“Falling through the cracks”; Stakeholders' views around the concept and diagnosis of mild cognitive impairment and their understanding of dementia prevention

Many people live with an awareness of mild cognitive changes that increase their dementia risk. Previous authors describe the uncertainties of this liminal state, between cognitive health and dementia, where being “at risk” can itself be an illness. We ask how services respond to people with memory concerns currently, and how a future, effective and inclusive dementia prevention intervention might be structured for people with memory concerns.


| INTRODUCTION
Dementia is the greatest health and social challenge of our time, with the number of people living with the disease projected to rise to 115 million by 2050. 1 As a sequelae of the drive across Western countries towards early dementia diagnosis, to optimise benefits from symptomatic treatments and advice, many people are being identified by health services as experiencing memory concerns and cognitive problems that put them at increased dementia risk. Mild Cognitive Impairment (MCI-objective cognitive symptoms and absence of dementia), affects a fifth of people over 65; 2 and Subjective Cognitive Decline (SCD-self-reported experience of cognitive decline and unimpaired performance on cognitive tests) half of people over 65. 3 In sociological terms, MCI and SCD are contested categories. Most people experiencing them do not develop dementia. While most people with memory concerns do not seek help from services, many do. Awareness of increased dementia risk can drive lifestyle change, 4 but also causes anxiety. 5 Dementia prevention is a political priority in many countries, fuelled by optimism about the preventative potential of lifestyle change and health promotion. 6 A third of dementia cases are attributable to potentially modifiable dementia risk factors. These include cardio-metabolic dysfunction (diabetes and cardiovascular risks), physical inactivity, social isolation, hearing loss, mental illness, alcohol and smoking. 7 There is also evidence that interventions targeting dementia risk factors reduce cognitive decline. 8 NICE (The National Institute for Health and Care Excellence) 9 and PHE (Public Health England) 10 guidelines advocate dementia prevention through planning health services that "emphasise the need for, and help people to maintain, healthy behaviours throughout life", with a focus on physical activity. 9 A recent review found limited evidence regarding the implementation of such policies. 11 No current evidence-based interventions exist that are scalable for delivery to large populations. 12 In this article, we ask how services and society respond to people with memory concerns, and how a dementia prevention intervention should be designed for people with memory concerns. Our primary aim is to explore how people with MCI or SCD and other stakeholders involved in their wellbeing and care (family members, National Health Service and third sector professionals and commissioners) consider future dementia prevention (APPLE-Tree Programme) should be designed for and delivered to this group.

| METHODS
The APPLE-Tree programme protocol is published elsewhere. 13 London (Camden and Kings Cross) Research Ethics Committee

| Sampling and recruitment
We interviewed people aged 60+ years with subjective or objective memory problems without dementia or significant functional impairment; family members in at least monthly contact with them, and health and social care professionals and third sector workers involved in commissioning or supporting this group. We purposively selected participants for diversity in gender, age, ethnicity and location (urban, semi-urban and rural). Participants who lacked capacity were excluded.
Participants were recruited from NHS memory, primary care and IAPT (Improving Access to Psychological Therapies) services; UCL and third sector organisations across London, South England and Northwest England. We recruited people who had been told they had MCI/ SCD; and those with memory concerns who had not sought NHS help, via third sector organisations.

| Procedures
We conducted semi-structured, face-to-face qualitative interviews with participants between July and September 2019. At their preference, two people with memory concerns were interviewed with their family member and two family members were interviewed together.
Other interviews were individual. One interview was conducted in Italian; the remainder were in English. Participants gave written, informed consent before interviews.
Topic guides were based on our research aim and revised iteratively to explore issues raised in interviews. We showed participants prompt cards listing potential lifestyle changes: being more physically or mentally active; looking after your physical and mental health/wellbeing and improving sleep; eating more healthily (and a description of a Mediterranean-style diet); and reducing alcohol.
Topic guides explored: how people with memory concerns are best motivated and supported to make targeted behavioural changes, potential barriers and facilitators to participating in an active dementia prevention programme and to its delivery (for professionals and third sector workers). The topic guide is appended as a supplementary file.

Key points
• We explored how services currently respond to people with memory concerns and how a future dementia prevention intervention might be structured for people with memory concerns.
• We found that people with memory concerns find themselves in a liminal, transitional state, between health and dementia, with professionals locating responsibility for managing and monitoring memory concerns with patients.
• Effective dementia prevention must empower individuals to make lifestyle changes within challenging contexts and prevention programmes must be sufficiently flexible to allow new activities to be fitted into people's lives.
Interviews lasted around 45 minutes. We ceased recruitment and determined that thematic saturation was reached when no new themes emerged. We asked people with memory concerns to complete the Quick Mild Cognitive Impairment screen (Qmci). 14

| Data analysis
We used NVivo 12 qualitative analysis software and an inductive thematic analysis approach. 15 Data collection and analysis was iterative.
Interviews were digitally recorded, anonymised and professionally transcribed. Ten transcripts were independently coded by co-authors and codes compared. MPo and HM then developed and applied a coding frame to remaining transcripts, in discussion with other authors; revising it to ensure it applied to all interviews (two further iterations). We identified and discussed emerging themes.

| Thematic analysis
We identified three main themes responding to our research question of how an active prevention intervention should be designed and delivered for people with memory concerns. These were: (1) acknowledging the liminal state, (2) enabling change in challenging contexts and (3) building on existing values, cultures, and routines.

| Theme 1: Acknowledging the liminal state
This theme reflects the "lostness" or state of "falling through the cracks" encountered by people with memory concerns who find themselves in a liminal, transitional state, between health and dementia. By contrast, professionals identified a clear dichotomy between dementia and wellness. As "well" people, those with memory concerns were considered fully able to manage their cognitive condition. In clinical encounters, NHS memory service professionals prioritised explaining risk of progression to dementia, the importance of seeking help if symptoms worsened, and advice to self-initiate lifestyle change-by "signposting" to appropriate service and groups. In some professional accounts, there was a sense that therapeutic nihilism may underlie this "signposting" away, but there was also a practicality to it, as memory services are only commissioned to treat people with dementia.
Memory concerns were conceptualised either as a risk factor, or a diagnosis that they had no mandate to treat. This approach often left people with memory concerns and their families feeling alone and unsure what to do; holding the responsibility to self-monitor and selfmanage, without support from services.
1.1 "The only plan I didn't have in place was to keep my memory" Interviewees living with memory concerns and their family members described a liminal state-between the normality of retirement, holiday plans, and doubts and fears about how memory concerns might affect them in the future. They described the uncertainty and fluidity of the experience of undiagnosed memory problems.
Because I was very busy as a businessman before I retired and I was very conscious before I retired that I had to have the plan in place to do things. The only plan I didn't have in place was to keep my memory, so there we go, that's where it is.
Person with memory concerns, not "diagnosed" by services (PWMC15).
Accounts of people with memory concerns and staff appeared to acknowledge a need to identify with wellness and move away from the illness position. This was implicit in the following account: I think you have got to be outward-looking and connect with people. If possible, with people who are in one or another way worse off than you are, so that you can help them in a little way. I think helping other people helps your mental wellbeing.
Person with diagnosis of MCI (PWMC5).
Most people with memory concerns referred to a clinical encounter where memory concerns were acknowledged and discussed, as a defining event in how they understood and responded to their memory concerns. There was a divergence between how professionals and people with memory concerns perceived these encounters, which we describe in the two subsequent sub-themes.

"People need to be given the facts"
Professionals located responsibility to reduce the risk of dementia and to contact services if symptoms worsened with the client: I think people need to be given facts. So they need to be told, you know, your chance of this progressing to full-blown dementia is 10-15% a year or whatever it is.
And we should tell them, this is not inevitable, that you will go on to have dementia. We've caught this early, you can change it by lifestyle. This is what you need to do. So you need to give them those very clear messages that the carrots, almost literally, rather than the sticks. So that they have got that incentive.
In this next quote, there is an implicit suggestion that the professional role is information provider-and it is up to the clients how they respond.
But we don't see them again. With MCI diagnoses, we're not commissioned to hold on to MCI patients […] we're doing an audit at the moment looking at how many people come back later and whether they come back later than we would have liked. Because I tend to feel like we did an MCI diagnosis three years ago.
They're now coming back and they're already moderate. And you think, why didn't you come back two years ago?  Some staff members reported providing advice directly to family members, which suggested they saw the family members as carers: The advice might be given to the carer rather than the person with memory problems.

| Theme 2: Enabling change in challenging contexts
We identified three sub-themes, describing potentially challenging contexts of which a dementia prevention programme should take account. Cognitive and physical frailties, and low self-confidence were barriers that people with memory concerns and professionals had experienced being overcome with support. This ran counter to professional accounts in the previous theme, which placed responsibility for introducing lifestyle changes solely on people with memory concerns and their families. Support worker, third sector organisation (ST16).

Increasing frailty and challenges
Memory concerns and concomitant physical frailties were challenging contexts in which to increase social engagement and activity.
Tailoring advice to the individual as part of the intervention programme was suggested by many to mitigate this: Third sector worker (ST18).

Need to build self-confidence
While physical and cognitive problems may prohibit involvement in some activities, associated loss of confidence was mostly perceived as the greatest barrier for joining a dementia prevention intervention group.
Yes, I think it's that confidence thing of feeling that they're losing part of themselves a little bit and they don't want to make a fool of themselves or they don't want to get things wrong. And they don't want to turn up on the wrong day or be late.
One person with memory concerns reflected on the value of supporting someone to overcome the barrier of joining a group, as after attending one session, it gets easier: Yes. You only walk in once, don't you? You are only a stranger once.

Role of gender
Similarly, it was also important that activities were tailored to possible perceived gender differences.
The other thing, now they've started a knitting group in our local library for any age group, but mainly women are in it and they're knitting.
Wife of person with MCI (FM3).
We have the yoga that isn't for everybody. And walking football obviously will only be for men. It isn't just for men, but obviously it would normally be men who'll be interested in it.

| DISCUSSION
We identified three themes: (1) acknowledging the liminal state, (2) enabling change in challenging contexts, and (3) 20 Perhaps the loss of fidelity to an evidence-based approach in these volunteer models finds compensation in sustainability, which may stem from their support of active citizenship. Our new coproduced APPLE-Tree intervention programme, which we developed from this work, is situated as a cognitive wellbeing course. It does not explicitly reference dementia.
Linked to this liminal position, between wellness and illness, our second theme acknowledges the challenging contexts in which a future, inclusive dementia prevention programme must operate-of social disengagement, frailty and lack of confidence. Outside the context of responsibility for service provision, professionals acknowledged that people with memory concerns required help and response to circumvent these challenges.
Our final theme acknowledges the importance of previous roles in the development and delivery of a future dementia prevention intervention. For any activity or lifestyle change to be successfully adopted and sustained, it should consider preferences, routines and habits potentially built over decades; and be realistic for the person's current abilities. Gendered expectations could limit engagement in certain activities but also create a sense of familiarity, comfort and belonging within certain groups. The importance of tailoring advice and activities to the individual in order for an intervention to be effective and acceptable for individuals with memory concerns was a clear message by our study participants. Tailoring interventions to individuals' interests and needs, especially for individuals with memory concerns, has been recommended elsewhere. 7,9,21 We recruited participants from a broad range of organisations that provide care to people living with SCD and MCI, including NHS and third sector organisations thereby maximising diversity and encompassing a wider range of professional expertise. We only included professionals who often worked with people with memory concerns. However, our sample were inevitably drawn from those who self-identify with memory concerns, and mostly from those who explicitly sought help for such services.

| CONCLUSION
Effective dementia prevention must empower individuals to make lifestyle changes within challenging contexts. Programmes must be evidence-based yet sufficiently flexible to allow new activities to be fitted into people's current lives. They must also be mindful of the risks of pathologising memory concerns. Most current memory services are neither commissioned, financially or clinically resourced to support people living with memory concerns without dementia. Effective, large scale dementia prevention will require a broad societal response.

SUPPORTING INFORMATION
Additional supporting information may be found online in the Supporting Information section at the end of this article.