Keywords
Sleep. Dementia. Mild cognitive impairment. Systematic Review. Meta-analysis. Protocol.
While sleep disturbance is common in advanced dementia, leading to negative outcomes, there is growing evidence that sleep disturbance begins early in prodromal dementia and may contribute to cognitive decline. Sleep is therefore an important treatment target throughout the natural history of dementia.
The existing literature regarding interventions for sleep in people with cognitive impairment is beset by wide heterogeneity in the sleep measurement tools used and outcome measures reported, which limits data synthesis and impedes the development of a robust evidence base for these interventions.
We propose to undertake a systematic review and meta-analysis to assess the evidence base from clinical trials for both pharmacological and non-pharmacological interventions to improve sleep in people with cognitive impairment. We will also extract data regarding the sleep measurement tools and outcome measures used in these studies to underpin the development of a core outcome set for future clinical trials in the area.
This systematic review and meta-analysis will include randomised and non-randomised controlled clinical trials that assess interventions to improve sleep in participants with mild cognitive impairment or dementia. Medline, CINAHL, PsycINFO, and the Cochrane CENTRAL database will be searched from inception to present. A flow chart will illustrate the study selection process in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-analysis (PRISMA) guidelines.
The methodological quality and risk of bias of included studies will be assessed. Heterogeneity will be tested. A narrative synthesis will be undertaken where meta-analysis is not feasible.
This systematic review and meta-analysis will help inform clinical practice by evaluating the available evidence for interventions to improve sleep in people with mild cognitive impairment or dementia. It will also facilitate the development of a core outcome set that will improve the coherence and utility of data emerging from future clinical trials.
Sleep. Dementia. Mild cognitive impairment. Systematic Review. Meta-analysis. Protocol.
Sleep is important for good physical and mental health1. Sleep patterns change as people get older2, however, with significant declines identified in Non-Rapid Eye Movement (NREM) slow wave sleep and sleep efficiency3. The prevalence of such sleep disturbance is even greater among older people with cognitive impairment4, and its prevalence continues to increase as cognitive impairment advances5.
While the association between sleep disturbance and cognitive impairment is well established6, the relationship has not always been clear. While sleep disturbance was previously presumed to be a symptom of cognitive impairment7, the relationship appears to be more complex with studies now suggesting that it may be both causal and bi-directional8.
There is, for example, plausible evidence that sleep disturbance plays a role in precipitating or accelerating deposition of the beta-amyloid and tau proteins that are characteristic of the pathophysiology of Alzheimer’s Disease (AD)9. Cerebrospinal fluid levels of beta-amyloid fluctuate diurnally, rising during wakefulness before declining during sleep10. The glymphatic system that clears toxic metabolic waste products from the brain is more efficient during sleep11. Sleep deprivation has been shown to increase accumulation and cortical deposition of both beta-amyloid12 and tau13. A recent meta-analysis suggests that up to 15% of cases of AD could be attributable to sleep disturbance14.
Although the recently published Dementia Prevention, Intervention and Care: 2024 report of the Lancet Commission15 has reaffirmed that additional research is needed to clarify the effect of sleep on cognitive decline, interest is nevertheless growing in sleep as a modifiable risk factor that may be a target for preventative intervention in mild cognitive impairment (MCI) or prodromal dementia16.
Improving sleep remains an important treatment target for patients with more advanced dementia. Regardless of the underlying reason, sleep disturbance tends to worsen with advancing dementia17, leading to strained relationships with caregivers who experience considerable physical and emotional distress and burnout as a consequence18. The increased carer burden caused by sleep disturbance in people with dementia has been shown to lead to sub-optimal care19 and early institutionalisation20.
A 2014 Cochrane review of pharmacological interventions for sleep in people with AD21 included only five studies, illustrating the dearth of research data to guide clinical decisions among this cohort. The review found that, although it had a favourable side-effect profile, melatonin did not have any positive benefit on sleep parameters. The atypical anti-depressant, trazadone, did result in significantly improved nocturnal total sleep time (NTST) and sleep efficiency, with no appreciable effect on daytime sleep or cognition. A 2020 Cochrane review of pharmacotherapies in dementia22 included nine studies, including two additional studies involving orexin antagonists that provided moderate certainty evidence of improved NTST and wakefulness after sleep onset. A more recent systematic review from 2024 involving participants with AD23 included 14 studies. Trazadone again demonstrated positive effects in treating insomnia in people with AD, with several studies also indicating a noticeable improvement in cognition. The sedative-hypnotic, eszopiclone, and orexin antagonists were both found to improve sleep efficiency and total sleep time, while melatonin did not. Although zopiclone, another sedative-hypnotic medication, improved NTST and the number of night-time awakenings, it exhibited a worrisome side-effect profile with risk of increased confusion, agitation and hallucinations. Each of the aforementioned reviews were limited by the small sample sizes in the included studies. The latter review concluded by saying “the research landscape surrounding sleep disturbances in people with AD is characterised by a conspicuous gap in knowledge and notable challenges in establishing a consensus regarding the most effective and safest pharmacological intervention”.
A 2019 systematic review and meta-analysis of non-pharmacological interventions to improve sleep in people with cognitive impairment24 found modest improvements in sleep efficiency with bright light therapy (BLT) and multi-domain interventions, usually incorporating BLT, but only when all data were pooled. Data were small in number, however, mainly involving residents in institutional care and few data were available for the effectiveness of these interventions in people with MCI. Further, most of the small number (n=3) of randomised controlled trials included in the meta-analysis showed unclear risk of bias. A 2023 Cochrane review of non-pharmacological treatments for sleep in people with dementia25 included 19 studies, 14 of which were conducted in nursing homes. It concluded with low certainty that physical and social activity interventions may improve NTST and sleep efficiency, while multi-modal interventions may improve NTST. The effects of BLT were uncertain, with conflicting results in the included studies.
A common thread running throughout reviews of both pharmacological and non-pharmacological interventions to improve sleep in people with cognitive impairment is the wide variety in research methodology, particularly in relation to the outcome measures employed21–26. This heterogeneity limits the comparability of data from different studies and impedes the development of a reliable evidence base to guide clinical practice.
Given the importance of sleep as a treatment target throughout the natural history of dementia, to potentially slow progression of early-stage disease and to reduce symptom burden in more advanced cases, we propose to undertake an updated systematic review and meta-analysis of clinical trials involving both pharmacological and non-pharmacological interventions to improve sleep in people with MCI or dementia. This will clarify the evidence currently available to support clinical treatment decisions. Furthermore, we will collate data regarding the sleep measurement tools used and the sleep outcome measures reported in included trials. This will inform the development of a core outcome set that will improve the coherence and comparability of data emerging from future clinical trials.
This protocol for a systematic review and meta-analysis will answer the following question: “What is the evidence from randomised and non-randomised clinical trials to support interventions that aim to improve sleep in people with MCI or dementia?”
The PICO Framework27 will be employed to examine randomised and non-randomised clinical trials among participants with MCI or dementia that assessed pharmacological and non-pharmacological interventions designed to improve sleep by comparing the studied intervention to at least one other intervention/non-intervention/placebo control group in which the goal of improving sleep was the desired outcome. (Table 1).
MCI = Mild cognitive impairment.
The Preferred Reporting Items for Systematic Review and Meta-Analyses - Protocols (PRISMA-P) guidelines28 were followed in producing this protocol (see completed PRISMA-P checklist in Table 2 [see Extended Data]). The protocol has been registered on the International Prospective Register of Systematic Reviews (PROSPERO), registration number: CRD42024556750.
This systematic review and meta-analysis will include randomised and non-randomised controlled clinical trials that assess pharmacological and non-pharmacological interventions to improve sleep in people with MCI or dementia.
Inclusion criteria
original peer reviewed articles
written in the English language
reporting on clinical trials of interventions to improve sleep, involving:
participants:
■ aged ≥18 years.
■ at least 80% of whom have a stated diagnosis of MCI or dementia or, if no such diagnosis is stated, at least 80% of whom score within the generally accepted threshold for cognitive impairment on standard neurocognitive testing or, if said percentage is impossible to calculate based on the information available, the standard deviation/interquartile range of scores on standard neurocognitive testing is within the generally accepted threshold for cognitive impairment. Trials will also be included if a distinct sub-group of participants, for whom results are separately reported, meets this inclusion criterion.
comparison of the studied intervention with at least one other intervention/non-intervention/placebo control group. While cross-over trials will be included, within-subjects trials that consecutively compare the studied intervention with another intervention or control within the same cohort of participants will not be included.
at least one primary or secondary outcome relating to sleep. Trials that use composite assessments involving sleep as outcome measures will be included only if the sleep component of the assessment is separately reported. For example, trials that utilise the Neuropsychiatric Inventory (NPI)29, a measure of non-cognitive symptoms in dementia, as an outcome measure will be included only if the score pertaining specifically to the sleep domain of the NPI is reported.
Exclusion criteria
The following databases will be searched from inception to date: Medline (Ovid), CINAHL, PsycINFO, and the Cochrane CENTRAL database. The search strategy for each database is detailed separately in Tables 3a–3d (see Extended Data). The interTASC Information Specialists Sub-Group (ISSG) Search Filter Resource was used to filter for clinical trials30. Given the high prevalence of cognitive impairment among nursing home residents31,32, it was felt prudent to include ‘nursing home resident’ (and synonymous terms) with cognitive impairment (and synonymous terms) in order to ensure capture of trials conducted in nursing homes that did not explicitly reference participants’ cognitive diagnoses but may nevertheless meet our inclusion criteria by reporting cognitive assessment scores. To illustrate this point, in a 2018 systematic review of non-pharmacological interventions to improve night-time sleep among residents of long-term care settings33, over half of included studies involved participants with dementia but this was not always explicitly stated in the studies themselves34. The search strategy was reviewed in accordance with the Peer Review of Electronic Search Strategies (PRESS) Guideline Statement35. Table 4 illustrates the predicted outcome for this search strategy (see Extended Data).
Search results will be merged using reference management software and duplicate records will be removed. Remaining titles will be reviewed to remove articles that are obviously irrelevant. Abstracts of retained articles will then be examined against eligibility criteria and those that are obviously irrelevant will be removed. Finally, the full text of remaining articles will be reviewed independently by two separate reviewers to determine which articles comply with eligibility criteria. Any disagreements regarding inclusion of articles will be resolved by discussion and consensus agreement, where possible, or by independent review by the project supervisor (ROC), where necessary.
Data will be extracted from the included articles using standardised forms that will facilitate recording of the data set out in Table 5 (see Extended Data).
Methodological quality of included studies will be assessed using the Joanna Briggs Institute Checklist for Quasi-Experimental and Randomised Controlled Trials36. Risk of bias will be assessed using the Cochrane Collaboration’s Tool for Assessing Risk of Bias37.
Heterogeneity will be tested using Higgins’ I2 statistic38. It is expected that there will be substantial heterogeneity both in the tools used to measure sleep and in the outcome measures reported26. Meta-analyses will be undertaken, where possible, if appropriate conditions are met in certain subgroups. Quantitative data will, where possible, be pooled in statistical meta-analysis to perform data synthesis. The significance threshold will be p<0.05 on two sides. A forest plot for each parameter will be constructed to indicate the weight ratio of each incorporated trial. Where meta-analysis is not feasible, a narrative synthesis will be undertaken focussing on population characteristics and the performance of interventions compared to controls.
Based on the results of previous systematic reviews and meta-analyses21–26, there is likely to be high levels of heterogeneity in sleep measurement tools used and sleep outcome measures reported, which will limit the scope for meta-analysis. Furthermore, studies involving participants with cognitive impairment are prone to inaccuracies if data collection is reliant on subjectively reported outcome measures. Limiting the scope of this study to four databases may result in some relevant articles being missed, although these are commonly accessed for studies examining sleep based on previous similar studies24.
Despite the need for further evidence to clarify the relationship between sleep disturbance and cognitive impairment15, there is growing interest in the possibility that improving sleep in the early prodromal stages of dementia may help prevent or delay cognitive decline16. Sleep disturbance is common at all stages of established dementia5,17, leading to negative outcomes for patients19,20 and their caregivers18.
Previous systematic reviews and meta-analyses have been piecemeal, either focussed entirely on pharmacological22 or non-pharmacological24 interventions, or have been restricted to specific causes23 or stages26 of dementia. Furthermore, all of these reviews have illustrated a relative dearth of existing research and have been limited by high levels of heterogeneity impeding data synthesis and the derivation of a reliable evidence base to guide clinical practice. The treatment of sleep in people with cognitive impairment is a dynamic area of increasing research interest, with growing evidence supporting its role, so it is important to update the aforementioned reviews with one review encompassing both pharmacological and non-pharmacological interventions to improve sleep throughout the spectrum of cognitive decline from MCI to dementia. This updated study will help chart the topography of the current research landscape, thus clarifying the existing evidence base and helping to identify gaps in the current knowledge that need to be bridged by future research.
Given the apparent paucity of relevant research heretofore, and the challenges inherent in conducting research among people with cognitive impairment39, it is important to ensure that future research is coherent so that its utility can be maximised. An ancillary objective of this review, therefore, will be to identify the sleep measurement tools used and the outcome measures reported in included trials so as to inform the development of a core outcome set that will guide the conduct of future clinical trials.
The planned systematic review and meta-analysis outlined in this protocol will help inform clinical practice by evaluating the available evidence for both pharmacological and non-pharmacological interventions to improve sleep in people with MCI and dementia. Additionally, the review will contribute to the development of a core outcome set for future clinical trials involving interventions to improve sleep in people with cognitive impairment, thus improving the coherence and utility of data emerging from those trials.
This systematic review and meta-analysis does not require ethical approval due to absence of human participants. Results of the systematic review will be published in a peer-reviewed journal and presented at relevant academic conferences.
Open Science Framework: A protocol for the systematic review and meta-analysis of clinical trials of interventions to improve sleep in people with mild cognitive impairment or dementia. https://doi.org/10.17605/OSF.IO/HXTYW
This article contains the following extended data40:
Tables 3a-d (Search strategies for Medline, the Cochrane CENTRAL database, CINAHL, and PsycINFO).
Table 4 (Search Strategy Predicted Outcome)
Table 5 (Data Extraction Form)
Open Science Framework: A protocol for the systematic review and meta-analysis of clinical trials of interventions to improve sleep in people with mild cognitive impairment or dementia. https://doi.org/10.17605/OSF.IO/HXTYW
This article contains the following reporting guidelines40:
Table 2 (Completed PRISMA-P Checklist)
PRISMA Flowchart
Data are available under the terms of the Creative Commons Zero "No rights reserved" data waiver (CC0 1.0 Public domain dedication).
Is the rationale for, and objectives of, the study clearly described?
Partly
Is the study design appropriate for the research question?
Partly
Are sufficient details of the methods provided to allow replication by others?
Yes
Are the datasets clearly presented in a useable and accessible format?
Not applicable
Competing Interests: No competing interests were disclosed.
Reviewer Expertise: Sleep with ageing, dementia, caregiving
Alongside their report, reviewers assign a status to the article:
Invited Reviewers | |
---|---|
1 | |
Version 1 07 Oct 24 |
read |
Provide sufficient details of any financial or non-financial competing interests to enable users to assess whether your comments might lead a reasonable person to question your impartiality. Consider the following examples, but note that this is not an exhaustive list:
Sign up for content alerts and receive a weekly or monthly email with all newly published articles
Register with HRB Open Research
Already registered? Sign in
Submission to HRB Open Research is open to all HRB grantholders or people working on a HRB-funded/co-funded grant on or since 1 January 2017. Sign up for information about developments, publishing and publications from HRB Open Research.
We'll keep you updated on any major new updates to HRB Open Research
The email address should be the one you originally registered with F1000.
You registered with F1000 via Google, so we cannot reset your password.
To sign in, please click here.
If you still need help with your Google account password, please click here.
You registered with F1000 via Facebook, so we cannot reset your password.
To sign in, please click here.
If you still need help with your Facebook account password, please click here.
If your email address is registered with us, we will email you instructions to reset your password.
If you think you should have received this email but it has not arrived, please check your spam filters and/or contact for further assistance.
Comments on this article Comments (0)