Sleep disorders increase the risk of dementia, Alzheimer's disease, and cognitive decline: a meta-analysis

Abstract

Sleep disorders, particularly insomnia and obstructive sleep apnea, are increasingly implicated as significant contributors to cognitive decline, dementia, and neurodegenerative diseases such as Alzheimer's disease (AD) and vascular cognitive impairment and dementia (VCID). However, the extent and specificity of these associations remain uncertain. This meta-analysis evaluates the impact of common sleep disorders on the risk of developing dementia and cognitive decline. A comprehensive search of the literature was conducted to identify prospective cohort studies assessing sleep disorders and dementia risk. Studies reporting risk estimates for dementia, AD, or cognitive decline associated with obstructive sleep apnea, insomnia, and other sleep disorders (e.g., restless legs syndrome, circadian rhythm sleep disorders, excessive daytime sleepiness) were included. Meta-analyses were performed using a random-effects model to calculate pooled hazard ratios (HRs) and 95% confidence intervals (CIs). Thirty-nine cohort studies were included, with subgroup analyses showing significant associations between all-cause dementia and obstructive sleep apnea (HR 1.33, 95% CI 1.09-1.61), insomnia (HR 1.36, 95% CI 1.19-1.55), and other sleep disorders (HR 1.33, 95% CI 1.24-1.43). Obstructive sleep apnea increased the risk for AD (HR 1.45, 95% CI 1.24-1.69), though its association with vascular dementia did not reach statistical significance (HR 1.35, 95% CI 0.99-1.84). Insomnia was significantly associated with increased risk for both vascular dementia (HR 1.59, 95% CI 1.01-2.51) and AD (HR 1.49, 95% CI 1.27-1.74). This meta-analysis highlights the critical role of sleep disorders in dementia risk, emphasizing the need for early detection and management of sleep disturbances. Targeted interventions could play a pivotal role in reducing dementia risk, particularly among high-risk populations.

Keywords: Aging; Apnoe; Circadian rhythms; Cognitive decline; Inadequate sleep; Neurodegeneration; Semmelweis Study; Sleep deficit; Sleep-disordered breathing; Stroke.

Fig. 1

Flow diagram showing the study selection process

Fig. 2

Summary of hazard ratios (HRs) for the association between apnoe, insomnia, or other sleep disorders and all-cause dementia risk across multiple studies. Each row represents an individual study, displaying the study author(s) and year of publication, the log hazard ratio (logHR), standard error (SE), weight assigned to the study in the random-effects model, and HR with 95% confidence intervals (CI). The red squares denote the HR for each study, with square size reflecting study weight. Horizontal lines indicate the 95% CI for each study’s HR. The pooled HR estimate (diamond shape) for all sleep disorders shows a significant association with an increased risk of dementia. Heterogeneity across studies is indicated by I². Abbreviations: CI, confidence interval; HR, hazard ratio; IV, inverse variance; SE, standard error

Fig. 3

Funnel plots depicting the relationship between hazard ratios (HRs) and standard error (SE) for the association between various sleep disorders and different cognitive outcomes: all-cause dementia (A–C), vascular dementia (D–F), Alzheimer’s disease (G–I), and cognitive decline (J–L). The plots are organized into these four cognitive outcome categories, with each section examining the relationship with apnea, insomnia, and other sleep disorders. The funnel plot shape and symmetry can provide insights into potential publication bias, with asymmetrical plots suggesting the possibility of selective reporting or publication of studies

Fig. 4

Meta-analysis exploring the relationship between sleep disorders and vascular dementia, with subgroup analyses for apnoe, insomnia, and other sleep disorders. Hazard ratios (HRs) and 95% confidence intervals (CIs) were calculated using a random-effects model. The combined analysis demonstrates a significant link between sleep disorders and vascular dementia (HR = 1.71, 95% CI 1.33–2.20). Substantial heterogeneity is reflected by an I² value of 93%. Abbreviations: CI, confidence interval; HR, hazard ratio; IV, inverse variance; SE, standard error

Fig. 5

Forest plot depicting hazard ratios (HRs) for the association between sleep disorders and Alzheimer’s disease risk. The analysis includes 32 studies stratified into three subgroups: apnoe (n = 7), insomnia (n = 8), and other sleep disorders (n = 17). The overall pooled effect shows a significant increased risk (HR = 1.45, 95% CI 1.28–1.63). Diamond markers represent pooled estimates; squares represent individual study effects with size proportional to study weight. Abbreviations: CI, confidence interval; HR, hazard ratio; IV, inverse variance; SE, standard error

Fig. 6

Meta-analysis results showing associations between sleep disorders and risk of cognitive decline across three categories: apnoe (n = 13), insomnia (n = 8), and other sleep disturbances (n = 9). The combined analysis of 30 studies demonstrates an elevated risk (HR = 1.51, 95% CI 1.37–1.67), with moderate heterogeneity across studies. Individual study estimates are shown as squares, with pooled estimates represented by diamonds. Abbreviations: CI, confidence interval; HR, hazard ratio; IV, inverse variance; SE, standard error