In adults with a primary diagnosis of dementia, what is the evidence for the effectiveness of multimodal (i.e. complex, more than one mode) non-pharmacological interventions for improving cognitive functioning?
• What is the evidence for what works and does not work?
• What is the evidence for different groups of people with dementia (early, mid or late stage) and potentially for different types (Alzheimer’s, vascular, etc.)?
• What are the strengths and limitations of different study designs used in testing these outcomes?
• What are the strengths and limitations of different evaluation tools used to assess the effectiveness of non-pharmacological multimodal interventions (NPMMIs) on improving cognitive functioning?
• What outcome measures have been used?
• What is the evidence for theory on the likely process of change for each mode of intervention?
• What is the effectiveness of different modes of delivery on the effectiveness of NPMMIs for improving cognitive functioning for people with a diagnosis of dementia?
Major healthcare Databases - MEDLINE, EMBASE, The Cochrane Central Register of Controlled Trials (CENTRAL), PsycINFO, AMED, CINAHL, ASSIA, LILACS, Science Citation Index, Scopus and the Cochrane Library.
ALOIS (www.medicine.ox.ac.uk/alois) — the Cochrane Dementia and Cognitive Improvement Group’s (CDCIG) specialised register.
Trial registers – PROSPERO, ISRCTN; the WHO portal (which covers ClinicalTrials.gov) and ISRCTN.
A hand search of Dementia, The Gerontologist and conference paper abstracts in the Journal of the American Geriatrics Society, as they may not be found in the databases.
Other resources - As this is an emerging area for research, conference papers, websites and news articles are expected to be an important source of information. Therefore, search of the published literature will be complemented by a search of the grey literature including ISI Web of Knowledge Conference Proceedings; by contacting experts and activists on this topic; by checking websites and specialist researcher websites such as ResearchGate; published conference abstracts; reference lists of included papers; Index to Theses; and Citations searches - following up articles which cite any of the included review papers.
Types of study to be included
We will conduct a mixed methods systematic review in which all study designs will be considered.
Condition or domain being studied
This review is designed to synthesise available evidence about the effectiveness of non-pharmacological multimodal interventions (NPMMIs) on the cognition of adults with dementia.
Adults (people over 18) with a diagnosis of dementia as their primary condition (not as part of the end stage of another disease like Parkinson’s for example); living in their own home or in a care home or nursing home.
The intervention must be complex, meaning that it has at least two components or modes. No mode can be pharmacological. If however it is a controlled trial and one arm is pharmacological and the other is non-pharmacological, then the non-pharmacological arm will be included for review if it has at least 2 modes.
Interventions, activities or components to include – Exercise, physical exercise, aerobic exercise, strength training, dance, participatory arts, music, movement, walking, yoga, meditation, mindfulness, diet, Mediterranean diet, food, nutrition, vitamins, herbs, medicinal plants, supplements, medium-chain triglycerides (MCTs), brain training, brain gym, cognitive stimulation/ rehabilitation, transcranical magnetic stimulation (TMS), outdoors, gardening, green gym, Traditional Chinese medicine, tai-chi, mah-jong, therapy, psychotherapy, hypnotherapy, horticulture and lifestyle change
The intervention can take place in any location including the person’s own home, medical clinic, GP surgery, health club/gym/pool, park or green space, activity room, school, village hall, community centre or care environment (such as day centre, care home or nursing home). Activity provision can be administered to a group of people or to individuals by health or social care professionals, providers or consultants, volunteers or members of a research team. It could potentially be self-directed lifestyle changes, whereby individuals carry out their own activities and report on their progress to the research team or healthcare provider who monitors the participants over the course of the study. All options will be included.
Studies in peer-reviewed journals, conference proceedings or theses; monographs, reports and articles in newspapers, magazines, print and other media will be included.
Languages: Articles, titles and abstracts that can be accessed in the English language.
Time and Place: Publications of any date.
Intervention: Studies that describe interventions with more than one (>1) component or domain, such as exercise, diet, cognitive stimulation, stress reduction, sleep hygiene, hypnotherapy, etc.; Studies in which the intervention is led, supervised or prescribed and also monitored (not simply advised or recommended); and studies involving participant self-report in which their activities and experiences are recollected.
Study participants: Data will be included from adults with a primary diagnosis of dementia.
Primary Outcome: Included studies will address the effectiveness of NPMMIs on cognitive function, (regardless of whether the outcome was found to be positive, negative or neutral). Studies will be included whether they address cognitive effect exclusively, or address it as one of a range of outcome measures.
Study designs and settings: All study designs and all settings. Studies may also include measures of self-report of cognitive function.
• Interventions being of only one mode - for example 3 different kinds of exercise would be excluded, (whereas an exercise activity with a dietary change and/or cognitive stimulation would be included).
• Studies involving non-human subjects
• Studies that utilise tracking, tagging or physical restraints
• Studies that specifically address delirium, pain, incontinence or Parkinson’s disease
• Studies in which the intervention delivers care staff education or training; studies specifically designed to improve the knowledge, abilities or wellbeing of carers, or to facilitate the caregiving process.
• Studies in which the interventions are specifically designed to improve the management or control of people with behavioural and psychological symptoms of dementia (BPSD) in a care setting, and are therefore not intentionally designed to improve cognitive function of the person with dementia, but rather to improve the knowledge and ability of care staff to cope and manage.
We will include controlled trials in which the effects of interventions are compared with controls which are not expected to have specific cognitively modifying effects. The control arms would typically involve placebo/social interaction or no intervention/usual care.
The number of people living with dementia worldwide in 2017 was estimated at 47 million and is projected to increase to 75 million in 2030, and almost triple to 132 million by 2050 (WHO, 2017). In the UK, 1 million people will be living with a dementia by 2025 and delaying the onset by five years would reduce deaths directly attributable to dementia by 30,000 a year with the financial cost of dementia to the UK is £26 billion per annum (Alz Soc, 2014). Dementia has now overtaken heart disease as the number one cause of death in the UK, with the socioeconomic cost now being higher than the cost of cancer and cardiovascular disease combined.
Prevalence, prediction and costs as well as failed drug trials and limited efficacy of existing drugs (Vandenberghe et al. 2016, Alz Soc 2016) have led to an urgent need to look for alternative solutions that focus on non-pharmacological approaches; this includes holistic approaches to dementia prevention and treatment involving lifestyle changes. Also, a newer view of Alzheimer’s is emerging, one in which the disease is determined by multiple factors and mechanisms, and not simply that caused by neuropathology. Alongside this we have seen the emergence of multimodal approaches to prevention (Barcelos et al. 2015; Barnes et al. 2013; Burgener et al. 2011; Cheng et al. 2014) including holistic and psychosocial ones (Van der Linden & Van der Linden, 2016; Whitehouse, 2013a).
A number of single studies undertaken across the globe point to the potential for these approaches to maintain, improve or even reverse the cognitive decline associated with dementias (Burgener, 2011; Sachdeva et al. 2015; Rodakowski et al. 2015). For example, the FINGER (Finnish Geriatric Intervention) Study to Prevent Cognitive Impairment and Disability (the first large randomised controlled multi-domain lifestyle intervention) showed that cognitive decline can be prevented by addressing several risk factors simultaneously among older at-risk elderly people (Ngandu et al. 2015). To date, however, we have little understanding of the size and quality of the evidence base for the effectiveness of multimodal non-pharmacological approaches. This study aims to address this gap through a systematic review of the mixed method and grey literature. The review approach is ‘integrative’ in which the contents of multiple studies is summarised and interpretation is minimised. Such an approach allows for “the simultaneous inclusion of experimental and non experimental research in order to more fully understand a phenomena of concern” (Whittemore & Knafl, 2005).
The outcome measure will be cognitive function, to include any effect on attention, episodic or semantic memory, perception, speech & language, verbal fluency, decision-making, speed of processing, executive functioning or executive control.
Effects will be assessed by any validated measure of cognition or cognitive function, for example, Alzheimer's Disease Assessment Scale-cognitive subscale (ADAS-cog) or The Mini Mental State Examination (MMSE). Effects can also be assessed by clinician report or self-report. No minimum treatment duration nor follow up is set.
Data extraction, (selection and coding)
Search terms and all their variations will be incorporated into a search strategy tailored to each database, drawing on specialist librarian support. A search log will be kept to record the databases and websites searched, key words used and results of the search. The titles and abstracts of articles to be considered for retrieval will be downloaded into reference management software, with a note of where they were found. If more than one report describes the same trial, we will include all of them to allow complete extraction of the trial details. All titles and abstracts of studies identified by the researcher through electronic searches will be initially reviewed by the researcher to extract duplicates and those clearly not meeting the inclusion criteria. Abstracts will then be reviewed by the whole team and the inclusion and exclusion criteria will be applied to judge relevance to the study questions. The researcher will retrieve and scan full text of the agreed abstracts and further extract those which on closer scrutiny do not meet the criteria. These decisions will again be agreed by the research team. Any selection queries arising from this process will be evaluated and resolved by the whole team. Papers that cannot be retrieved will be listed in a database.
Data relating to study design, participants, interventions and outcomes will be extracted from included studies in a standardised way into structured tables by the researcher and checked for accuracy by the research team. Studies will be grouped by intervention, target population, outcomes and context. We will also follow the PRISMA guidelines (Liberati et al. 2009) for reporting reviews of studies evaluating healthcare interventions, with the checklist of items to include.
Risk of bias (quality) assessment
Once these disparate data have been extracted, the quality of the individual studies will be assessed. We will draw upon the quality appraisal tool of Hawker, et al. (2002) to assess the qualitative studies. Due to the design of the tool it is also appropriate for critical appraisal of conference posters & papers and articles in print media. For quantitative intervention studies we will use the Jadad scale (Jadad et al. 1996) which is a procedure to independently assess the methodological quality of a clinical trial. Any disagreements on the risk of bias/quality assessment will be resolved by the research team.
Strategy for data synthesis
Results will be synthesised narratively, as the studies in the review are expected to be heterogeneous in their research design. We will be guided by Popay, et al. (2006), which presents ‘narrative synthesis’ as a textual approach that “tells the story of the findings of included studies.” Standard procedures will be followed, including the development of a theory of how the interventions work and for whom, and a preliminary synthesis of studies. Textual descriptions will also be produced at this stage. Established techniques within narrative synthesis will then be used to explore relationships within and between studies and assess the robustness of the synthesis.
Further guidance by Hong et al, 2017 on the conducting and reporting of mixed studies reviews will be incorporated into the strategy regarding synthesis methods and the integration of qualitative and quantitative data. The emerging analysis and main themes will be reviewed regularly with the research team.
Analysis of subgroups or subsets
If there is evidence for different groups of people with dementia (early, mid or late stage) or for different types of dementia (Alzheimer’s, vascular, etc.), then subgroup analysis will be possible.
• Submission to a peer-reviewed academic journal
• Findings presented to the funder by the research team in 2018
• Findings presented at an international dementia conference
• Dissemination will also be undertaken through social media (LU website, Facebook & Twitter).
Contact details for further information
Centre for Ageing Research (C4AR)
Division of Health Research
Faculty of Health and Medicine
Lancaster LA1 4YG UK
Organisational affiliation of the review
Centre for Ageing Research, Division of Health Research, Faculty of Health and Medicine, Lancaster University
Dr Garuth Chalfont, Division of Health Research, Faculty of Health and Medicine, Lancaster University Professor Christine Milligan, Division of Health Research, Faculty of Health and Medicine, Lancaster University Dr Jane Simpson, Division of Health Research, Faculty of Health and Medicine, Lancaster University
Anticipated or actual start date
01 May 2017
Anticipated completion date
28 February 2018
Lancaster University and the AIM Foundation
Conflicts of interest
Subject index terms status
Subject indexing assigned by CRD
Subject index terms
Cognition; Dementia; Humans; Multimodal Imaging
Stage of review
Date of registration in PROSPERO
30 June 2017
Date of publication of this revision
30 June 2017
Stage of review at time of this submission
Piloting of the study selection process
Formal screening of search results against eligibility criteria
Risk of bias (quality) assessment
PROSPERO This information has been provided by the named contact for this review. CRD has accepted this information in good faith and registered the review in PROSPERO. CRD bears no responsibility or liability for the content of this registration record, any associated files or external websites.