Prognosis of Alzheimer’s disease; what do patients and care partners want to know?

Background: Prognostic studies in the context of Alzheimer’s disease (AD) mainly predicted time to dementia. However, it is questionable whether onset of dementia is the most relevant outcome along the AD disease trajectory from the perspective of patients and their care partners. Therefore, we aimed to identify the most relevant outcomes from the viewpoint of patients and care partners. Methods: We used a two-step, mixed-methods approach. As a rst step we conducted four focus groups in the Netherlands to elicit a comprehensive list of outcomes considered important by patients (n=12) and care partners (n=14) in the prognosis of AD. The focus groups resulted in a list of 59 items, divided into ve categories. Next, in an online European survey, we asked participants (n=232; 99 patients, 133 care partners) to rate the importance of all 59 items (5-point Likert scale). As participants were likely to rate a large number of outcomes as ‘important’ (4) or ‘very important’ (5), we subsequently asked them to select the three items they considered most important. Results: The top-10 lists of items most frequently mentioned as ‘most important’ by patients and care partners were merged into one core outcome list, comprising 13 items. Both patients and care partners selected outcomes from the category ‘cognition’ most often, followed by items in the categories ‘functioning and dependency’ and ‘physical health’. No items from the category ‘behavior and neuropsychiatry’ and ‘social environment’ ended up in our core list of relevant outcomes. Conclusions: We identied a core list of outcomes relevant to patients and care partner, and found that prognostic information related to cognitive decline, dependency, and physical health are considered most relevant by both patients and their care partners.


Introduction
Alzheimer's disease (AD) is a progressive neurodegenerative disorder, characterized by cognitive decline and dementia. 1 Dementia is in fact a late stage of this disease, that takes decades to develop. Predementia stages of AD include preclinical AD, which may express as subjective cognitive decline (SCD) and mild cognitive impairment (MCI). 1 Advances in biomarker-based diagnostic testing allow for an AD diagnosis in pre-dementia stages. 2 However, the question asked immediately after "doctor, what is wrong with me?", is "what can I expect?". 3 The establishment of an individual prognosis is still challenging due to the high variability in disease course and phenotypic manifestation. 1 In addition, it is not fully known which information patients and care partners would value most about their course of the disease.
Until now, prognostic studies mainly attempted to predict progression to dementia. While the syndrome diagnosis of dementia is an important endpoint in research, it may be rather an arbitrary point along the entire disease trajectory from the perspective of patients and care partners. 4,5 Which endpoints are most relevant to patient has not yet been thoroughly assessed. In addition, factors that are important for patients to know about the future may differ from those of care partners. As care partners have an signi cant role when making (dementia) care-related decisions, it is important to identify both patientand care partner-relevant outcomes in the prognosis of AD. 6,7 In addition, it can be debated whether the data we collect at memory clinics during routine follow-up, cover the prognostic information most relevant to patients and care partners. There is increasing interest in so called patient-reported outcomes (PROs). 8 These include aspects such as (health related) quality of life, activities of daily living and interpersonal functioning from the perspective of patients and care partners, which may be missed during our current, routine check-ups. Some existing outcome instruments, such as TOPICS-MDS, are already based on patient report. 9,10 These outcome sets however have mostly been developed without consulting patients and caregivers about what is relevant to them. A few former studies that took into account patients' perspectives mostly focused on speci c disease outcomes or on experiences with speci c treatments. 11,12 The few available studies which focused on relevant outcomes from the perspective of patients were based on interviews in small samples. 13,14 In the current study, taking the viewpoints of patients and care partners as the starting point, we aimed to identify outcomes in the disease course of AD relevant to patients and care partners from multiple European countries.

Design
In this mixed-method, multi-center study, we used a two-step approach with four focus groups followed by an online survey that was distributed in multiple European countries. Written informed consent was obtained from participants of the focus groups and digital informed consent was obtained from participants of the online survey. Local boards of the Medical Ethics Committee reviewed and approved this study.

Focus groups (step 1)
We took on a qualitative, bottom-up approach, using focus groups to elicit a comprehensive list of patient and care partner relevant outcomes in the progression of Alzheimer's disease to serve as input for the online survey (step 2).

Participants
We conducted four focus groups. Three focus groups were conducted in an academic hospital (Alzheimer center Amsterdam, Amsterdam UMC) and one focus group in a local hospital (Alrijne Hospital Leiden) between July 2019 and October 2019. Participants were recruited from the memory clinics in both hospitals. We purposefully selected participants to compose heterogeneous groups in terms of gender, age, education and disease stage (AD dementia, MCI, or SCD). Patients were eligible if they were diagnosed with AD dementia, MCI, or SCD at the memory clinic. Patients were not eligible if (1) their MMSE score was less than 18, (2) they had a psychiatric disorder, (3) they reported hallucinations or (4) they did not speak or understand Dutch. Care partners were eligible if they were a relative or loved one involved in care for someone with a diagnosis AD dementia, MCI, or SCD. Care partners were not eligible if they were younger than 18 years old or if they did not speak or understand Dutch.

Procedures
Each focus group consisted of two parts. In the rst part, we used a photo elicitation method, aimed at stimulating participants to think about and express what is most important to them in life. We showed the participants thirteen pictures related to everyday situations and activities. We asked the participants to select one or more pictures that re ected what is important to them in life. Then, we asked them to explain why they selected this picture. The second part of the focus groups was aimed at obtaining a comprehensive list of concrete outcomes. We therefore asked the participants to formulate as many answers as possible to two main questions: 1. What do you want to know about the course of the symptoms? 2. If there was a treatment for Alzheimer's disease, on which speci c aspect should this have an effect?
For each question, participants were rst given ve minutes to write their answers on individual sticky notes. Then, one by one participants were asked to read out loud and explain one of the outcomes they had written down. We collected and placed the sticky notes on large paper sheets, broadly clustering them into categories. In subsequent rounds, the participants expanded these lists with outcomes that had not yet been mentioned. After all outcomes had been read out loud and broadly categorized, participants were asked if they could think of any additional outcomes of relevance to them, which were then added.
Audio recordings were made of all focus groups, and the sticky notes on the paper sheets were also saved.

Analysis
One author (AM) transcribed the audiotaped focus groups using intelligent verbatim transcription. A second author (EDB) checked the transcripts for completeness. The transcripts were analyzed in MAXQDA software 15 using a process of inductive and deductive thematic analysis (directed content analysis). One author (AM) generated a 'start list' of codes based on literature, prior to coding. Subsequently, two authors (AM and LNCV) coded the transcripts independently, adding, adjusting and categorizing the codes based on the transcripts and the answers from the sticky notes. Afterwards, they compared and discussed the differences in coding until consensus was reached. After the fourth focus groups, data saturation was achieved, i.e., no new outcomes were identi ed based on the data from the fourth focus group compared to the previous three focus groups.
The nal list consisted of 60 outcomes, which were subdivided into ve categories. The item 'How long am I able to decide whether I want euthanasia?' is only relevant in the Netherlands, because euthanasia is not legally permitted in the other European countries, and the nal list used in the European survey thus included 59 items.

Online survey (step 2)
Subsequently, we employed a quantitative, online European survey to determine the most relevant items among the outcomes identi ed in the focus groups.

Participants
Participants were recruited via multiple routes. We recruited patients and care partners through memory clinics in the Netherlands, Slovenia and Slovakia, who were then sent a link to the online survey. In addition, the online survey was sent to members (patients and care partners) of the dementia association Demensförbundet in Sweden. Finally, Alzheimer Europe sent a link to the online survey to members (patients and care partners) of its European Working group of People with Dementia. Alzheimer Europe also invited three of its member associations (i.e. national Alzheimer associations) to share this link with members of other similar English-speaking national working groups of people with dementia and carers.
Patients were eligible to participate if they had a diagnosis of dementia, MCI or SCD. Care partners were eligible if they were a relative or loved one involved in caring for someone with a diagnosis of dementia, MCI or SCD. Care partners were not eligible if they were younger than 18 years old.

Design
The survey was created in the online survey tool Survalyzer. 16 We translated the survey into ve different languages (Dutch, English, Slovak, Slovenian and Swedish) and created separate versions for patients and care partners. Both versions consisted of two parts. In part 1, we collected background information about the participants, including age, gender, ethnicity, education, disease stage and year of diagnosis. In part 2, we rst asked participants to rate their importance of all 59 items (1-unimporant, 2-not very important, 3-neutral, 4-important, 5-very important). Participants had the option to add missing items. As we anticipated that participants would rank most items as (very) important, we subsequently provided them with the items that they rated as 'important' or 'very important' and asked them to select the three most important items. This stepped approach facilitated the identi cation of the most relevant items out of the long list of 59 items that resulted from the focus groups.

Analysis
Data was analyzed using version 22.0 of SPSS for Windows. Frequencies and percentages were calculated, resulting in two lists of the 10 items most frequently rated as most relevant by patients and care partners respectively. We assessed concordance of these two lists between patients and care partners and subsequently compiled a nal list of important items merging the top-10 lists of both patients and care partners.

Focus groups
In total, 12 patients and 14 care partners participated in the focus groups. Seven patients and care partners participated as a couple in the focus groups. Median age (IQR) of the patients was 66 (61-72) years and n = 6 (50%) were female. There were n = 5 (41%) patients with AD dementia, n = 3 (25%) with MCI and n = 4 (33%) with SCD. Mean ± SD time since diagnosis was 3 ± 1 years. Care partners had a median (IQR) age of 69 (54-74) years and n = 9 (64%) were female. N = 11 (79%) were care partner of a patient with AD dementia and n = 3 (21%) of a person with MCI. Mean time since diagnosis was 3 ± 1 years. Most care partners were a life partner (spouse) of the patient (n = 9; 64%) and was living with the patient (n = 9; 64%).
The focus groups revealed 59 outcomes of relevance, which we subdivided into ve broad categories: 1) Cognition (12 items); 2) Functioning and dependency (18 items); 3) Behavior and neuropsychiatry (14 items); 4) Social environment (8 items); 5) Physical health (7 items). The appendix provides an overview of the list of 59 outcomes, ordered by ve major categories.

1) Cognition
This category included outcomes such as learning, recognizing loved ones, planning and organizing. The following quote is a concrete example of an AD dementia patient who wants to know how the memory symptoms will progress. Quote 1: "How quickly does my memory deteriorate?"

2) Functioning and dependency
Patients with AD dementia gradually lose the ability to make important decisions and eventually become legally incapacitated. The following quote of an AD dementia care partner illustrates how patients and care partners expressed an interest in predicting the patient's independency and ability to make decisions. Quote 2: "My mother has always been an independent woman who does not accept help from others. So the rst thing that came to my mind was: when is she no longer able to make decisions on her own?" Closely related to dependency, patients and care partners mentioned outcomes related to everyday functioning and activities. The following quote by an individual with SCD illustrates the importance of the ability of performing hobbies, and other enjoyable activities.
Quote 3: "Singing, gardening, attending workshops. At this moment, I am able to do my hobbies. If I am no longer able to do my hobbies, than it is a great loss. I think it is important that I am able to do the things that I enjoy."

3) Behavior and neuropsychiatry
This category comprises outcomes such as anxiety and depressive symptoms, living in accordance with personal values and beliefs and behavioral changes. Many participants pointed out the importance of maintaining, and being able to predict changes in, their personality and identity, as illustrated by, the quote below by an individual with SCD. Quote 4: "My personality is important to me. I am always happy, that is what I am known for. I would like to know; when am I no longer myself?"

4) Social environment
The category social environment includes outcomes such as patient and care partner quality of life and having meaningful moments with loved ones. The following quote of an AD dementia care partner shows the nancial importance of knowing how long you are able to do activities together.
Quote 5: "We consciously decided to make memories and do activities together as long as we are able to. And sometimes that costs money. If you know how long you can do meaningful activities together, then you can take that into account nancially."

5) Physical health
This category comprises outcomes such as the progression of problems regarding motor skills and biomarkers in the brain, as illustrated by the quote below by an AD dementia care partner.
Quote 6: "How long is he able to cycle?" Also, some participants experience AD as a hopeless process, and they therefore want to know what their life expectancy is, like this AD dementia patient: Quote 7: "It is a hopeless and long process. When can I expect the end?" For most participants, it appeared to be important to know more about the progression of the symptoms. However, a few participants expressed not being interested in the course of the symptoms, as illustrated by this quote from an AD dementia care partner. Quote 8: "We live our lives from day to day. So, I do not want to know anything about symptom progression." 3.2 Online survey Table 1 presents the sample descriptives of the 232 participants from 13 European countries who completed the online survey. Of the 99 patients, n = 58 (59%) were female and median (IQR) age was 67 (61-73) years. There were n = 42 (42%) patients with dementia, n = 29 (29%) with MCI and n = 28 (29%) with SCD. The 133 care partners had a median (IQR) age of 62 (52-72) years and n = 78 (59%) were female. There were n = 103 (77%) care partners of a patient with dementia, n = 23 (17%) with MCI and n = 7 (5%) with SCD. The majority of the care partners was a life partner (spouse) of the patient (n = 76; 57%) and was living with the patient (n = 83; 62%).   Figure 1 and the appendix display how frequent every item, grouped by category, was selected as one of the three most important outcomes by patients and care partners. For both patients and care partners, cognition-related items were most frequently selected, followed by functioning and dependency. Compared to patients, care partners selected more items related to 'behavior and neuropsychiatry' and 'physical health' as most important. Seven patients (7%) and three care partners (2%) indicated they did not want to know anything about the progression of the symptoms.
The top-10 lists of items most frequently mentioned as most important by patients and care partners were quite concordant, since seven items (70%) were selected by both groups. The overlapping seven items were memory deterioration, recognizing friends and family, patients' ability to convey what they want to say, take care of their self, no longer recognize others, loss of legal capacity and live in their own home. In addition, the following items were most frequently identi ed as most important by patients only: ability to participate in conversations, learning and planning and organizing. Orientation in place, life expectancy and expected cognitive symptoms were most often selected as most important by care partners only. We merged the seven matching items and the six non-matching items, into a core outcome list of 13 most important items as shown in Table 2. The majority (9 items) of these 13 items is part of the category 'cognition', followed by three items in the category 'functioning and dependency' and one item in the category 'physical health'. There were no items from the category 'behavior and neuropsychiatry' and 'social environment' in the core outcome list. Of note, the item 'How long am I able to decide whether I want euthanasia?', which was only included in the Dutch survey, was the 5th most important outcome in the Dutch sample, selected by six (16%) of Dutch participants.

Discussion
In this pan-European study, we identi ed a core list with outcomes relevant to patients and care partners in the disease trajectory of Alzheimer's disease (AD). Patients and care partners largely identi ed similar outcomes as most important, including memory deterioration, loss of legal capacity, no longer recognizing (signi cant) others, and the patients' ability to convey what they want to say, take care of their self and live in their own home. Most selected outcomes were from the category 'cognition', followed by items in the categories 'functioning and dependency' and 'physical health'.
Many patients perceive the neuropsychological examination as stressful, yet our results show that they highly value a prognosis in terms of cognitive outcomes. Patients and care partners are keen to know about the (expected) course of the cognitive outcomes. 17 The current study underlines that we should bridge the gap between cognitive test scores and daily functioning, by making an effort to translate numerical cognitive test results to the cognitive outcomes such as mentioned in this study.
Contrary to our expectation, outcomes from the category 'behavior and neuropsychiatry', such as anxiety and apathy, were not often selected by our participants as most important. This nding is in correspondence with results from a previous study, reporting that anxiety and depressive symptoms were not highly prioritized by patients with mild cognitive impairment (MCI) and their care partners. 11 Nonetheless, behavioral and psychological symptoms are highly prevalent across the entire spectrum of AD, with 96% of the patients showing at least one symptom. 18,19 Former studies reported that these symptoms were associated with increased risk of institutionalization, di culties in daily functioning and caregiver burden. 20,21 We could speculate that symptoms from the category 'behavior and neuropsychiatry' may be more relevant at a later stage, or that patients and care partners may not have realized that these symptoms are related to AD. Alternatively, these symptoms may not be the most striking aspect of managing AD in everyday life, and this could be different in other types of dementia such as dementia with Lewy bodies or frontotemporal dementia, where behavioral and psychological signs and symptoms are more in the foreground. 22,23 We observed large variability in the importance rating of individual items, as 33 out of the list of 59 were rated as 'most important' by at least one patient or caregiver. When scrutinizing the individual items, some showed some overlap with other items from the same category. For example the items 'How long am I able to recognize my family and friends' and 'When do I no longer recognize anyone?' from the category cognition showed resemblance since they both focused on the ability to recognize others. Nevertheless, on a category level, it was clear that most patients and care partners identi ed items related to cognitive decline and functional dependence as most important. In view of the large variability in items selected as most important, it was even more remarkable that outcomes deemed relevant by patients and care partners were very similar. Seven out of ten items in both the patient and care partner top10-lists were concordant, justifying us merging these lists into one core outcome list of 13 items. For both patients and care partners, outcomes from the category 'cognition' and 'functioning and dependency' were identi ed as most important. However, the differences are also interesting. Care partners for example indicated a strong wish to know the patient's life expectancy. Patients on the other hand emphasized the importance of maintaining their ability to participate meaningfully in conversations. By merging the top-10 lists of both patients and care partners into one core outcome list, the perspectives of both stakeholder groups are equally well represented.
To our knowledge, this is the rst study that has investigated outcomes of relevance to patients and care partners in the disease trajectory of AD (AD dementia, MCI and SCD) in multiple European countries, using a mixed-method approach. A former review reported comparable results in relevant outcomes of patients with MCI or AD and their care partners and health care providers. 24 This review identi ed some outcomes mentioned in our study, e.g. memory, mental health, activities of daily living and maintenance of identity. Outcomes that we did not replicate are eating behaviors, apathy and self-e cacy in the ability to manage memory impairment. In addition, a recent qualitative study examined what matters most to patients and care partners across the AD continuum, by means of interviews only. 25 They also observed that memory (e.g. forgetting friends/family) was reported as one of the most challenging issues by the majority of the participants. In addition, they reported that the impacts of AD vary across the diseaseseverity spectrum. Emotional impacts (e.g. more frustrated) and social impacts (e.g. decreased social activities) were most commonly reported in the early stages of AD. However, care partners of patients with mild to severe AD reported impacts on their own daily responsibilities as most important.

Strengths
One of the strengths of our study is the two stepped, mixed-methods approach, whereby we rst used a qualitative bottom up strategy to generate a comprehensive list of outcomes, and then a quantitative approach to provide a selection of most important outcomes from the perspectives of a large sample of patients and care partners. Also, we included participants from 13 different European countries, which greatly enhances generalizability.

Limitations
Among the potential limitations is our use of an online survey, which may have led to an underrepresentation of patients with severe cognitive symptoms and computer illiterate patients. On the other hand, one might argue that with diagnosis of AD occurring in an increasingly early stage, the preferences and wishes with respect to prognosis of these groups are most relevant in this context.
Another potential limitation is that identifying the most important items in the online survey was challenging for participants. To support participants in identifying what mattered most to them, we used a stepped approach that made it easier for the participants to select the three items they considered most important. In addition, the fact that behavior and neuropsychiatry were not mentioned as relevant outcomes may be partly due to patients and care partners being unaware that these outcomes are possible and quite likely consequences of AD.

Implications for clinical practice and further research
Information on individual disease trajectories can support patients, care partners and health care providers to plan and organize care. We will use the results of the current study to collect patient-reported outcomes (PROs) in the context of the European ADDITION project. Subsequently, we will integrate these PROs into a disease progression model, which could help to provide patients and care partners with personalized information on their expected disease course. A previous study constructed biomarkerbased models for prediction of progression to dementia in MCI patients. 26 The next step is to predict other outcomes relevant to patients and care partners.
Based on our results, it can be debated whether the data we routinely collect during follow-up at the memory clinic (i.e. medical exam, cognitive testing) su ciently re ects the outcomes that are most relevant. We should focus more on data collection of PROs, such as identi ed in the current study.
Moreover, previous studies investigated the views of healthcare professionals regarding important outcomes in AD, and reported driving, 13 patient and care partner quality of life, 13, 27 maintaining personality and identity 13 and behavior changes 13, 28 as relevant. These did however not end up in our core outcome list from the perspective of patients and care partners. This would suggest that important outcomes from the perspective of healthcare professionals are additionally relevant and could also be taken into account when investigating PROMs in AD-studies. Of note, seven (7.1%) patients and three (2.3%) care partners reported that they did not want to know anything about the progression of the symptoms in our study. Therefore, it is important to adopt a process of shared decision making in memory clinics , 29 in which patients, care partners and clinician decide together on the prognostic information about the course of AD which is most relevant to patients and their care partners.
In addition, to follow patients more e ciently, we could make more use of online opportunities, and this has particularly become relevant in the current times with COVID19 related regulations that hamper physical visits. We will incorporate our core outcome list in a large-scale online data-collection of PROs, supplemented with outcomes that are deemed relevant from the perspective of the professionals (e.g. quality of life and behavior changes). In our online data collection, in addition to questionnaires on quality of life and activities of daily living, we include questions about their social environment, patients' ability to participate in conversations, to recognize loved ones and to keep up hobbies. Some outcomes identi ed in our study are too abstract or otherwise impossible to inquiry by means of a questionnaire (e.g. how long am I able to convey what I want to say?). Acknowledging these uncertainties and the impossibility to adequately answer these questions might help them cope. 30

Conclusion
This study resulted in a comprehensive core list of 13 outcomes in the AD trajectory relevant to both patients and care partners. These outcomes can be incorporated in patient care, research, for example via online survey. Integration of these outcomes into a disease progression model can help to provide patients and care partners with personalized information on prognosis of AD. Centre Ljubjana and Institute of Neuroimmunilogy AD centre Slovakia). All participants provided informed consent.

Consent for publication
Not applicable.

Availability of data and materials
The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request.  Figure 1a and 1b show how frequent every item, grouped by category, was selected as one of the three most important outcomes by patients ( gure 1a) and care partners ( gure 2b). The items corresponding to each number can be found in the appendix. The asterisks indicate the 10 items most frequently identi ed as most important by patients and by care partners, included in the core item list. Since seven items overlap, the core list comprises of 13 items.

Supplementary Files
This is a list of supplementary les associated with this preprint. Click to download.