Challenges in disclosing and receiving a diagnosis of dementia: a systematic review of practice from the perspectives of people with dementia, carers, and healthcare professionals

ABSTRACT Background: Disclosing a diagnosis of dementia is a key process involving people with dementia, carers, and healthcare professionals (HCPs) that can facilitate access to treatment and support. Receiving a diagnosis of dementia may represent a change in identity and loss of a planned-for future, resulting in an emotional impact for both people with dementia and carers. Delivering the diagnosis of dementia can be difficult and draining for HCPs. Methods: We conducted a systematic review that included studies which explored the experience of giving or receiving a diagnosis of dementia from the perspectives of people with dementia, carers, or HCPs. All study designs were eligible except for previous literature reviews. Findings were analyzed thematically and grouped into categories and then synthesized into a narrative review. The quality of all included studies was assessed. Results: Fifty-two studies were included in this review. Findings indicated that receiving a diagnosis is generally a negative process for people with dementia, carers, and HCPs and leaves carers in particular feeling uncertain over the prognosis and future of the person they care for. Disclosing a diagnosis of dementia is a difficult and complex process, for which formal training and guidance is lacking. Carers in particular would welcome more opportunity for realistic and hopeful discussions of the implications of receiving a diagnosis of dementia. Conclusions: Changes in some aspects of disclosure, such as providing a truthful diagnosis to the person with dementia, have occurred over the last decade. A process approach involving pre-diagnostic counseling and follow-up appointments could enable discussions regarding prognosis and the future, create opportunities to clarify the diagnosis, and reduce emotional burden on HCPs. There is a need for more objective evidence that considers the perspectives of all individuals involved.

Receiving a diagnosis of dementia is an important step for people with dementia because it facilitates access to interventions, and potential support from health and social services and third-sector organizations (Robinson et al., 2015). Receiving a diagnosis may be an emotionally charged experience, representing a transition in identity for people with dementia and their carers requiring an emotional readjustment and reappraisal of their future (Robinson et al., 2011).
The proportion of those developing symptoms of dementia who receive a diagnosis has increased in the UK (Abhayaratne et al., 2019) and internationally (World Health Organization, 2020), partly in response to policy changes (Department of Health, 2015), and due to greater awareness and acceptance of dementia. A recent systematic review (van den Dungen et al., 2014) using a pooled average based on 9,065 respondents from 23 studies demonstrated that approximately 85% of people with cognitive impairment would wish to be told a diagnosis if one were made. Knowledge of prognosis and illness trajectories enables people to plan and ensure their affairs are in order or undertake lifestyle changes (Woods et al., 2019). Typically, in developed countries, diagnosis disclosure occurs in secondary care settings, conducted by professionals such as psychiatrists, clinical psychologists, or specialist nurses. However, some diagnosis disclosures occur in primary care, delivered by a primary care physician (PCP). The time from developing awareness of cognitive changes to receiving a diagnosis varies across individuals, but can be a lengthy process; a delay of 3 years is not uncommon (Chrisp et al., 2011). The communication and delivery of the diagnosis require careful management to account for a variety of reactions, needs, and levels of understanding from both the person with dementia and carers accompanying them to the disclosure meeting (Bunn et al., 2012).
There are differences between healthcare professionals' (HCPs) perceptions of best practice when disclosing a diagnosis of dementia and the perceptions of people with dementia or carers, which raises ethical dilemmas about how to respect these different needs (Dooley et al., 2015). An earlier literature review (Lecouturier et al., 2008) identified a list of best practice behaviors and combined this with a qualitative exploration and consensus approach to identify eight categories of best practice for disclosure. These categories comprised preparing for disclosure, integrating family members, exploring the patient's perspectives, disclosing the diagnosis, responding to the patient's reactions, focusing on quality of life (QoL) and well-being, planning for the future, and communicating effectively. While comprehensive, this review is now over 10 years old and the authors did not critically appraise the included research. Another review (Werner et al., 2013) covering a similar topic and time frame addressed a broad range of topics related to dementia diagnosis disclosure, but without a specific focus on which practices are typical and how they might be perceived. Given the recent policy initiatives around early diagnosis and changes in societal awareness of dementia over the last decade (Department of Health, 2015), an updated review of the evidence regarding disclosure from the perspectives of people with dementia, carers, and HCPs is required.

Objectives
This review aimed to explore the experiences of giving or receiving a diagnosis of dementia. It identified common disclosure practices, challenges associated with disclosing the diagnosis, and needs of different individuals involved in the disclosure. It extended earlier work by including a systematic search strategy and a critical appraisal of the results using an established quality assessment tool and went beyond describing the range of possible practices and behavior to identify what is reported as typically occurring in practice and how this is perceived.

Search strategy
A search of the PubMed, CINAHL, and Web of Science databases was conducted on 31October 2018; updated on 7 February 2020. Reference sections of included papers were hand-searched to identify further publications. The terms "dementia" AND "diagnosi*" were used to search titles and abstracts. A second search was conducted which combined these with the terms "disclosure" OR "best practice," used to search in all fields. The study flowchart is shown in Figure 1. Due to the large number of studies conducted on this topic, the search was restricted to the last 10 years (2008 onward).

Study selection
Studies were included if they were written in English and provided information on delivering or receiving a diagnosis of dementia. All study types were eligible except for papers that only included a review of previous literature. Studies from the perspectives of HCPs, patients, and carers were all considered. The two primary reviewers (JY and MS) screened the titles, excluding papers that were not relevant, and then reviewed the abstracts. Full texts were retrieved if the title and abstract suggested that an aspect of delivering or receiving a diagnosis of dementia was explored. JY and MS conducted this task jointly, resolving differences in judgment through discussion and agreement.

Data collection and synthesis
Data were extracted using a table developed by JY and MS to capture the study details (Table 1). Initially, JY and MS extracted the data together to ensure consistency, before continuing to extract data independently and amalgamating tables once complete. As data were extracted, a number of similar ideas and issues were identified in the findings. These were collated in a separate document and added to by JY and MS until data extraction was complete. Similar findings were grouped to form thematic categories. These themes were refined and amalgamated where possible through discussion and during the writing of the narrative by the whole research team. Themes were reported in a narrative format.
Quality review Methodological quality was assessed by JY and MS using the Joanna Briggs Institute (JBI) checklists (Aromataris and Munn, 2017). The corresponding checklists for each study design were used. A rating of methodological quality that was comparable across different designs was made by calculating the number of items endorsed on each checklist for each paper and converting this into a percentage. Ten percent of the papers were sampled using a random number generator and assessed for quality by a third reviewer (AH). Reviewer scores were compared for consistency using a paired samples t-test. No significant differences were found (t(4) = 2.33, p = .080), suggesting a robust quality assessment.

Results
Fifty-two studies were included. Main findings are summarized in Table 1. Findings are categorized under the following themes: content of the diagnosis disclosure; emotional impact; communication of the diagnosis; people involved; attitudes toward diagnosis disclosure; use of diagnostic tests and assessments; truth telling and deception; timeliness; and training and skills.

Quality assessment
Quality assessment scores are shown in Table 1. Detailed assessments can be obtained on request. The scores represent the percentage of items on each study design-specific checklist endorsed, and higher percentages reflect higher quality. Scores ranged from 60% to 100%, and of the 52 studies, 13 scored 100%, seven scored 90-100%, 14 scored 80-90%, 12 scored 70-80%, and six scored 60-70%. Studies scoring lower tended to have qualitative designs. The most frequent quality issues included a failure to specify the underpinning philosophy of the research approach, so the congruence between the philosophical perspective and the research methodology was unknown, and a lack of reflexivity to locate the researcher culturally and theoretically, acknowledging the impact of the researcher on the research and vice versa. Studies scoring higher tended to utilize survey designs, consensus methods, and observations.

Study design types
This review included 1 case report, 20 crosssectional studies yielding quantitative data, 5 studies that were opinion pieces, commentaries, or utilized consensus methodologies with experts, 24 qualitative studies, 2 quasi-experimental studies, and 1 randomized controlled trial.

Participant characteristics
Participants were HCPs, people with dementia, carers, and people who had attended memory assessment services but not received a diagnosis of dementia at the time of participation. Sixteen studies Dementia diagnosis disclosure: A systematic review 1163 To investigate how PCPs in Japan approach disclosure of a dementia diagnosis to patients and family members.
Diagnoses were disclosed jointly to the person with dementia and their family members for all participants, and cooperation and well-being of family members was important to physicians in maintaining QoL for the person with dementia. Physicians were concerned about the potential negative psychological impact on patients, the degree to which patients understood the information provided, and the stigma associated with dementia. Almost half of the sample reported they did not receive training in disclosing a diagnosis of dementia and were unsure if their approach was appropriate. To investigate the views and experiences of doctors making and delivering diagnoses in memory clinics.
Organization of some memory clinics involves allied health professionals conducting assessments, so doctors disclosing the diagnosis can be meeting that patient for the first time, which feels unnatural and makes it difficult for the doctor to gauge the patients' readiness and expectations. The diagnosis was seen as a useful framework to involve other services and open discussions about possible therapies. Participants reported managing the different audiences (patient and companion) in the consultation was challenging and balancing the needs of both could lead to conflict. Few participants reported receiving specific training and most relied on basic principles taught in medical school, and there was a lack of supervision and support in discussing the emotional impact on doctors themselves. Participants tended to use the word dementia but some approached it gradually by introducing the idea of memory problems first. There was a tension between presenting a hopeful picture, while also being honest, and reflecting that there is uncertainty in the diagnosis. Participants reflected that due to service pressures there just wasn't enough time to cover everything they wanted to in a disclosure meeting and consequently follow-up appointments are crucial.  To evaluate safety and tolerability of disclosure of elevated amyloid levels shown on PET scans.
There were no differences between the amyloid elevated group and the amyloid non-elevated group for depression at any time point. Anxiety increased in amyloid elevated participants at a low level on the day of disclosure but was not sustained at 6 weeks or 6 months. Higher levels of anxiety and depression at baseline were modestly predictive of levels of anxiety and depression at follow-up. Amyloid-elevated participants did not regret learning the outcome of their scan. The authors note that discussion of the scan frames the results conceptually as a risk factor for development of AD rather than as a diagnostic tool. Further, the authors stress using terms "elevated" rather than positive to avoid confusion of positive being beneficial.  Most individuals and their companions seeking a dementia evaluation do not experience adverse psychological reactions when they receive diagnostic feedback. Symptoms of anxiety and depression remain stable or even decline immediately after diagnosis. Severity of dementia, age, sex, and education appear to have no significant effect on this outcome. Little change was seen in depressive symptoms. In contrast, symptoms of anxiety seem to decrease after diagnostic feedback. In particular, individuals who started the evaluation process with high levels of anxiety were likely to experience significant relief after they received diagnostic information.

80%
Caruana-Pulpan & Scerri 2014: Practices in diagnosis, disclosure and pharmacotherapeutic management of dementia by GPs: a national survey.
193 GPs in Malta Survey, Malta National survey To explore, via a national survey, the practice patterns of GPs with respect to their ability to diagnose, disclose, and pharmacotherapeutically manage dementia in their primary care setting A considerable number of GPs (62.2%) indicated that they do not routinely disclose the diagnosis, even though more than half of the respondents were in favor of the fact that disease disclosure may actually help the patient and the caregiver in future planning and treatment decisions. On disclosing, GPs are unsure of the suitable terminology to use in describing the condition to their patients and/or caregivers, with dementia being the most likely used term (33.7%) followed by memory problems (26.4%). Most physicians would prefer to disclose on being sure of a correct diagnosis and would do so to both the patient and the caregiver or relative. To understand the informal caregiver's experience of being present with a loved one when he or she receives a diagnosis of dementia, and what the constituents of the meaning that they assign are.

80%
Caregivers are usually aware that there is something wrong with the person they care for and can experience feelings of relief and validation when the diagnosis is disclosed, which helps lead to a sense of acceptance. Alongside, this were emotional responses that involved feeling sad, terrified, worried, and overwhelmed, but coupled with a need to stay strong. Caregivers watched for an emotional response in the person with dementia when the diagnosis was disclosed and reported seeing fear, sadness, and perhaps confusion, but sometimes no response. Caregivers reported wanting a road map to help them navigate the future and understand how they might care for the person with dementia. The primary purpose of the present study was to examine potential benefits of and barriers to diagnosis from the perspective of black and white adults directly affected by AD; a secondary purpose was to explore black-white differences in these perceptions.
Family members affected by AD endorsed a wide range of the benefits but few of the barriers to obtaining a diagnosis examined in this study. The most frequently endorsed benefits pertained to obtaining information, finding out what was wrong, and prompting future plans. In addition to the lack of a cure for AD, the beliefs that little can be done for someone with AD, that there is a lack of effective treatment, and that obtaining a diagnosis was a demanding process for families were the barriers most frequently endorsed. Black respondents expressed more positive views of obtaining a diagnosis than their white counterparts. Black respondents more likely to strongly endorse a number of benefits, but they were also less likely to view the fact that there is currently no cure or effective treatment as a barrier to diagnosis To achieve a greater understanding of dementia diagnosis from the perspective of GPs.
GPs reported that rather than disclose the diagnosis using explicit terms, a discussion over a number of consultations is preferred, where the PwD can begin to realize for themselves that they have memory problems. GPs made judgemnts about whether the PwD would be comfor- To explore the construction of morals while highlighting the social, collaborative, and processual nature of reaching a diagnosis.
Each person should be treated individually and there are no text book rules about how to disclose. Clinicians expressed a tension between recognizing that the patient is an individual with the capacity to make free and reasoned choices, and that the diagnosis cannot be disentangled from the perceptions, needs, and concerns of other people in the patient's life who might need more information. Disciplinary and organizational cultures play a role in framing the consultation, where the impetus is on achieving the correct diagnosis in a reliable way, rather than on how it is delivered. Clinicians face a tension between wanting to use time as a resource to build up to delivering the diagnosis, but also feeling under pressure to not provide another appointment and instead to apply a diagnostic label, diagnose promptly and free up appointments for other people in the face of increasing caseloads. Visual evidence from CT scans can help clinicians frame the diagnosis in an objective way and show the transition from memory problems to a diagnosis and move the conversation toward treatment and intervention plans. Innes et al., 2014: Dementia diagnosis and postdiagnostic support in Scottish rural communities: experiences of PwD and their families 18 participants (6 PwD and 12 carers)

Scotland
Qualitative interviews To report service user views about diagnostic processes in a remote and rural region.
Most diagnoses were delivered face to face by a clinician. Participants wanted to receive an explanation and more information and support regarding available care. Participants reported that nothing was explained when they were diagnosed. Participants reported feeling the diagnosis came as a shock, despite an awareness of their memory problems prior to assessment.

70%
Kaduszkiewicz et al., 2008: Telling "the truth" in dementia -Do attitude and approach of GPs and specialists differ?
30 GPs participated in qualitative interviews 307 respondents to the questionnaire (211 GPs and 96 specialists)

Germany
Qualitative interviews and a questionnaire developed from the interview data To systematically investigate differences between GPs and specialists with regard to diagnosis and disclosure of dementia.
GPs reported that patients deny their diagnosis when it is disclosed and try to convey that all is well. GPs worried about ruining the doctor-patient relationship and avoid conflict with the patient. 21 GPs reported using language other than dementia or Alzheimer's and instead use terms like normal aging process or attribute it to circulatory issues. Both specialists and GPs tended to report more details of the diagnosis to the relatives rather than the patient and use explicit terms. Only one-third of specialists and half of GPs used the terms Alzheimer's and dementia with patients. A large proportion of GPs strongly (35%) or partially (40%) confirmed that patients felt ashamed when confronted with signs of cognitive impairment. Results from specialists were similar. 50% of GPs and 30% of specialists were interested in further training concerning communication with PwD and relatives. Many patients expressed disappointment or discontent following the consultations due to an absence of tangible solutions or due to the physician's style of interaction and communication with the patient and companion. For example, patients felt disempowered when physicians spoke directly to the companion instead of them. There is a difference in perspective between patients who endorsed their memory difficulties and those who did not, with the latter experiencing more offense. Companions wanted an explicit explanation of what they can and should do, where they might receive support and guidance. However, companions expressed that they were not given solutions and guidance. Some companions felt a follow-up session to deepen their understanding was necessary. The lack of details provided in the session left companions feeling insecure about how to proceed. Some companions reported that the physician expressed impatience and even contempt toward the PwD, making rude interruptions. Some companions, however, expressed that they were glad someone was listening and was sympathetic to their feelings. Companions whose struggles and efforts were recognized and acknowledged felt reassured and empowered. To evaluate the consequences of amyloid beta (AB) disclosure on mood, subjective sense of memory impairment, lifestyle, and perceived risks of AD.

75%
Participants who had a negative AB result were relieved but conscious and it could change over time. Participants with a positive AB result were anxious but not surprised and had shared this information with others and sought additional information on the internet. AB positive participants had made lifestyle changes and experienced no negative effects on their mood or subjective sense of memory impairment. A psychoeducational brochure provided helped participants to understand the link between environmental and genetic risk factors and the future progression to AD, and they found it useful and informative. The brochure did change perceptions of risk for participants (either AB positive or negative). Surveys revealed that participants were satisfied with the disclosure process. All participants had also taken a health literacy test, and half of those with MCI showed a lack of health literacy, confirming the need to have disclosure sessions with a companion present. A theme from the focus group of best practice recommendations suggested the following ideas: offer pretest counseling, use clear graphics, review patients' brain image scans in the disclosure, offer take-home materials describing follow-up options, call patients postdisclosure to answer emerging questions, and communicate seamlessly with primary care providers. A second theme was that knowledge is power, and despite the outcome of the tests participants generally felt that knowing enabled them to be strong and to make decisions about their futures. Patients and carers felt that communication was not necessarily a dialog as although there was time for questions in the consult, many did not know what to ask or what information they might need. Some felt they received enough information, but others felt their concerns and questions were disregarded by physicians telling them they were okay. Support tended to be generic rather than person-centered and in response to their concerns. Several patients reported that the disclosure should be direct, even though it would be a shock, although some patients felt it depended on the individual and should be done sensitively. Carers' views were very similar. Patients felt that the doctor should be sure of the diagnosis before it was disclosed. Attributes of doctors considered a requirement for this role were supportive, being a good listener, and being easy to understand. Carers felt that they did not receive enough information during the disclosure. 55% of patients preferred to receive all information about the diagnosis, treatment, and prognosis all at once and up-front, whereas 35% preferred to receive it in stages. 75% of patients felt that written information should 65% Reactions to receiving a diagnosis described by both patients and clinicians involved shock, fear, and acceptance, but the most common reactions are denial and avoidance. Patients use avoidance to manage their loss of sense of self and anxieties about the future, and clinicians use avoidance by not using explicit terms due to a lack of certainty about the diagnosis, and a desire to minimize distress, as well as respect for patients' wishes not to know the diagnosis. Supportive staff at the MAS put patients at ease and help them to engage with the process. However, a diagnosis delivered without provision of information contributed toward anxiety. Clinicians reported that working with other members of health and social care helped to facilitate post-diagnostic support. Working with other staff also helps to increase confidence in the diagnosis. Clinicians reported that disclosure was draining and that patients' distress does affect them. Being able to share this 29% of patients declared that they had suffered from the disclosure, but caregivers reported only 15% of patients were still suffering. 85% of caregivers felt the diagnosis had been useful. Patients who could remember their diagnosis significantly more frequently reported that they had suffered compared to those who could not recall, and that they were more anxious. Caregivers of patients who could remember their diagnosis reported significantly more frequently that the patient had suffered and was still suffering from the disclosure, with a reaction of sadness, depression, anxiety, but also satisfaction to receive an explanation. The disclosure of AD was responsible for anxiety or sadness in approximately one-third of patients.

60%
Nielsen et al. 98% reported that disclosure is almost always/often performed by a physician and 61% reported that a nurse almost always/often participated in these meetings. 54% never/rarely disclosed on the first encounter. 56% had the opportunity to find out expectations before the meeting, but for 31% this was never/rarely possible. 98% of participants almost always/often discussed pharmacological treatments at the disclosure meeting and 64% offered follow-up (although 28% never/rarely did so). 95% provided information on psychosocial support. 81% provided written materials to support the verbal information. Patients were encouraged to bring a family caregiver to the disclosure meeting and most participants disclosed to them jointly. 78% of participants regularly used the term AD or dementia of the AD type (52%), or dementia (39%). No respondents used terms senility or age-related impairment. Most respondents shared results from brain scans/assessments and informed on etiology, progression, causes of symptoms. However, few respondents gave information on prognosis and future symptoms, and the amount of information given was dependent on the degree of cognitive impairment, the specific type of dementia, and the level of emotional distress. 100% of respondents discussed issues such as handling medications and 87% discussed driving. Page et al., 2015: "Rydym Eisiau Gwybod" the dementia diagnosis disclosure preferences of people in North Wales 253 PwD attending memory assessment services North Wales, UK Clinical audit To quantify patients' preferences for diagnostic disclosure 84.1% of patients included in the audit preferred their diagnosis to be disclosed jointly to themselves and a relative or 2.7% to themselves and a third party (a carer). 8.3% of patients preferred to receive their diagnosis alone. All but one of the 4.7% of those who did not want to know wanted their diagnosis to be disclosed to a relative. Patients who expressed a preference for non-disclosure were 5.74 times more likely to not have their preferences actioned than those requesting disclosure. Patients requesting joint disclosure were less likely to have their preferences ignored, suggesting the companion represents a reassuring presence not just for the PwD but also for the healthcare professional. To understand why previous literature labels the disclosure process as difficult and to generate patient-oriented advice on the process Patients tend not to agree or acknowledge the news delivered by the doctor, even when doctors take an affiliative conversation style by including the patient in the conversation. Bad news tends to be "shrouded" with positives. Some interactions do not necessarily need the diagnostic labels to be explicit for the diagnosis to be communicated. Unambiguous naming of dementia/AD does not always help the patient understand what is happening. Diagnosis is an ongoing, repetitive discussion between healthcare professionals, patients, and carers. Patients themselves tend to normalize their experience, and doctors interact with this, which presents the doctor with a situational dilemma of having a smooth consultation but also not deceiving patients and giving them the opportunity of a timely diagnosis. Disclosure is easier if the patient is already thinking about that diagnosis. GPs in this study reported feeling a sense of trepidation about disclosure as patients were often fearful of the diagnosis. GPs felt that if there was a well-developed doctor-patient relationship, they should be the one to disclose though, rather than a specialist who may not be known to the patient. However, some participants felt that the specialist could occupy a position of blame, and the GP was then in a position to support their patients. Some GPs felt that disclosure would impact the doctorpatient relationship. Family members were felt to be helpful as they usually wanted to know more information, which the GP could focus on, especially if the patient discounted the implications of the diagnosis. Providing education to carers about the patient's behavior was felt to be important to help them cope with the consequences of the condition. GPs felt it was important to offer hope during the disclosure by confirming peoples' fears and giving them a constructive way to move forward. Some GPs reported using terms such as memory impairment rather than dementia to avoid negative connotations of the word dementia. To consider if and how diagnosis should be disclosed in two cases from an ethical perspective.

80%
Patients should be given the choice whether to receive the results or not before the assessments take place, and patients should have the chance to change their mind at any point. The authors suggest that due to the lack of effective treatments, a truthful diagnosis disclosure is not necessary to guarantee the best available treatment in prodromal AD. Disclosure and the terminology used should be based on the patient's current socio-personal context, considering their mood and family support. Biomarkers should be treated with caution as the presence of biomarkers does not necessarily translate into the same clinical situation in the future. Many participants in this study felt supported by practitioners at the time and that details of their condition had been explained to them. Some participants described feeling shock, loss, grief, and dejection, and some of these felt that the process of imparting the diagnosis has increased these feelings. Participants wanted to know the long-term prognosis but felt that a staged process of diagnosis disclosure would have helped them to take the news on board, as they had little time to discuss concerns and questions. Carers were keen to talk to the PwD before speaking further to the practitioner because this interaction tended to generate further questions. Participants expressed that they had lacked information and had only been given basic information, and practitioners had disregarded their need to know more. Not all types of dementia were supported, for example, participants who received a diagnosis of vascular dementia in this study were discharged without support from secondary care because they were not eligible for medication and reported feeling helpless, shocked, lost, and confused. To understand geriatricians' attitudes toward disclosure of AD, and whether these opinions differ from those of neurologists and neuropsychiatrists.

60%
Participants preferred to use the term memory disease (59%) over AD (48%) with PwD but preferred to use AD with relatives (87%). More than 55% of participants ask the PwD if they want to know their diagnosis, but only 39% ask the PwD if they can reveal it to their family. 18% asked the family if they should reveal the diagnosis to the PwD. 71% of participants provided information about the prognosis and progression of AD to the PwD and 94% did to the family. Most cited reasons for not informing PwD of this was that they did not ask (94%), or the participant was fearful of a depressive reaction (71%), or that the knowledge was useless to the PwD (29%) The most frequent emotional reactions of PwD on hearing the diagnosis were depressive reaction (50%), indifference (51%), anxiety/nervousness (46.8%), and denial/ minimization (40%). The most frequent emotional reaction from family members was anxiety (50% Participants wondered why professionals had not been more forthcoming with information and wondered if this was deliberate or accidental. Carers expressed a wish for information about typical manifestations of dementia and practical information about health, social, and financial support. Some carers were able to use the medical explanation provided as a framework to explain the PwD's behavior and changes. Others found the medical framework incomplete for understanding the condition, and this seemed to occur when there was a lack of explanation of the condition. For most participants, disclosure of the dementia diagnosis was paradoxical in that it provided relief and an explanation of the behaviors their loved one had displayed, but left carers in a state of uncertainty, with no idea about the course of the disease or expectations for the future. Carers receive a new role at the disclosure of the diagnosis, in an instant they go from not being a carer to being a carer, but without any education as to what this new role involves. Many respondents were annoyed at the lack of information. Disclosure took away hope and trust in the future for some participants as they associated dementia with a humiliating, progressive process involving poor care, and social exclusion. A small minority of carers preferred to receive the disclosure themselves and not tell the PwD. The disclosure protocol should be a repeated and planned announcement where the diagnosis is disclosed by the specialist, and again by the PCP to reaffirm. PCPs need training opportunities to integrate this disclosure into their consultation. Disclosure must occur before care planning and provide basic education to the family. Care plans should be presented to avoid causing excessive concern to PwD and families about the progression of dementia. Legal issues such as advance directives, designated people of trust, and driving can be covered at this time. Participants' understanding of well-being was underpinned by a nihilistic and reductionist view of dementia, embedded within the medical model that dementia represents a threat to well-being directly associated with a decline in functioning. Diagnostic disclosure meeting viewed as a key event in a person's life and an opportunity to discuss well-being and engagement with services, although some participants felt the shocking nature of the disclosure meant that well-being should be addressed at a later time point. Disclosure is only helpful in supporting well-being if people are ready to hear their diagnosis, and that early diagnosis can be detrimental to well-being. Participants felt that it was right to tailor language even though this raised dilemmas for whether to disclose the diagnosis in full and explicit terms. Participants expressed tensions between positivity and reality, as they wanted to provide positive messages that were genuine and realistic about the future. Participants felt that the diagnosis was a separate part of the discussion to well-being. Participants described a discrepancy between the care they wished to provide and the care they did provide, for example, the practical and emotional difficulties of disclosing a diagnosis to someone with whom they do not have a therapeutic relationship. To examine the carers' experience of the diagnostic process, disclosure of diagnosis, and impact of diagnosis over time.

90%
47% of carers felt the diagnosis was delayed, and contributing factors were the PwD refusing to be assessed (38%) and negative professional attitudes (33%) where the first professional seen did not consider anything wrong and being told there was no point in a diagnosis (7%). Carer ratings of quality indicators of sharing the diagnosis were generally favorable, except for pre-diagnostic counseling (did PwD want to know?), for the carer being able to speak to the healthcare professional alone, and for having a written summary. Higher-quality diagnosis sharing was associated with lower sadness, depression, and despair, and greater acceptance, reassurance both immediately after diagnosis, and beyond (average of 4 years). To quantify the content of diagnostic conversations to examine the prevalence of different types of discourse across stakeholders. To investigate relationship between elements of discourse to dementia severity and psychological outcomes.
With increasing dementia severity, patients spoke less in the interaction. Providers (healthcare professionals) spoke most frequently, accounting for an average of 73% of the dialog. Patients used lifestyle/psychosocial information and emotional rapport building features of dialog more than companions. Providers and companions gave more medical/therapeutic information as dementia severity increased. More positive rapport building from providers was associated with lower depression and anxiety scores for PwD.

90%
Xanthopoulou et al., 2019: Patient and companion concerns when receiving a dementia diagnosis: an observational study of dementia diagnosis feedback meetings 19 healthcare professionals 60 patients 59 companions

UK
Video observations of disclosure meetings This study explored concerns expressed in HCP-patientcompanion communication in specialist memory clinics during the dementia diagnostic feedback meeting.
Topics of concern for patients and companions included the dementia diagnosis itself, symptoms of dementia, impact on family, physical health of the patient, frustration with the symptoms of dementia, mental health, medication, prognosis, the role of the companion, lack of social connection, and other. All concerns were given a response by the healthcare professional, where 62% were encouraged to elaborate and 38% were discouraged from elaborating. When concerns were elicited by the healthcare professional, they were more likely to respond by encouraging elaboration, but comparatively when concerns were volunteer by the patient or companion the healthcare professional was more likely to respond by discouraging elaboration. To characterize the extent to which physicians adopt a patientcentered approach in disclosure. To evaluate whether patient-centered behaviors differ in relation to patient characteristics.
The most frequent patient-centered behavior was positive rapport building, in particular physician agreement with PwD/companion and showing approval. Facilitation and patient activation was the next most frequent category of behaviors, with back-channeling most commonly used (signaling interest with non-lexical utterances), followed by checking their understanding, and the PwD/companion's understanding. Physicians asked for PwD/companion opinions, permission, or reassurance less frequently. Emotional rapport building was the least used category of behaviors and occurred infrequently. Physicians rarely offered statements of reassurance or optimism and infrequently expressed concern or worry.

80%
Few instances of empathy were demonstrated, PwD/companion feelings were infrequently legitimized, and physicians tended not to self-disclose or create a sense of partnership with the PwD/companion. There was considerable variation between physicians, but physicians were generally consistent in their behaviors within themselves. To provide a systematic evaluation of agreement among physicians, PwD, companions, and other professionals about the results of diagnostic workup.
Consensus was moderate but far from perfect, and generally PwD had only fair agreement with other sources. PwD and companions felt feedback was thorough, suggesting self-perceived comprehension was high. Physicians felt that PwD who agreed with the disclosure understood the feedback better than those who did not agree. PwD tended to report no dementia when other sources believed they had dementia. To obtain the attitude of carers toward disclosure of AD to patients.
For carers of PwD who knew the diagnosis, half felt the PwD had suffered emotionally from knowing. 95.7% of carers would want to know themselves if they had dementia. 97.6% of carers would want the diagnosis disclosed to a family member if they had dementia. Reasons for disclosure included having a right to know, slowing the disease, facing the condition positively, know what is happening to them, and plan for the future.
100% AD: Alzheimer's disease, GP: general practitioner, MAS: memory assessment service, OAMHTs: Older Adult Mental Health Teams, PwD: people with dementia included only the perspectives of HCPs, four studies included only the perspectives of people with dementia, one study included the perspectives of people without a diagnosis of dementia, and seven studies included only the perspectives of carers. Sample sizes ranged from 2 to 1409. The term "carer" was used in most studies to refer to the individual accompanying the person receiving the diagnosis to the disclosure meeting. Other terms including "relative" and "companion" were also used (Table 1). For brevity, the term "carer" is used in this review.

Themes
Content of the diagnosis disclosure In a survey of specialist physicians in Denmark, most respondents reported that they always or often provided information on etiology, progression, and causes of symptoms, and they tailored the information provided depending on the degree of cognitive impairment, specific type of dementia, and level of emotional distress (Nielsen et al., 2018). People with dementia and carers in other studies reported receiving only basic information and felt they lacked information (Samsi et al., 2014), or that nothing was explained (Innes et al., 2014). Messages conveyed should be positive and hopeful, while still being realistic (Vince et al., 2017;Phillips et al., 2012;Lee and Weston, 2011;Lecouturier et al., 2008;Grossberg et al., 2010). People with dementia and carers indicated that they wanted to know more about the future and prognosis (Lecouturier et al., 2008;Laakkonen et al., 2008;Grossberg et al., 2010;Lee and Weston, 2011), sources of support, and local community health and social services (Foley et al., 2017;Innes et al., 2014). The diagnosis disclosure should include information regarding well-being, such as how people with dementia can continue with life as much as possible, maintain their sense of self, and accept their identity as a person with dementia (Vince et al., 2017). Carers in two studies reported being unsure what to ask in the disclosure meeting (Manthorpe et al., 2013,;Laakkonen et al., 2008) and felt that approaching diagnosis in stages might be preferable to enable people with dementia and carers to absorb and process the information (Laakkonen et al., 2008;Lecouturier et al., 2008). However, in one study, 55% of respondents reported preferring to receive the whole disclosure up-front (Mastwyk et al., 2014).
The diagnostic process should be structured to ascertain the beliefs, expectations, and potential misconceptions that people with dementia or carers might have. This might include determining what people with dementia already know about dementia (Lee and Weston, 2011) and identifying patientcentered informational needs (Bennett et al., 2018) and may take place in pre-diagnostic counseling ahead of disclosure meetings (Lecouturier et al., 2008).

Emotional impact
Eighteen studies explored the emotional impact of receiving a diagnosis of dementia. Studies reporting on the emotional impact of the diagnosis on people with dementia suggested they had experienced negative emotions (Zou et al., 2017;McCabe et al., 2019), including feelings of depression (Segers, 2009;Laakkonen et al., 2008), anxiety or nervousness (Segers, 2009;Mormont et al., 2012;Lim et al., 2016), shock (Samsi et al., 2014;Milby et al., 2017;Innes et al., 2014), loss (Samsi et al., 2014), grief (Samsi et al., 2014;Laakkonen et al., 2008), dejection (Samsi et al., 2014), sadness (Mormont et al., 2012), fear (Milby et al., 2017;Phillips et al., 2012), and shame (Kaduszkiewicz et al., 2008). One study reported that the diagnosis disclosure also brought satisfaction at receiving an explanation (Mormont et al., 2012). Three studies reported that some people with dementia were either indifferent upon receiving the diagnosis (Segers, 2009), had not experienced significant distress (Zaleta et al., 2012), or showed initial anxiety that was not sustained over time (Burns et al., 2017). A further study reported that anxiety decreased in people with dementia after receiving their diagnosis, particularly for those with high anxiety prior to the diagnosis (Carpenter et al., 2008). Studies on the perspectives of HCPs suggested that some people with dementia deny or minimize the diagnosis (Segers, 2009;Kaduszkiewicz et al., 2008). Studies exploring the perspectives of carers or family members found they reported feeling anxious (Segers, 2009;Laakkonen et al., 2008), uncertain (van Wijngaarden et al., 2018Laakkonen et al., 2008), hopeless (Laakkonen et al., 2008), lonely (Laakkonen et al., 2008), that their hope and trust in the future was taken away (van Wijngaarden et al., 2018), and anxiety and uncertainty due to not knowing where to get help or what to do next (Laakkonen et al., 2008). One study found that carers reported a sense of relief and validation that their observations about something being wrong were correct, which in turn led to a sense of acceptance. However, this study also found that carers felt sad, terrified, overwhelmed, and worried (Champlin, 2020). HCPs reported feeling a need to manage their own emotional journey (Bennett et al., 2018). General practitioner (GPs) reported a sense of trepidation regarding the disclosure (Phillips et al., 2012) and were hesitant and expressed worry regarding the negative psychological impact upon people with dementia of the diagnosis disclosure due to stigma (Abe et al., 2019).

Communication of the diagnosis
Communication was explored in 21 studies. There were conflicting findings regarding using explicit terms such as "dementia" or specific labels like "Alzheimer's disease" when communicating the diagnosis to people with dementia and carers. Five studies reported that such terminology should be used (Bennett et al., 2018;Grossberg et al., 2010) and was commonly used Nielsen et al., 2018;Bailey et al., 2019), but eight reported that HCPs tended not to use these terms (Zou et al., 2017;Zaleta et al., 2012;Segers, 2009;Phillips et al., 2012;Moore and Cahill, 2013;Milby et al., 2017;Kaduszkiewicz et al., 2008;Hansen et al., 2008). Reasons for not using specific terminology were that ambiguous language may be preferable when HCPs themselves are uncertain of the diagnosis (Zaleta et al., 2012;Milby et al., 2017), and unambiguous terms were not always helpful in facilitating the person with dementia's understanding (Peel, 2015). Other reasons included minimizing distress (Milby et al., 2017;Bailey et al., 2019), avoiding negative connotations of the word "dementia" (Phillips et al., 2012), and avoiding pathologizing the condition (Abe et al., 2019). Cultural differences were highlighted, suggesting that colloquial terms were used because they were more familiar to the local community (Abe et al., 2019).
Considering communication patterns in consultations, one study found that HCPs tended to dominate the disclosure meeting by talking on average for 83% of the time. Emphasis shifted from the person with dementia to the carer even when the person with dementia was in the mild stages of the condition and still able to speak and contribute information (Wynn and Carpenter, 2017). While patient-centered communication was noted in one study, findings suggested that aspects of communication such as emotional rapport building were infrequent (Zaleta and Carpenter, 2010). Three studies indicated that written information should accompany disclosure meetings (Nielsen et al., 2018;Mastwyk et al., 2014;Grossberg et al., 2010).

People involved
There were mixed findings regarding who should disclose the diagnosis. One study suggested specialists should always take this role (Villars et al., 2010) with another suggesting that specialists could occupy a position of blame in contrast to the GP, who could provide support to patients (Phillips et al., 2012). However, this study also reported that GPs should take on the role of disclosure when a welldeveloped patient-doctor relationship existed because they would be known to the patient compared to a specialist. A Swiss survey reported that 75% of GPs disclosed the diagnosis themselves (Giezendanner et al., 2018). Other professionals have roles in disclosure, suggesting that discussing well-being should be undertaken by a multidisciplinary team (Vince et al., 2017), although psychiatrists wanted more time to discuss well-being during the disclosure to avoid being seen as someone who only diagnoses conditions and prescribes medication (Vince et al., 2017).
The majority of papers agreed it was beneficial for carers to accompany the person with dementia to disclosure meetings (Phillips et al., 2012;Page et al., 2015;Nielsen et al., 2018;Grossberg et al., 2010;Bradford et al., 2011;Lingler et al., 2016;Abe et al., 2019;Bailey et al., 2019;Giezendanner et al., 2018). Family members often wanted more information and could receive education to help them cope in the future (Phillips et al., 2012;Bailey et al., 2019).

Dementia diagnosis disclosure: A systematic review 1185
Carers represent a reassuring presence for the HCP (Page et al., 2015), can help them communicate the diagnosis more effectively to the person with dementia (Grossberg et al., 2010), and help to recall details of the diagnosis (Bradford et al., 2011;Bailey et al., 2019). One study found that, within its sample, people with MCI showed a lack of health literacy, suggesting carers were necessary to help process and understand the information provided (Lingler et al., 2016). Further, one study suggested that carers are usually most troubled by symptoms of dementia, and using disclosure meetings to reassure them is important (Abe et al., 2019). HCPs in one study suggested that fuller details of the diagnosis were disclosed to carers rather than people with dementia (Kaduszkiewicz et al., 2008). A key point here is the lack of people with dementia as participants in the included studies, which limits the ability to draw conclusions regarding their preferences for their own and others' involvement.
Attitudes toward diagnosis disclosure Studies investigating HCPs' attitudes toward disclosing the diagnosis found it can be perceived as draining (Milby et al., 2017); they might feel like the "grim reaper" when delivering bad news (Bennett et al., 2018) or worry that the process could ruin doctor-patient relationships (Kaduszkiewicz et al., 2008). Some psychiatrists may hold a medically rooted nihilistic and reductionist attitude due to a sense of hopelessness around potential interventions for people with dementia and a lack of appropriate services to support well-being (Vince et al., 2017). Geriatricians who did not provide information on progression and prognosis of Alzheimer's disease to the person with dementia reported this was because the person with dementia did not ask, or the HCP was either fearful of a depressive reaction or felt the knowledge was of no use (Segers, 2009).
Carers typically approached diagnosis disclosure as a useful event (Mormont et al., 2012) and felt that people with dementia had a right to know in order to begin treatment, face the condition positively, know what is happening to them, and plan for the future (Zou et al., 2017). Carers were divided on whether the diagnosis should be revealed in an up-front approach, or in stages (Mastwyk et al., 2014). Those who preferred an up-front approach felt it necessary for planning and managing issues such as driving cessation. Carers felt empowered when HCPs explained how and where to get treatment (Mastwyk et al., 2014) and tended to prefer the HCP openly informing the person with dementia of the diagnosis (Laakkonen et al., 2008). A good quality disclosure was associated with better carer adjustment in relation to sadness, depression, and acceptance (Woods et al., 2018).
Only two studies considered attitudes toward diagnosis disclosure from the perspective of people with dementia themselves, with one finding that people with dementia tended not to agree with or acknowledge the information disclosed by the HCP (Peel, 2015). The other study found that people with dementia would prefer to receive a direct disclosure even though it might be a shock (Mastwyk et al., 2014).

Use of diagnostic tests and assessments
Four studies suggest that, prior to the disclosure meeting, HCPs should discuss what people with dementia and carers can expect from assessments, especially from brain scans and biomarker testing, and potential limitations or implications of diagnostic tests (Visser et al., 2012;Porteri et al., 2010;Grill et al., 2017;Lingler et al., 2016). People with dementia and carers should have the choice to receive the results from assessments or not and should have the opportunity to change their mind at any point (Porteri et al., 2010).
Three studies suggested that brain scans should be reviewed in disclosure meetings (Nielsen et al., 2018;Lingler et al., 2016;Grill et al., 2017), so people with dementia and carers can see the images. Two studies offered caution regarding how results of amyloid scans were communicated, where explanations should state that presence of amyloid indicates a risk for developing dementia while absence of amyloid does not indicate absence of illness (Grill et al., 2017), and that biomarkers are not yet a diagnostic tool in themselves (Burns et al., 2017). One study advocated for provision of take-home materials to supplement the explanation of scans and assessments (Lingler et al., 2016).

Truth telling and deception
Three studies considered the aspect of truth telling or withholding the truth of the diagnosis from people with dementia. One study reported that HCPs felt a truthful diagnosis was not necessary due to a lack of effective treatments and potential impact on preexisting symptoms of anxiety or depression (Porteri et al., 2010). Conversely, a different study found that providing a truthful diagnosis did not create a negative emotional impact for people with dementia or carers and instead helped to relieve anxiety or depression for both (Carpenter et al., 2008). The third study reported difficulties in delivering the diagnosis as people with dementia may try to normalize their experience, and HCPs must balance this with not deceiving the person with dementia as to the diagnosis (Peel, 2015).

Timeliness
Three studies considered the timing of receiving a diagnosis of dementia, with one study reporting that nearly half of carers surveyed across five European countries felt diagnosis was delayed (Woods et al., 2018). Reasons for delay included people with dementia refusing to be assessed, negative professional attitudes, and being told there was no point in a diagnosis. Receiving timely diagnoses could assist people with dementia in remembering their diagnosis (Bradford et al., 2011), but another study indicated potential detriment to well-being if provided before a person had processed what is happening and accepted the changes they are experiencing (Vince et al., 2017).

Training and skills
As far as can be determined from the included studies, there is no specific training available in how to disclose a diagnosis of dementia. In two studies, HCPs reported receiving training in breaking bad news, attending conferences on the topic, or using general guidelines for diagnostic disclosure (Vince et al., 2017;Nielsen et al., 2018). Despite disclosure often being performed by specialists, two studies indicated that PCPs should have training on disclosing sensitively (Foley et al., 2017) and integrating disclosure into their consultations (Villars et al., 2010).

Discussion
This review provides the only up-to-date comprehensive account, based on a rigorous systematic literature search, of common practice around the world in disclosing a diagnosis of dementia from the perspectives of people with dementia, carers, and HCPs. These findings suggest that telling the person with dementia the diagnosis is now common practice. However, HCPs may experience discomfort in revealing the diagnosis and consequently use euphemisms to describe the syndrome rather than explicit terminology. Generally, if diagnosis disclosure is provided as a one-off event, and without discussion of hopeful aspects or emotional rapport building, it has a negative impact on people with dementia and carers. Carers report needing to know more about the future and prognosis, and their preference is for the diagnosis to be disclosed in a realistic yet hopeful way. HCPs would benefit from guidance and support in how to disclose the diagnosis, and the disclosure should be approached as a process involving a multidisciplinary team rather than as a singular event. The findings suggest progress in normalizing the diagnosis and reducing stigma but highlight that there is still more to be done.
Disclosure was viewed as a negative process by HCPs who reported an attitude of hopelessness, and uncertainty around the exact diagnosis, prognosis, and how much people receiving the diagnosis already knows or can understand. HCPs highlighted that the information needed by people with dementia and carers is difficult to deliver due to the delicate balance between honesty and hope (Bailey et al., 2016). HCPs do not appear comfortable with their position as someone who delivers bad news and can only provide limited treatment options. This review also identified that little consideration was given to patient-centered communication and expectations extending to cultural differences.
Our findings suggest that disclosures can cause people with dementia to suffer emotionally (Zou et al., 2017). Similarly, emotional rapport building is identified as important yet may not be frequently used in disclosures (Zaleta and Carpenter, 2010), suggesting that HCPs are not prioritizing the emotional impact of receiving a diagnosis as a key consequence of the meeting. While disclosure tended to have a negative impact for people with dementia and their carers, carers did report that disclosure was a useful event that enabled them to organize treatment and support, and discussion of etiology, progression, and causes was helpful. However, carers expressed a need to know more about the future and prognosis following diagnosis. This finding shows consistency with a previous review (Werner et al., 2013), but the present review shows a step forward in the dialog around truth telling, in that simply knowing the diagnosis itself is not enough. Carers need to know what will happen to the person with dementia in the future, beyond the diagnosis, and what ramifications this will have for themselves as carers.
Consistent with earlier reviews (Lecouturier et al., 2008;Werner et al., 2013) is the preference for a process approach to disclosure. This is supported by guidance from the UK Alzheimer's Society (Alzheimer's Society, 2018) and the British Psychological Society (Watts et al., 2018) which suggest that delivering information in one session is overwhelming and instead disclosure should be embedded within the care pathway. People with dementia and carers need more time and space to process information provided in the disclosure, which could be achieved using follow-up appointments. Using follow-up appointments could aid with retaining information, as in the general population 34-88% of what is said in medical appointments is forgotten, depending on the type of information provided, where general recall is higher for a discussion, but lower for specific information such as regarding lifestyle interventions (Richard et al., 2017). This is important, given the range of topics covered in a diagnosis disclosure, and it is likely that this figure is higher for people experiencing cognitive changes. More opportunities to impart the same information, or presenting smaller amounts of information more than once, could help people remember the diagnosis and its implications, while providing space to explore the emotional impact of the diagnosis. However, it must be noted that increasing the number of appointments would require greater resources, without which an increase in waiting times is likely and this is not desirable either. Involving other agencies, such as third sector agencies, could be useful.

Strengths and limitations
This review used a comprehensive search strategy across several databases to identify as many relevant papers as possible, although some papers indexed in other databases may have been missed. The rigor of the systematic search and quality review can be seen in that searching was completed by two reviewers, allowing for discussion and justification for inclusion or exclusion of each paper, and a sample of papers were assessed for quality by a third, independent reviewer which resulted in a high level of agreement. Quality assessment was not used in previous reviews of this topic area (Werner et al., 2013), and showing how this review has moved the discussion forward by highlighting methodological issues that must be considered in future work. Despite studies being of a variety of designs and conducted in different settings, the findings are generally consistent, suggesting that this review has identified realistic trends. The perspectives of people with dementia, carers, and HCPs were captured in this review, ensuring that findings are relevant to the experience of all people involved in the disclosure, and any implications can consider each set of needs.
Only studies written in English were included, and therefore some findings may have been missed. However, studies from across the world were included and despite some contradictory findings, generally there was agreement, suggesting that studies in languages other than English may not have significantly changed the results if included. Studies of all designs were included and, although this renders it impossible to synthesize numerical findings into meaningful quantitative analyses, the integration of qualitative and quantitative findings should be considered a strength. One methodological consideration to acknowledge is that much of the research included in this review is reliant on self-report techniques, either through use of surveys or qualitative interviews. While surveys can reach a broad audience, and interviews can elicit rich details about the experience of disclosure, both methods of data collection can be affected by bias, and it is challenging to reliably compare or contrast practices reported by HCPs with the impact they have on recipients (Plejert et al., 2017). Only 8 of the 52 included studies used observational methods (Hillman, 2017;McCabe et al., 2019;Peel, 2015;Sakai and Carpenter, 2011;Wynn and Carpenter, 2017;Xanthopoulou et al., 2019;Zaleta and Carpenter, 2010;Zaleta et al., 2012).

R E C O M M E N D A T I O N S F O R C L I N I C A L P R A C T I C E
Communication skills do not reliably improve with experience alone (Cantwell and Ramirez, 1997) and consequently further training, supervision, and regular reflection on practice may support HCPs in navigating uncertainties and managing their own emotional journey. For HCPs in specialist roles, training should be specific to disclosing a diagnosis of dementia rather than generally about breaking bad news, and supervision should cover managing the emotions of the person with dementia and the carer, as well as their own emotional responses. Communication skills training is mandatory in the majority of training programs for all HCPs (Moore et al., 2018), but training to develop skills in shared communication, to raise awareness of supporting people with dementia and carers to assert themselves and maintain their agency and power during the disclosure, is needed. HCPs may benefit from guidance in understanding and supporting cultural differences and sensitivities in diagnosis disclosure. However, it is recognized that training is resourceintensive and relies upon having appropriately trained individuals who are available to train and support others. This is likely to be difficult across disciplines and in time-pressured healthcare environments.
Pre-diagnostic counseling prior to disclosure could explore beliefs and expectations and potentially mitigate negative impacts by ensuring people with dementia and carers are prepared. This may support people with dementia and carers to consider questions they could ask between meetings and provide greater opportunity to speak during disclosure meetings. Follow-up meetings could enable tailoring of information, an opportunity to reconfirm the diagnosis, and discussions of prognosis, future considerations, and access to local support. Involving multidisciplinary teams could help to provide an holistic and hopeful disclosure, reducing the negative emotional burden on individual HCPs while also reducing the likelihood that the disclosure is viewed as a singular negative event by people with dementia and carers (Vince et al., 2017). Reassurance and a realistic sense of hope about the future should be emphasized throughout, and HCPs could discuss how people with dementia and their carers can maintain quality of life after diagnosis.

R E C O M M E N D A T I O N S F O R P E O P L E W I T H D E M E N T I A
A N D C A R E R S Information about memory assessments and what typically happens could be provided through channels including posters or leaflets in GP surgeries and community spaces. Information about what to expect when receiving the results of assessments may be provided ahead of disclosure meetings, although this could be distressing if information were given without having someone available to provide support or further explanations. People with dementia should be supported to ensure they can have someone attend the disclosure meeting with them if they wish. Encouraging people with dementia to arrange a follow-up appointment with their GP could be beneficial for asking further questions and considering the future implications.

R E C O M M E N D A T I O N S F O R R E S E A R C H
Much of the research included in this review relies on recalling the disclosure experience, and there is disparity between viewpoints of people with dementia, carers, and HCPs. A sensible next step is to conduct research that directly observes disclosure meetings, considering all individuals involved. While work in this area has already been undertaken involving people with dementia, carers, and HCPs (Peel, 2015), further work currently underway focuses on doctors only (Bailey et al., 2019;Xanthopoulou et al., 2019). More research that considers the three-way dialog between people with dementia, carers, and HCPs, analyzing what each brings to the disclosure and what each takes away, is needed. Most research to date is from the perspective of carers, with some from the perspective of HCPs. More research considering perspectives of people with dementia on diagnosis disclosure is needed to fully understand their wants and needs. Research combining all three perspectives is required to develop a realistic understanding of how disclosures can be best conducted to benefit everyone involved, more closely reflecting real practice.
Researchers must respect the agency and rights of people with dementia to participate in research. Earlier reviews (Werner et al., 2013;Lecouturier et al., 2008) have not explored power dynamics between individuals involved in disclosures, or the agency that people with dementia possess in the interaction. Only two studies in this review specifically considered attitudes toward diagnosis disclosure from the perspective of the person with dementia, although other studies have explored difference facets of the disclosure process. Less than half of the studies in this review recruited people with dementia, and of those that did, the number of participants with dementia was small. Developing an understanding of disclosure without involving people with dementia is inconsistent with patient-centered care. The lack of information regarding cultural differences in disclosure clearly shows that future research must explore this area to develop practices that support a culturally sensitive disclosure process.
Future qualitative studies could demonstrate greater methodological quality by stating the underpinning philosophical perspective adopted by researchers during data collection and analysis, including reflexive statements to locate researchers culturally and theoretically. Increased awareness by journal editors and peer reviewers regarding the reporting of qualitative methodology would encourage researchers to be transparent in their theoretical approaches.

Conclusions
Disclosing a diagnosis of dementia is a key process for people with dementia, carers, and HCPs. This review indicates that while changes in some aspects of disclosure have occurred over the last decade, disclosing a diagnosis of dementia remains a difficult and complex process, for which formal training and guidance is lacking. Receiving a diagnosis is generally a negative process for people with dementia, carers, and HCPs and leaves carers in particular feeling uncertain over the prognosis and future of the person they care for. Pre-diagnostic counseling and follow-up appointments could enable realistic and hopeful discussions of the implications of receiving a diagnosis of dementia, while reducing emotional burden on HCPs. This review highlights a need for more objective evidence that considers the perspectives of all individuals involved.

Conflict of interests
None.

Author contributions
JY and MS conducted the systematic review and the quality analyses of the included articles. JY wrote the article and MS, RS and LC all contributed to writing the article. All authors designed the review.
Dementia diagnosis disclosure: A systematic review 1189