabia S, Fayosse A, Dumurgier J, et al. Association of sleep duration in middle and old age with incidence of dementia. Nat Commun 2021;12:2289.
Summary Points
- Data from 7,959 individuals with a mean age of 50.6 years in the 1980s was used as the basis for this 25-year study investigating an association between sleep duration in middle age and development of dementia in later years.
- With “short sleep” defined as less than six hours nightly, short sleep at age 50 years and 60 years (adjusted for multiple variables) was associated with a hazard ratio of 1.22 and 1.37, respectively, for development of dementia over the follow-up time period.
- The lowest incidence of dementia was found in participants reporting seven hours or more of sleep nightly; there was no association noted between sleep duration of greater than eight hours in mid-life and development of dementia in later years.
Years of research have confirmed the importance of sleep to overall health and have expanded our understanding of its medical powers.1 Yet, in many ways, the field still is in its infancy, with many questions remaining about the relationship between sleep and disease states.
In an effort to address one unanswered question, Sabia et al designed a 25-year prospective study using data from a national survey to investigate how and if sleep duration in years prior to onset of dementia is associated with development of the disease. The authors noted that there have been conflicting results from other studies looking at these relationships. They hypothesized that one limit to such other studies has been the relative short-term follow-up periods and the mixture of ages at the onset of the study.2,3
Dementia typically involves pathophysiologic changes over 15 to 20 years, thus, Sabia et al noted that shorter-term studies may confuse cause and effect. Additionally, the authors noted the importance of controlling for mental health, since disorders, such as depression, are important risk factors both for the development of dementia and sleep disorders (and sleep disorders often are a symptom of depression). The Whitehall II study, begun in the 1980s, periodically surveys a cohort of British civil servants to investigate reasons for a “social gradient” in health and outcomes.4 For the purposes of the Sabia et al investigation, questionnaire responses to the item, “How many hours of sleep do you have on an average weeknight?” were extracted at ages 50, 60, and 70 years (± 5 years) from 1985-1988, 1997-1999, 2002-2004, 2007-2009, 2012-2013, and 2005-2016. Three categories of sleep duration then were created, with short sleep defined as fewer than six hours a night, normal sleep defined as seven hours a night, and long sleep defined as eight hours or more a night. Accelerometer data was collected in a portion of the respondents in 2012-2013. This allowed a subgroup of participants with objective measurement of sleep to be included in the Sabia et al study.
Out of the initial 10,308 Whitehall II participants, 7,959 participants were eligible for the Sabia et al study. Eligibility required sleep duration information, relevant covariate information, and age 50 years (± 5 years; mean age 50.6 years) in the first phase of the study. Among this group, 521 developed dementia. The mean age of dementia diagnosis was 77.1 years of age (standard deviation [SD] = 5.6 years; range 53.4-87.6 years old). The number of participants fell over time, with 7,164 participants when statistical analysis was performed at age 60 years and 6,516 participants when the group reached 70 years of age.
Results
Results of the fully adjusted model (adjusted for multiple variables, including health-related factors, disorders of mental health, and sociodemographic variables) are displayed in Table 1. This shows the hazard ratio and P values for incident dementia for each age group, according to sleep duration when comparing short or long sleepers to individuals reporting normal sleep duration.
Table 1. Development of Dementia According to Age and Duration of Sleep
|
|
Age 50 Years
|
Age 60 Years
|
Age 70 Years
|
|
Short sleep duration
(≤ 6 hours)
|
HR = 1.22; P = 0.04*
|
HR = 1.37; P = 0.005*
|
HR = 1.24; P = 0.10
|
|
Normal sleep duration
(seven hours)
|
1 (referent)
|
1 (referent)
|
1 (referent)
|
|
Long sleep duration
(≥ 8 hours)
|
HR = 1.25; P = 0.07
|
HR = 1.15; P = 0.34
|
HR = 1.15; P =0.60
|
|
HR: hazard ratio
* Statistically significant value
|
Notably, short sleep appears to associated with an elevated risk of incident dementia later in life for the 50- and 60-year-old respondents. However, the association with short sleep and risk of incident dementia does not reach statistical significance for the 70-year-old respondents. Curious about this progression with age, Sabia et al next looked at changes in sleep duration over time and any association with development with incident dementia.
Table 2 displays the results when looking at this trajectory of sleep duration or changes in sleep duration over time and any connection or association with development of dementia. Notably, individuals reporting persistent short sleep duration had the highest risk of developing dementia.
Table 2. Trajectory of Change in Sleep Duration Between 50 and 70 Years of Age and Hazard Ratio for Development of Dementia
|
|
Hazard Ratio
(95% CI)
|
P Value
|
|
Change from short to normal sleep
|
1.20 (0.89-1.63)
|
0.23
|
|
Change from normal to long sleep
|
1.02 (0.73-1.42)
|
0.90
|
|
Change from normal to short sleep
|
1.13 (0.79-1.62)
|
0.50
|
|
Persistent short sleep
|
1.30 (1.00-1.69)*
|
0.048*
|
|
Persistent normal sleep
|
1 (referent)
|
|
|
Persistent long sleep
|
1.28 (0.88-1.85)
|
0.20
|
|
CI: confidence interval
* Statistically significant value
|
Included in the analysis were 3,888 respondents with objective measures of sleep (via accelerometer) in 2012-2013. This subgroup had a mean follow-up of 6.4 years. Patterns of association with incident dementia were similar to the self-reported data, with short sleep (≤ 6 hours/night) associated with a higher risk of incident dementia and normal or long sleep not associated with an elevated risk of dementia.
COMMENTARY
This long-term prospective study adds depth to our understanding of the relationship between sleep duration and dementia, specifically pointing toward an association between short sleep duration (defined as six hours or less) in middle-age and development of dementia years later.
Other observational studies in this field have linked both short and long sleep with a higher risk of dementia; however, interpretation and generalization of results are limited by the small number of participants, short follow-up periods, and difficulty controlling for confounding factors, especially disorders of mental health.2,3 Sabia et al designed this study with an eye on attacking all these limitations.
However, there do remain some limitations to this work, including the observational design, a relatively homogenous respondent pool (inherent to the Whitehall II respondents, all of whom were employed in public service at recruitment), and that most of the sleep data were based on self-report.
Future studies with more diverse populations incorporating “smart” devices monitoring sleep may be able to address a portion of these limitations and allow further advances in our understanding of the role of sleep in dementia.
Notably, dysregulation of sleep is common in many types of dementia. In Alzheimer’s dementia, sleep problems may appear before cognitive impairment, and progression of sleep disorders often aligns with progression of disease.5 Thus, as was mentioned earlier, the long-term design of this work provides a glimpse of sleep character prior to onset of pathophysiologic changes expected in the early stages of dementia. The question of causality remains; Sabia et al propose a bi-directional relationship between short sleep and incident dementia, but more research is needed to determine causality and whether increasing sleep time can reduce the risk of dementia.
Interestingly, Sabia et al focused only on sleep duration and not on sleep quality or restfulness of sleep. These other aspects of sleep also may be useful to investigate in future studies as these may be confounding variables.
It is useful to note that, although there are genetic risk factors for some types of dementia, the biggest overall risk factor for development of dementia is age.5,6 Thus, all patients in primary care practices are at risk of developing this progressive, debilitating disease process. Finding interventions for prevention with clinical relevance is a priority for individuals and populations alike.
It bears repeating that no causal relationship between sleep duration during middle-age and the development of dementia has been established by this study. However, a clinical message is nonetheless evident. When working with middle-aged patients, the primary care provider is in a good position to discuss sleep and, specifically, sleep duration. Given that there is no indication that improving sleep duration is associated with health risks, patients stand to benefit from knowing that improving sleep hygiene and prioritizing longer sleep duration can have positive long-term effects.
REFERENCES
- Worley SL. The extraordinary importance of sleep: The detrimental effects of inadequate sleep on health and public safety drive an explosion of sleep research. P T 2018;43:758-763.
- Westwood AJ, Beiser A, Jain N, et al. Prolonged sleep duration as a marker of early neurodegeneration predicting incident dementia. Neurology 2017;88:1172-1179.
- Larsson SC, Wolk A. The role of lifestyle factors and sleep duration for late-onset dementia: A cohort study. J Alzheimers Dis 2018;66:579-586.
- Marmot M, Brunner E. Cohort profile: The Whitehall II study. Int J Epidemiol 2005;34:251-256.
- Liguori C, Placidi F, Izzi F, et al. Sleep dysregulation, memory impairment, and CSF biomarkers during different levels of neurocognitive functioning in Alzheimer’s disease course. Alz Res Therapy 2020;12:5.
- Edwin TH, Strang BH, Persson K. Trajectories and risk factors of dementia progression: A memory clinic cohort followed up to 3 years from diagnosis. In Psychogeriatr 2021;33:779-789.
