Sleep disorders predict risk of dementia years before diagnosis, study shows

Sleep disorders predict risk of dementia years before diagnosis, study shows

New research finds that sleep disturbances may signal future risk of Alzheimer's, Parkinson's and other dementias years before symptoms, offering hope for early intervention and prevention.

In a study recently published in the journalNPJ dementiaThe researchers examined the neurodegenerative effects of clinically identified sleep disorders and related disorders in later life. They collected biobank data from more than 1 million participants in Finland, Wales and Great Britain (UK). Study results showed a significant association between these sleep disorders and several neurodegenerative diseases (NDDs), including dementia, Alzheimer's disease (AD), and Parkinson's disease (PD).

In particular, sleep disorders were able to predict NDD risk as early as 5-15 years before the disease was diagnosed. In Alzheimer's disease this risk appeared to be largely independent of genetic predisposition, while in Parkinson's disease an interaction with genetic factors was observed. These findings underscore the long-term impact of conditions such as sleep apnea and other formally identified sleep disorders and underscore the importance of sleep interventions in maintaining neurological health in late life.

background

Sleep is a nearly universal, fundamental biological process that is essential for optimal cognitive function and overall health. Several studies have established strong, bidirectional relationships between sleep and neurodegenerative diseases (NDDs), showing that certain sleep disorders and significant sleep disturbances can exacerbate both short-term cognitive impairment and long-term dementia risk.

As a result, the World Health Organization (WHO) has emphasized the importance of sleep as a critical health behavior, advocating research and interventions to address sleep disorders and improve sleep quality in diverse human populations. Unfortunately, sleep disorders are a common and growing global health problem, with reports estimating that 25% of all Europeans have insomnia.

Despite research elucidating multiple genetic and environmental factors in sleep disorders, the mechanisms underlying the role of sleep in NDD etiology are poorly understood. The extent to which specific, clinically recognized sleep disorders may predict NDD risk remains similarly inconclusive. Most studies examining sleep-NDD associations have limited sample sizes, inadequate follow-up durations, and focus on one of a few NDDs, complicating attempts to ascertain these outcomes.

About the study

The present study aims to further investigate the associations and potential causal links between sleep and NDDs by using a large medical electronic health record (EHRS) database of more than 1 million individuals in Finland, Wales and the United Kingdom (UK) and analyzing EHR data from a 20-year period (1999-2018). Study data were obtained from the secure anonymized information linkage database, Finngen datasets and the UK Biobank (UKB).

Participants' NDD and sleep disorder diagnoses were classified using the International Classification of Diseases 10th Revision (ICD-10) (e.g., G30 for Alzheimer's disease and G47.3 for sleep apnea) to ensure that the study focused on clinically documented conditions rather than per se reported symptoms. Cohort-specific medical histories were further used for statistical modeling and meta-analyses, including calculation of Cox proportional hazard ratios (HRS), polygenic risk scores (PRS), and logistic regression models.

To isolate the behavioral effects of sleep (exposure) on NDD, the models controlled for participants' genetics, age, gender, and other confounding variables. To facilitate the generalizability of the results and improve the accuracy of the results, all analyzes were replicated across multiple populations.

Study results

Regression and HR analyzes showed strong relationships between ICD-10 coded sleep disorders and a spectrum of late-life NDDs. Circadian rhythm-associated sleep disorders (ICD10 code G47, which include conditions such as insomnia, narcolepsy, sleep apnea and parasomnias) have been identified as significant risk factors in the subsequent development of Alzheimer's disease (AD; HR = 1.15), Parkinson's disease (PD), dementia and VASCular-dementia (HR = 1.4), disease (PD), and dementia and VASCular dementia and dementia and vascular Dementia and Dementia and Vascular Dementia and Dementia and Vascular Dementia and Dementia and Vascular Dementia and Dementia and VASC.

Non-organic sleep disorders (ICD10 code F51, such as nightmares and generalized insomnia not attributable to substances) were similarly associated with increased dementia (HR = 1.67), PD and vascular dementia (HR = 2.05). The study also found that the severity of certain sleep disorders, as indicated by recurrent clinical diagnoses, tended to increase the risk of some NDDs. While sleep apnea has been shown to be associated with amyotrophic lateral sclerosis (ALS), a lack of sufficient ALS data prevented the generalizability of these findings.

Notably, many identified associations persisted even after adjustment for genetic risk factors. In Alzheimer's disease in particular, the contributions of diagnosed sleep disorders to neurodegeneration risk appeared to be largely independent of genetic factors. However, in Parkinson's disease, the study found evidence of an interaction between genetic risk and certain sleep disorders.

Individuals with low genetic predisposition to NDDs still showed high NDD HRs associated with these sleep conditions, suggesting that such disorders are significant risk factors, particularly in individuals with lower genetic susceptibility.

All associations identified were observed to precede NDD diagnoses between 5 and 15 years, pointing to sleep assessments as an early indicator of future NDD risk. These results highlight the potential of sleep interventions in mitigating neurodegenerative diseases (NDDs) late in life and highlight the importance of early detection and treatment of sleep disorders to improve overall neurological well-being.

Conclusions

The present study uses the largest sleep data set to date to elucidate the relationships between clinically documented sleep disorders and late NDD risk. It analyzed 20 years of EHR data drawn from broader records of more than 1 million participants and found clear associations between such sleep disorders and late-life NDDs. These associations often persisted after participants' genetic predispositions were adjusted for diseases such as Alzheimer's, although interactions with genetic risk were found for Parkinson's disease.

While the use of a predominantly European cohort and exclusive EHR data (as opposed to blood tests) prevents the global generalizability of these results, this study represents an ideal first step in non-invasively addressing late-life neurodegeneration.

Notably, formally identified sleep disorders have been found to be accurate and stable predictors of future neurodegeneration risk, suggesting the assessment of disorders as an early indicator of AD, PD, dementia and vascular dementia, but also highlighting their treatment as modifiable and treatable pathways for healthy neurological aging.

Sources: