Dementia wellbeing and COVID‐19: Review and expert consensus on current research and knowledge gaps

Abstract Objectives In response to a commissioned research update on dementia during the COVID‐19 pandemic, a UK‐based working group, comprising dementia researchers from a range of fields and disciplines, aimed to describe the impact of the pandemic on dementia wellbeing and identify priorities for future research. Methods We supplemented a rapid literature search (including unpublished, non‐peer reviewed and ongoing studies/reports) on dementia wellbeing in the context of COVID‐19 with expert group members' consensus about future research needs. From this we generated potential research questions the group judged to be relevant that were not covered by the existing literature. Results Themes emerged from 141 studies within the six domains of the NHS England COVID‐19 Dementia Wellbeing Pathway: Preventing Well, Diagnosing Well, Treating Well, Supporting Well, Living Well and Dying Well. We describe current research findings and knowledge gaps relating to the impact on people affected by dementia (individuals with a diagnosis, their carers and social contacts, health and social care practitioners and volunteers), services, research activities and organisations. Broad themes included the potential benefits and risks of new models of working including remote healthcare, the need for population‐representative longitudinal studies to monitor longer‐term impacts, and the importance of reporting dementia‐related findings within broader health and care studies. Conclusions The COVID‐19 pandemic has had a disproportionately negative impact on people affected by dementia. Researchers and funding organisations have responded rapidly to try to understand the impacts. Future research should highlight and resolve outstanding questions to develop evidence‐based measures to improve the quality of life of people affected by dementia.


| INTRODUCTION
The COVID-19 coronavirus pandemic disproportionately affects people living with dementia who have a substantially increased risk of infection and subsequent death, 1,2 accounting for 31% of all COVID-19 related deaths. 3 People living with dementia are also at risk of and affected by the social isolation and reduced access to health and social care that accompany pandemic restrictions. In the UK, people affected by dementia (individuals with dementia, their carers and social contacts, health and social care practitioners and volunteers), were 'worst hit' by the pandemic. 4, 5 Dementia services, research activities and funding have been reduced, while health inequalities have widened. 6 Initiatives to investigate and address these impacts have begun to identify learning from pandemic experiences, but research that includes people living with dementia has been affected because of their vulnerability to  For those working to support people living with dementia and carers, it is important to understand findings from existing research on COVID-19 and dementia, and how to make the best use of that research to inform policy and practice and identify key research gaps to direct future work. In response to the impact of the pandemic on dementia wellbeing, NHS England published a guide to delivering policy aspirations in the context of COVID-19, based on the NHS England Dementia Well Pathway for delivering best dementia care 7 .
This pathway provides a framework of care and support that people with dementia, and their carers, need at each stage of their journey, from prevention to end-of-life, 8 so that dementia care, research and awareness can be optimised. 9 The guide to dementia wellbeing in the COVID-19 pandemic 7  social care who work on dementia-related topics (see Table S1 for further details of the working group). Initial discussions were supplemented by a literature search to identify published and unpublished research on COVID-19 and dementia at the end of 2020. We identified areas of consensus on the important gaps in the literature and future directions for research. We report findings aligned with the six steps of the COVID- 19 Dementia Wellbeing Pathway and specifically mention the UK studies.

| Literature search and study selection
One author (KL) led the literature search, reviewed titles and abstracts for relevance and then extracted data from relevant full texts.
Databases were searched using specific search terms (Table 1) on 9th December 2020, which was updated on 7th January 2021 in consultation with members of the expert group.
Included studies were research articles in the English language from any country on COVID-19 and dementia or mild cognitive impairment in individuals of any age. Individuals with mild cognitive impairment are often assessed and/or followed up by memory services, and a significant proportion are experiencing the early stages of a neurodegenerative illness and will progress to dementia. We included primary studies and reviews, as well as case series, commentaries, position papers and policy documents if they reported quantitative or qualitative findings in relation to COVID-19 and dementia wellbeing. We excluded studies that investigated risk for development of dementia secondary to COVID-19, the mechanisms of disease, or the impact of dementia on COVID-19 mortality or infection, without identifying specific risk factors within dementia.
Case reports were excluded, as studies of individuals were judged to be insufficiently informative for the purpose of this study, as were articles based purely on data that were not obtained during the COVID-19 pandemic period (before November 2019).
Any additional studies or reports that the expert group judged to be consistent with the aim of identifying the scope of existing research on COVID-19 and dementia wellbeing, but that had not emerged from the literature search, were combined with the search results and reported ( Figure 1). For example, some specific studies on people living or working in long-term care facilities, such as care homes, did not explicitly report that participants with dementia were included in the study, but were judged to be relevant as more than 70% of UK care home residents are estimated to have dementia. 14,15 Some expert members were also aware of ongoing funded studies that had adjusted their existing project to explore the impact of COVID-19, non-peer reviewed UK survey-based reports from charities, 4,5,16,17 as well as two reports published by the 'LTCcovid' collaboration (www.ltccovid.org), which was set up to gather resources to support long term care responses to COVID-19. 3,18 Some members were aware of funded studies that have yet to be announced; we have not included them in the tables but provide some details in the main text.

| RESULTS
The database searches and screening of articles identified 141 relevant studies (see Figure 1, 2 and Tables 2-7). We describe the main findings from these studies and key gaps in research identified by expert consensus in relation to each of the six steps of the COVID- 19 Dementia Wellbeing Pathway, summarised in Box 1.
Preventing Well: Keeping fit and active to prevent the mental and physical consequences of isolation ( Table 2).
The overall negative impact of COVID-19-related restrictions on people living with dementia and their carers' wellbeing, mental health and functioning was explored by over 30 publications, which included participants with a range of different dementias. 129 A number of studies included in the Supporting Well and Living Well domains, described below, have the potential to help prevent the negative impact of further waves of the pandemic on dementia wellbeing.
Most of the dementia-focused studies used cross-sectional surveys and relied on subjective retrospective ratings, and few compared findings with pre-pandemic outcome data, revealing a need for further quantitative and longitudinal research. and individuals and families were identified by the expert group and are listed in Table 2 and described in more detail in Table S2.

T A B L E 1 Databases and search terms used for the literature search
Another theme, related to the prevention of COVID-19 in dementia, was the reduced ability of many people living with dementia to understand and comply with pandemic restrictions, social distancing and the use of face masks, 50 has the potential to worsen functioning and cognitive impairment. 138 The potential impact of post-COVID syndrome on the dementia workforce and carers is also unknown. Members of the expert group were aware of NIHR and UKRI-funded studies on the longer term biological and health impacts of COVID-19, that is, post-COVID syndrome, to be announced in early 2021, 139  Teleconsultation used in memory clinics during lockdown 80,81,87 was helpful for some carers, but others were uncertain about their purpose and usefulness. 5 An adapted UK study (PriDem) found that many people living with dementia and their family carers were avoiding contact with primary health services and had mixed opinions of the value of telephone contact from primary care. 88 Worsening of dementia-related behaviours seen during the pandemic, such as anxiety, agitation and psychosis, was associated with an increase in antipsychotic prescribing. 95 Table S2.
Lastly, several dementia-related resources and projects have been set up or adapted to support the national responses to COVID-19 in the UK, 117 Canada, 121 and internationally 3,10 ; including middleincome countries. 120 Living Well: Optimising the lived experience of dementia (Table 6).
Published and ongoing studies on the potential of technology to deliver physical, psychoeducational and psychosocial interventions remotely to people with dementia; 122  � Quantitative and longitudinal studies on the impact of COVID-19 and isolation. Studies should be inclusive and assess the impacts of inequalities. � Impact of restrictions on the wellbeing of care home residents with dementia, care home staff and health and social care professionals who work alongside people with dementia. � Findings specific to people working with or affected by dementia within broader studies, and dementia-specific findings within care homes with residents who have/do not have dementia should be analysed. � Monitoring effects of vaccination in people with dementia, as they were not included in published vaccine trials. � Longitudinal data on the impact of COVID-19 on dementia risk.
Diagnosing well: � Reduced access to dementia diagnostic services and reduced diagnosis rates during the pandemic. � COVID-19 in dementia is often asymptomatic or atypical and may include hypoactive delirium, making accurate PCR testing critical to control spread. � At the start of the pandemic in some areas, people with dementia in care homes and psychiatric/mental health hospitals were only able to access COVID-19 testing at a later stage compared to others. This likely contributed to local outbreaks. � Remote assessment and virtual diagnosis of dementia in home-dwelling participants during the pandemic may be feasible.
� Impact of loss of face-to-face assessments and access to diagnostic services during the pandemic on dementia diagnoses and mental health diagnoses in people with dementia. � Replication of studies conducted during the start of the pandemic at later periods, e.g., subsequent lockdowns, would establish whether lessons were learned and if they (and/or widespread antibody seropositivity) had any impact on infection and mortality rates. � Optimise remote assessments and ensure those who cannot use or possess the technology are not left behind. � Studies on post-COVID syndrome ('Long COVID') to include people with dementia to investigate the prevalence and impact of this condition on dementia risk, diagnosis and progression. The potential impact of post-COVID syndrome on carers and the dementia workforce. � Longitudinal biomarker studies (imaging, fluid, cognitive) to investigate rates of dementia progression following COVID-19.
Treating well: � Reduced access to and provision of primary care and memory clinic services, alongside an increase in antipsychotic prescribing. � Responses to telehealth were mixed. � People with dementia who are hospitalised have specialised needs, but reduced access to health care, which may not be justified. � Clinical research activities, such as participant recruitment, have been affected.
� Longer-term impact of reduced access to primary and secondary health services. � The acceptability and effectiveness of telehealth for people living with dementia. � Factors associated with positive treatment outcomes in people with dementia to inform inappropriate restriction to healthcare and use of DNAR orders. � Impact of COVID-19 on dementia research, including on existing and future projects, early career researchers, and participant recruitment initiatives.
Supporting well: � The pandemic has reduced care home residents' and staff wellbeing. Telehealth may benefit some residents. � Informal or paid home care, day care and other services can mitigate carer stress, but many carers worried about their availability and infection risk and did not access services. Many carers were unable to access remote consultations or support. Financial assistance may be beneficial. � Reduced access to and use of social care services, e.g. respite care. � Resources and projects set up or adapted to support national responses to COVID-19.
� Best practice in care homes, relating to visiting, remote communication, infection control whilst reducing loneliness and protecting human rights. � How to best support dementia care at home, including social, digital, financial and other interventions to reduce carer stress and burden. � Social care research findings related to older people in particular may be extrapolated to dementia, and specific dementia-related findings should be analysed and reported within broader studies that include these groups.  (Table 7).
We found one published review 127  The decision-making process in older adults may need to happen more rapidly in the context of COVID-19, 128 and two ongoing UK studies are developing advance care planning and decision-making tools for people affected by dementia in care homes with nursing (Brazil et al., UKRI, Table 7) and family carers. 144 UKRI, Table 7). The latter study has produced a freely available evidence-based decision tool to help people with dementia and their family carers make difficult decisions. 144 Such decisions and consultations are likely to be hampered by the loss of visitor access to many hospitals and care homes during the pandemic, but quantitative research in this area is lacking.
Additional studies as well as the regulator's investigations 126 on the use of advance decisions to refuse treatments and misuse of DNAR orders may help improve advance care planning for people living with dementia.

| DISCUSSION
Based on the findings that emerged from this initiative, there were directions for future research within each domain that the consensus group considered important (Box 1). Some themes applied across all domains, such as the potential benefits and risks of remote healthcare to deliver interventions and support, the need for longitudinal studies to monitor the longer-term impacts, the risk and impact of widening health and other inequalities, the need for sociodemographically inclusive studies, and the benefit of reporting dementia-specific findings within broader studies.
As this is a fast-moving and rapidly changing area in terms of policies and adaptations, which may outpace the rate of the normal publication process, we have tried to swiftly incorporate research findings from a range of non-peer reviewed sources, including registered systematic reviews in progress, preprint databases and COVID-19-specific websites and reports. This paper presents findings from a literature review and expert consensus, intended to provide comprehensive coverage of several aspects of dementia research related to COVID-19, and was not driven by a specific research question. Our study was not designed to be a traditional systematic or scoping review 145 and we did not formally quantify the quality of studies. Therefore, our findings may be less comprehensive compared to a protocol-driven search strategy that follows published guidelines and/or systematic critical appraisal.
Members of the consensus group are involved in online projects that can rapidly update evidence relating to COVID-19, for example, the LTCcovid collaboration (www.ltccovid.org). While peer-review remains the gold standard for research publication, databases such as these as well as preprint repositories can provide valuable, timecritical resources during the pandemic.
Another limitation of our study was that some potentially relevant studies of long-term care facilities may have been omitted if they did not explicitly mention inclusion of people with dementia: members of the consensus group identified some such studies, and these are described in the text. Non-specific findings should be interpreted cautiously, including those from large population-based studies/surveys of individuals, families and the health and social care workforce. For example, only 22% of respondents in the Carers UK survey (Table 2)  � Remote delivery of psychosocial, education and physical interventions have the potential to benefit wellbeing, but the caveats of access to and capacity to use technology also apply.
� Ongoing and further research to assess the success of remote interventions to improve wellbeing. � Rehabilitation needs of people affected by dementia, e.g., due to post-COVID syndrome or prolonged isolation.
Dying well: � Difficult decisions may need to be taken more rapidly in COVID-19, and decision aids for people with dementia and their families are now available.
� Primary research studies on palliative care involving people with dementia. � Studies to help improve advance care planning. the NHS England Dementia Well Pathway, but studies were not conducted with this framework in mind. Thus, research findings may be relevant to more than one domain and the categorisation of studies, by necessity, may sometimes have been imprecise. As this study was prompted by a question from the English Dementia Programme Board on research on dementia and COVID-19, and our consensus group were English-based experts, our findings and consensus views were skewed towards research based in England.
Overall, the COVID-19 pandemic has had a disproportionately negative impact on dementia wellbeing, and researchers and funding organisations have responded rapidly to try to understand the impacts.
We have identified potential directions for future research to explore these further, so that evidence-based measures can be developed to improve the quality of life of people affected by dementia.