Chapter 2. Identifying people with dementia
This chapter reviews what countries are currently doing to better identify people living with dementia. Many people with dementia go too long without receiving a diagnosis, and when they are diagnosed, the information they receive – about the disease, and about the services available to help them – is often inadequate. In many cases, primary care has assumed an increasingly important role in diagnosis and care for dementia. Specialist services remain an important part of the diagnostic pathway in many countries, with memory clinics increasingly common in many OECD countries. Access to timely diagnosis remains poor, but better training and well-designed incentives for primary care physicians, improved standardisation of specialised services, and strengthened data linkages between health settings, can help ensure more people receive a diagnosis when they need it.
The statistical data for Israel are supplied by and under the responsibility of the relevant Israeli authorities. The use of such data by the OECD is without prejudice to the status of the Golan Heights, East Jerusalem and Israeli settlements in the West Bank under the terms of international law.
2.1. Access to timely diagnosis remains poor
Although there is no cure or disease modifying treatment for dementia, the outcomes that people with dementia and their families experience can be significantly improved by the right health and social care services. A diagnosis allows health professionals to direct people living with dementia and their families towards the services that will be most beneficial to them. Those with early dementia may be offered medications that improve symptoms, even if only temporarily. Psychosocial interventions such as cognitive stimulation therapy can help to improve cognition. Families and carers can be trained to provide better care for people with dementia and manage the impact of caring on themselves more effectively.
Developing and being diagnosed with dementia can be distressing for people with dementia and their families, so it is important that diagnoses are accurately detected and effectively communicated. But diagnosing dementia can be difficult in part because symptoms are hard to differentiate from other conditions such as depression (Robinson et al., 2015), and it is not always done well.
A formal diagnosis of dementia (see Box 2.1) is also required for access to certain services in some countries. Anti-dementia drugs such as anti-cholinesterase inhibitors almost always require a formal diagnosis of dementia prior to prescription. In many countries, a formal diagnosis means that people living with the disease can be reimbursed fully or partially for a range of social and clinical support services, including medications and home care.
Despite these benefits, diagnosis rates remain low. Many countries reported that half or more of all people with dementia are thought to be undiagnosed, leaving thousands of people without access to the health and social care services that could improve their lives. Improving these rates is crucial to the wellbeing of people with dementia and their families.
A “diagnosis” can mean different things in different countries and in different settings. At the simplest level, a diagnosis of dementia means someone is told by a health professional that, in their opinion, they have dementia. In this report, we refer to this type of interaction as an indicative diagnosis.
Where a diagnosis of dementia gives someone access to certain health and social care services, or additional reimbursement, this may require certain specific tests to be carried out, specialists to be consulted, or for the diagnosis to be made in a specified setting. In this report, we refer to a diagnosis that meets the criteria for accessing all relevant support in a certain country as a formal diagnosis. In some countries, there are no specified criteria for accessing support, so indicative and formal diagnoses would be the same.
2.1.1. Primary care doctors are central to diagnosis, but do not always have the skills and resources required to succeed in this role
Primary care is a central part of the dementia diagnostic process in most countries
Primary care doctors are the first point of contact for people with dementia in most countries. In nearly all OECD countries, people who are concerned that they might have dementia go first to their primary care doctor. Of 29 OECD countries reporting, 26 reported that primary care doctors are always or sometimes the first point of contact for dementia (see Table 2.1). The role of primary care in dementia detection is often formalised or highlighted in national dementia plans and strategies, including those of Australia, Chile, Denmark, Ireland, and Mexico, and is often linked with goals to promote timelier diagnosis.
In a small number of countries, people who think they might have dementia can bypass primary care and go directly to specialists. In Estonia, for example, primary care practitioners serve as only partial gatekeepers: while referrals are needed for consultations with neurologists, psychiatrists – who also diagnose dementia – can be consulted directly. Primary care practitioners similarly play only a small role in dementia diagnosis in Greece, with most patients bypassing the primary care system in favour of going directly to a neurologist or other specialist. In Germany, people can also directly access a specialist when symptoms of dementia are present. The Japanese health system does not register general practitioners in a separate category from specialists; insured patients are allowed to go directly to any physician (Nakanishi and Nakashima, 2014; OECD, 2015a).
In nearly all countries, primary care physicians perform at least a basic cognitive assessment and provide people with an indicative diagnosis. In the United Kingdom, for example, people with memory complaints will most commonly go first to their primary care physician. After conducting a simple memory test and basic blood tests to exclude other possible causes of cognitive problems, the general practitioner will in most cases refer the patient on to specialist care, such as a memory clinic or memory assessment service.
The actual role of the primary care physician in dementia care can vary from cursory screening and referral to a more comprehensive examination and care management. In nine countries, primary care doctors also provide a formal diagnosis (Table 2.1). In Chile, for example, a shortage of specialists in geriatrics and neurology has led to primary care being the focal point for all diagnostic services (as well as ongoing dementia management). However, in most countries the initial tests conducted by primary care physicians serve to rule out alternative explanations and confirm suspected cases of cognitive impairment for onward referral. People with suspected dementia are then referred to specialist services for further tests before a formal diagnosis is made.
Primary care doctors use a range of standardised tools and tests to assess people with suspected dementia (Box 2.2). The most common is the Mini Mental State Examination (MMSE), with the clock test and the Montreal Cognitive Assessment (MOCA), a screening tool for mild cognitive impairment also frequently administered at the primary care level. In some countries (Belgium, Denmark, and Sweden) blood tests are carried out in primary care to rule out other conditions that may mimic symptoms of dementia-related cognitive impairment. While more advanced testing and neuroimaging is most frequently administered by specialists, primary care physicians in Sweden and Norway have recently begun conducting CT scans as part of their initial screening process.
Primary care doctors are not always able to provide effective diagnostic services
In some cases primary care doctors lack the skills and experience to provide effective diagnostic services. For example, a survey of general practitioners in Denmark suggested that less than half of primary care practitioners were able to perform a good assessment of dementia (Waldorff and Moller, 2001). Other studies have found that primary care doctors correctly identify only around 75% of dementia cases, and often fail to record the diagnoses that they do make. The performance of primary care is worse with more complex cases, such as early-stage cognitive impairment (Mitchell et al., 2011).
Primary care doctors may also be reluctant to take on the central role that OECD health systems expect them to play in dementia diagnosis, because of a lack of confidence or a lack of capacity. An evaluation of diagnosis in primary care in the United Kingdom identified lack of confidence as a significant obstacle to diagnosing dementia in primary care (Dodd et al., 2014); while in Slovenia, a shortage of general practitioners has led to high patient loads and a reluctance to take on additional cases of dementia. There is also evidence that where clinicians feel little can be done to help people living with dementia, such as if they are unaware of existing treatments or local resources and services for people with dementia and their carers, they may be more reluctant to diagnose or communicate a diagnosis in the first place (Bradford et al., 2009).
As the role of general practitioners in assessment and diagnosis of dementia has grown, a range of cognitive assessment tools have been developed to assist physicians in determining whether their patients have become cognitively impaired. Some assessment tools, such as the general practitioner assessment of cognition, have been developed specifically to be used in primary care settings. The most widely used cognitive impairment screening tools share a number of common characteristics: they are short (intended to be used within a regular doctor’s visit), easy to administer, and easy to score. The design of different screening tools nevertheless introduces different biases into the assessment, including related to level of education, cultural background, and how symptoms of dementia or cognitive impairment manifest in different people.
Mini mental state exam (MMSE)
The MMSE is a 30-point, 11-question assessment that tests cognitive function in the areas of orientation, registration, attention and calculation, recall, language, and visual construction. Originally proposed in 1975, it is the most frequently used cognitive assessment tool for dementia. While easy to administer, it is judged to be less effective for detecting mild cognitive impairment, while its structure – notably its use of verbal responses and written questions – may contribute to performance biases based on education level and language fluency. Unlike most cognitive assessment tools, the MMSE is copyrighted, somewhat limiting its use. Because of its copyright, the MMSE is somewhat less frequently recommended as an assessment tool in clinical guidelines.
General practitioner assessment of cognition
The general practitioner assessment of cognition was explicitly developed for use in primary care. It can be completed in less than five minutes and includes both a cognitive assessment of the patient and, in certain cases, a questionnaire for a family member or carer.
Montreal cognitive exam
The Montreal Cognitive Exam (MoCA) is a 10-minute, 30-point exam designed to screen for cognitive impairment. Scores from MoCA and the MMSE have found to be similar, particularly for people with dementia, though a number of studies indicate that MoCA may be more sensitive in detecting cases of earlier (mild) cognitive impairment (Nasreddine et al., 2005, Zadikoff et al., 2008, Dong et al., 2010, Trzepacz et al., 2015).
Memory impairment screen
The memory impairment screen is a four-minute delayed and cued recall test to evaluate memory impairment. Its short length and straightforward scoring make it easy to administer and evaluate. It is considered to be as sensitive as the MMSE in detecting cognitive impairment, though it does not assess cognitive impairment in the domains of visuospacial ability or executive function.
Mini cognitive assessment instrument
The mini-cognitive assessment (“mini-cog”) is a three-minute, three-part test that incorporates recall and delayed recall exercises with a clock test drawing. Its simplicity and brevity may be advantageous compared with longer assessment tools, such as the MMSE, given the short appointment times primary care practitioners often have with their patients. The mini-cog has been found to be as sensitive as the MMSE and other assessment tools in detecting cognitive impairment (Borson et al., 2003).
2.1.2. Guidelines are important to improving diagnosis in primary care
Many countries have developed clinical guidelines to improve dementia diagnosis in primary care
A number of countries have taken steps to promote high quality dementia diagnosis and care through the development of guidelines. Guidelines frequently include recommended assessment tools for diagnosis, principles of communation and care, and in some cases recommended pathways for care following referral or diagnosis. At least 16 OECD countries have guidelines for dementia diagnosis and care (see Table 2.2).
In some cases, guidelines have been developed specifically for, or tailored to, the primary care setting, as in NHS England’s Dementia diagnosis and management: a brief pragmatic resource for general practitioners. Similar primary care-specific guidelines are being developed in Belgium (by the University of Leuven) and Switzerland (by the Association of Swiss Memory Clinics).
Publicising guidelines and providing better training can help primary care doctors to comply with best practice
While guidelines can serve as a useful information resource for physicians, they are non-binding and may not always be followed. This may simply be because primary care doctors are not aware that guidelines exist or do not know where to find them. In many countries, guidelines are not developed by one single clinical body but may be produced by medical associations, nongovernmental bodies, or other associations. In the United States, for example, the Gerontological Society of America convened a Workgroup on Cognitive Impairment Detection and Earlier Diagnosis and developed a toolkit to help primary care physicians detect cognitive impairment and diagnose dementia (Gerontological Society of America, 2017). Publicising the existence of guidelines to primary care doctors may therefore help to improve compliance. However, there is also evidence that where doctors feel that they lack the skills to apply the recommendations found in guidelines, they may disregard them (de Lepeleire et al., 2008). This suggests that, to be effective, guidelines must be supported by training.
2.1.3. Adequate training for primary care doctors is essential for improving diagnosis
Across the OECD, primary care physicians have access to at least some dementia training, either during medical school or as part of their continuing education. Dementia training during undergraduate or graduate medical school remains limited. In most countries, primary care physicians are required to maintain their skills through a minimum number of continuing education hours per year. Training in dementia management is offered as an option to fulfil these requirements in many of these countries, including in Australia, Austria, Belgium, Hungary, the Netherlands, New Zealand, Sweden, and the United States.
Dementia training tends to be limited during medical school, though continuing education on dementia is available in most countries
Dementia training during undergraduate or graduate medical school remains limited, with many countries reporting that medical students receive very little training – often just one or two days – during their studies. Across 10 European countries, medical students average just 12 hours of dementia training during their studies (Tsolaki et al., 2010). In almost all countries, additional training on dementia is offered as part of continuing education programmes. . While training for dementia care is often offered as one of a suite of continuing education programmes available to primary care physicians, and can satisfy minimum continuing certification requirements, the choice of training topics are usually left to the clinician. Enrolling in dementia courses often requires clinicians to spend extra hours in training, which can result in lost income or cut into free time. Courses – particularly when offered by nongovernmental organisations – may also rely on external financing, including support from employers or municipalities. Where such reimbursement is not forthcoming, participation may suffer.
However, few incentives and barriers to participation mean take-up can be low and participation relies largely on the personal or professional interests of the physicians. Though they can help to increase participation rates, incentives for participating in dementia training are offered in very few OECD countries. The few incentives that are promoted focus primarily on either financial reimbursement or special certification rewards. Primary care physicians in Denmark do not have any continuing education requirements to practice, though training programmes are widely available. A reimbursement scheme has been developed that allows primary care practitioners, as well as specialists in private practice, to be reimbursed for the income they lose while undergoing training.
Developing nationally-available, dedicated dementia training courses may be another way to expand the reach of dementia training programmes and increase quality of care. In 2016, Australia launched a national training program providing dementia training to primary care providers, acute care, and aged care providers across specialities and levels of experience. Courses cover a range of issues, including diagnosis and care, medications management, and managing behavioural and psychiatric symptoms of dementia (BPSD). The training courses, which comprise a suite of options ranging from online short courses to tertiary-level qualifications, are administered by a consortium that includes five Australian universities, as well as the nongovernmental organisation Dementia Australia. In Denmark, the Danish Dementia Research Centre, the Danish Medical Association, and other providers offer both in-person and online training to primary health physicians and other health professionals.
Working with the Swedish Dementia Centre, the Karolinska Institute launched a Master’s in Dementia Care for Physicians in 2012, a web-based, part-time two-year program for physicians intended to improve dementia management for themselves and their clinics. The program requires all students to be funded by their employers. In Sweden, demand for the program from physicians has outstripped the willingness to pay by county councils, who are responsible for financing continuing education. The program has also accepted students from abroad, including from Japan and Germany.
Online training courses can reach more doctors more cheaply, but may not deliver the same outcomes
Online training programmes are increasingly used to increase access to information about dementia management that accommodates health professionals’ geographic or time restraints. In Mexico, the government recently launched a massive open online course (MOOC) focusing on dementia management. It has so far reached 5 000 health professionals, including primary care physicians, since its launch. Some of New Zealand’s continuing education training programmes are available later as online resources. The Danish Dementia Research Centre has identified e-learning as a key opportunity for training and knowledge dissemination (Tannebaek et al., 2014). More than 600 primary care practitioners – more than 15% of the country’s primary care physicians – have participated in an ABC e-learning module launched by the Research Centre in November 2016. When appropriately designed, e-learning modules may also help to reduce quality discrepancies between programmes, by allowing a large number of participants to benefit from the same, carefully designed curriculum.
Offering internet-based training courses can expand the reach of dementia education to a broader population of physicians and health professionals, though the quality of online courses and their impact on physician practice may not be as strong as more traditional training approaches. While even short-term training programmes may help increase physician’s confidence in their ability to care for people with dementia, there is some evidence to suggest that in-person training programmes and decision support interventions may be more effective than individual e-learning (Downs et al., 2006, Lathren et al., 2013).
Providing more intensive training to a small number of doctors may be a more realistic strategy to ensure access to expertise
Despite the increasing prevalence of dementia, not all primary care services will treat a high volume of cases Training all primary care physicians in dementia care management may not be the best use of limited resources if caseloads for many are low. And even where physicians see a large number of older patients – as is increasingly the case across OECD countries given rapid population aging – a policy that tries to increase the skill level of all doctors equally will only be able to cover basic dementia care training, and risks increasing skill level only marginally.
Investing resources to intensively upskill a limited number of primary care physicians may be a more effective approach. Physicians who express particular interest in dementia, or those who have a high caseload of older patients, could be targeted to receive additional training. In concert with more broadly available dementia training courses for primary care practitioners, a number of OECD countries have begun developing training programmes that aim to strengthen cooperation between primary care physicians and harness individual knowledge to improve dementia management across clinics. Together with an umbrella organisation of general practitioners, the University of Leuven in Belgium has begun developing a program that would identify and train specific primary care physicians to act as “reference” physicians for dementia for other practitioners handling fewer cases of dementia. In Mexico, a six-month training program targeting primary care professionals (including both physicians and nurses) takes place largely online. The program also includes an in-person training component with other participants, with the explicit purpose of developing a network of health professionals trained in dementia management.
2.1.4. Population screening for dementia is uncommon, but increasing
In an effort to promote early detection and increase diagnosis rates, countries have undertaken population screening programmes for a number of diseases, such as breast cancer and colon cancer. Population screening has proven to be a popular approach in OECD countries when detection rates for various diseases are suboptimal. Across the OECD, for example, more than three-fifths of women are screened for breast cancer (OECD, 2017). Unlike screening programmes for cancers, where the evidence relating to improved outcomes from early detection and treatment is more developed, the effectiveness of screening programmes for dementia is much less clear. Countries have largely refrained from or actively rejected implementing population screening programmes to detect dementia, despite persistently low diagnosis rates.
The effectiveness of screening programmes for dementia is unclear
Concerns about developing population screening programmes for dementia rest largely on the insufficient evidence of effect they have demonstrated, and the lack of cure for the disease. With no prospect for a cure, receiving a diagnosis before a person is ready to may lead to substantial emotional and psychological distress, with little additional benefit (Lecouteur et al., 2013, Turner, 2013, Mate et al., 2017). When people with dementia receive a diagnosis they are not expecting – which can particularly be the case for people with early and mild forms of the disease – the news can have a devastating effect.
Furthermore, there is no guarantee that people diagnosed through screening programmes will receive adequate follow-up care after diagnosis. Treatment options for dementia are extremely limited already, and there is some evidence suggesting that the follow-up action and care patients receive after a diagnosis through screening is low, particularly for younger patients and people with less than severe cognitive impairment (Boustani et al., 2005, Borson et al., 2007).
Screening programmes also run the risk of mistakenly identifying people without dementia as having the disease. General population screening programmes, even when restricted to older populations, have been found to have false-positive screening rates that may be as high as one in five among those initially screened positive for cognitive impairment. Because of the low prevalence of dementia in the general population, the positive predictive value of the screening assessments will also be lower in a general screening program than in more targeted settings (Boustani et al., 2005, Mate et al., 2017). The additional costs of screening, lack of preparedness of primary care physicians to undertake such screening programmes, and the additional time burden have further been identified as arguments against implementing broader screening programmes for dementia.
General population screening for dementia is uncommon, though some countries have recently introduced programmes
Most OECD countries have not implemented general population screening programmes for dementia. In a number of countries, including the United Kingdom and Australia, national clinical guidelines advise against the practice, while others, including the United States, have concluded that current evidence is insufficient to develop further recommendations (LeCouteur et al., 2013, U.S. Preventive Services Task Force, 2014, UK National Screening Committee, 2015, Guideline Adaptation Committee, 2016).
Nevertheless, four countries reported that they have developed cognitive screening programmes for older populations: Chile, Finland, South Korea, and the United States (Table 2.3). South Korea has implemented national population screening for dementia through the National Dementia Early Detection Program (NDEDP) since 2010. As part of its goal of strengthening its diagnostic system for dementia, Chile has similarly planned to integrate a clinical evaluation for dementia into its annual preventive health exam for older adults, the Examen de Medicina Preventiva del Adulto Mayor (EMPAM) (Chilean Ministry of Health, 2015). Three communities were selected to begin implementing the country’s dementia plan in primary care in 2017.
Cognitive screening was recently introduced to Medicare recipients in the United States. Under the Patient Protection and Affordable Care Act (ACA), Medicare recipients are entitled to an annual Wellness Visit, which includes an assessment to detect “any cognitive impairment” (Hughes et al 2011, Borson et al., 2013). Some concerns have been raised over the possibility of an increase in false positives as a result of this program; to help reduce this risk, clinicians have proposed a short screening indicator to help general practitioners identify patients at high-risk of cognitive impairment (Barnes et al., 2014).
Other countries have considered developing screening programmes, or have plans to do so. In the Czech Republic, a two-year pilot project has seen general practitioners include basic dementia screening, including blood tests and a mini-cog examination, in biennial preventative exams available to all patients above the age of 60. The Ministry of Health is also in the process of developing a special project focused on expanding preventative screening programmes beyond cancer, and is considering dementia screening as one possible option. Beginning in, 2019, screening for dementia will also be rolled out in primary care in Turkey, as part of its efforts toward the earlier detection of dementia.
Future medical advancements could change the benefit-risk calculation behind population screening for dementia
The balance of harms and benefits of population screening may change in the coming years. More recently, the question of whether countries should develop general population screening programmes has been revived by advancements in the understanding of the onset of dementia. After a number of promising anti-dementia medications, including Eli Lilly’s solanezumab and Merck’s verubecestat, failed during late-stage clinical trials, researchers have increasingly focused on developing treatments for dementia that must be administered years before any symptoms emerge. A number of ongoing clinical trials are testing this theory by administering anti-dementia medications in pre-symptomatic populations. Initial results will likely be published in 2019 and 2020. Should these interventions work, governments will be faced with new questions over how to identify the populations that would best benefit from these new disease-modifying treatments.
2.2. Specialist services continue to play an important role in diagnosing dementia
Even with the increasing focus placed on primary care management for dementia, formal diagnosis continues to take place at the specialist level, and often in memory clinics, in nearly all countries. The reasons for referral to specialists can vary, but can include primary physicians being unwilling to make a formal diagnosis, the need for a differential diagnosis, or insurance requirements that dementia be formally diagnosed by a specialist before available support can be accessed.
Specialists are needed to identify dementia sub-types and may often be more experienced in caring for people with dementia, including identifying the best available treatment options and local services available to a person with dementia and their family. Neurologists, psychiatrists, and geriatricians, are often needed to diagnose specific sub-types of dementia, including Alzheimer’s disease, dementia with Lewy Bodies, and vascular dementia. Receiving a differential (sub-type) dementia diagnosis is important to ensure the best treatment pathways can be adopted. Different dementia sub-types have been found to respond differently to existing dementia treatment options, including memantine and anti-cholinesterase inhibitors (Livingston et al., 2017).
2.2.1. The availability of specialists varies significantly across OECD countries
Following a visit with a primary care physician, people suspected of having dementia are regularly referred to a specialist – most frequently geriatricians, psychiatrists, and neurologists – for further testing and to confirm the diagnosis. However, the number of specialists ranges substantially across OECD countries, indicating access to specialist care is not always consistent. For psychiatrists and neurologists, physician density varies more than three-fold across the OECD, with just 17 per 100 000 in Portugal, compared with 57 per 100 000 in Switzerland (see Figure 2.1).
Source: OECD Health Statistics 2017, https://doi.org/10.1787/health-data-en ; Eurostat Database 2018.
To facilitate diagnosis, many countries have set up specialist memory clinics to help with diagnosis and care. As attention to dementia management has grown in OECD countries, the role of the memory clinic has taken an increasingly prominent position in national plans or strategies, and in some cases formalised diagnostic pathways. While specialists can also operate independently, in the past decades the memory clinic structure has emerged as an important component of diagnosis or care for people with dementia. Of the 31 OECD countries interviewed for this report, only six countries – the Czech Republic, Estonia, Hungary, the Slovak Republic, Slovenia, and Turkey – reported having no memory clinics in their country. Between 1993 and 2000, the number of memory clinics in the United Kingdom more than doubled, from 20 to 58 (Lindesay et al., 2002). As of 2017, there are approximately 200 memory clinics in England. Similarly, of eight memory clinics identified in New Zealand in 2008, six had been established within the previous five years (Cheung and Strachan, 2008).
Despite the rising profile of memory clinics in dementia care, what a memory clinic is has different interpretations in different settings. Originally memory clinics served as specialised clinical settings where health professionals – often neurologists, geriatricians, or psychiatrists – provided advanced diagnostic services.
In this report, we define memory clinics most broadly as any clinical setting where health professionals provide advanced diagnostic services and define themselves, or are defined externally (such as by insurance companies), as memory clinics. More specifically, we consider clinics that provide diagnostic services but no or only limited follow-up care to be diagnostic memory clinics. Memory clinics that are integrated into defined care pathways or otherwise engage in care management, education or training, and other components of dementia care are defined here as comprehensive memory clinics.
Few countries have developed clear guidelines or definitions around the staff profile, organisational structure, or location of memory clinics, or the services that they are expected to deliver. Consequently, most countries have a mix of diagnostic and comprehensive memory clinics. In some countries, the services provided by memory clinics may be case-dependent, or otherwise not clearly defined, in which case the broadest definition will be used.
Memory clinics play a major role in formal and differential diagnosis
Memory clinics operate as diagnostic centres for at least some cases of dementia in nearly all countries. A formal assessment at a memory clinic has become part of the recommended diagnostic pathway in some countries. Guidelines by the National Institute for Health and Care Excellence (NICE) in the United Kingdom identify memory assessment services as the “single point of referral for people with possible dementia” following a primary care assessment (NICE, 2006). Memory clinics are sometimes the only setting available for specialist consultation related to dementia. For example, Denmark has no geriatric specialists in private practice, while neurologists and psychiatrists in private practice almost never accept patients with dementia. While there are some neurologists and psychiatrists in private practice, they almost never take cases of dementia. As a result, people suspected of having dementia are nearly always referred to a memory clinic if follow-up specialist assessment or care is needed.
In some circumstances, memory clinics have been identified on the diagnostic pathway as important points of referral for more complex cases of dementia. Complex or rare cases of dementia can include dementia in younger persons, early stage dementia, people presenting with complex comorbidities, and people with minority language or cultural backgrounds. In Norway, people with suspected cases of dementia are recommended to be seen by their primary care physician and the municipal memory team. New guidelines further recommend that complex or unusual cases be referred to a memory clinic. In the Netherlands, where primary care is encouraged to participate in dementia care management to the greatest possible extent, both memory clinics and even more highly specialised university Alzheimer’s centres are available for diagnostic support, care, treatment, and research for people with complex or difficult cases.
Some countries have developed specific services for people with early onset dementia. People with early onset dementia – which is characterized by the development of dementia before the age of 65 – often face a different or additional set of challenges to those experienced by older people with the condition. Among other considerations, they and their carers may still be working. In France, a special national reference centre dedicated to people with early onset dementia has been established to address their unique needs and ensure they have access to clinical services and adapted support throughout their lives.
The characteristics, concentration and distribution of memory clinics differs between countries
In most cases, memory clinics in OECD countries are multi-disciplinary and led by specialists. A review of memory clinics by Jolley, Benbow and Grizzel (2006) identified a number of common specialist settings for memory clinics, including psychiatry, neurology, and geriatrics, with most clinics based in geriatric psychiatry services (Cheung and Strachan, 2008; Lindesay et al., 2002). One or more of these specialties is found in most memory clinics in the OECD. In Austria, memory clinics are typically led by a neurologist or psychiatrist, who works alongside a psychologist and nurse in making diagnoses and sometimes offer further post-diagnostic support to people with dementia and their families. Greek memory clinics are similarly structured, with the majority led by neurologists working in concert with psychologists and nurses. Memory clinics in Israel are most commonly run by neurologists or geriatricians and often supported by a multidisciplinary team that can include neuropsychologists, psychiatrists, nurses and social workers. In England, memory clinics are normally led by old age psychiatrists, with support from medics, nurses, psychologists, and neuropsychologists. Unlike in most other health care systems, neurologists play a very small role in diagnosing and care for dementia in Norway, where specialists more often come from geriatrics or psychiatry. In the eight memory clinics in New Zealand identified in a review by Cheung and Strachan (2008), psychogeriatricians, geriatricians, psychologists, and occupational therapists were found to act as lead clinicians. No memory clinics were led by neurologists.
Specialists most likely to diagnose and care for people with dementia can differ geographically. In the north of Sweden, for example, geriatricians typically lead memory clinics, while in the south, psychiatrists are more common. There is similar regional variation in New Zealand. In most cases, whether a person with dementia sees a neurologist, geriatrician, or psychiatrist depends less on their symptoms and more on the staff profile of nearby memory clinics.
As many as 25 out of 31 OECD countries interviewed for this report have developed memory clinics. Data from countries with longitudinal information indicate that clinics have also expanded rapidly in recent years. The number of memory clinics doubled in the United Kingdom between 1993 and 2000 and quadrupled in New Zealand between 2003 and 2008. In France, 28 resource and research memory centres for complex and atypical cases and 500 to 600 memory consultation centres provide diagnostic services for people suspected of having dementia. The number of people aged 65 and over per memory clinic varies widely across countries, from more than 170 000 people 65 and over per memory clinic in Belgium to about 23 000 per memory clinic in France (Table 2.4). While the concentration of memory clinics within countries likely also differs, this national variation also indicates that different countries rely to different extents on memory clinics as a part of the clinical pathway for dementia.
Few countries have developed guidelines and standards to help define the role of memory clinics
Almost no OECD countries have developed official definitions related to the operations of memory clinics. This has led to significant variation in the services offered by clinics both between and within countries. An important exception is France, which has developed guidelines outlining requirements for both its Centres Mémoire de Ressource et de Recherche (CMRR) and Centres Mémoire (CM). In the absence of national or regional guidelines, the comprehensiveness of care offered in a memory clinic often depends on the interests and capacities of the managing physician and their connection to nearby services. In Switzerland, the Swiss Memory Clinics association has had quality standards for admission into the association since 2011. A similar network of memory clinics was recently set up in the Netherlands. The Dutch Memory Clinic Network was launched in 2016 to promote knowledge sharing and improve care across memory clinics. As part of its work, standards for memory clinics are in the process of being developed. In the United Kingdom, the Memory Services National Accreditation Programme accredits memory services that meet national guidelines, including the NICE Quality Standard for Dementia (Royal College of Psychiatrists 2018).
Guidelines for memory clinics developed in Switzerland and Denmark can be seen as “minimum standards” that ensure clinics provide quality services that reflect national or globally developed standards, particularly for diagnosis. Both guidelines cover the minimum caseload expected of clinics, as well as the professional composition of memory clinic teams.
Swiss Memory Clinics Association
In order to be considered for membership, memory clinics must demonstrate that they conduct at least 100 diagnostic evaluations for dementia every year. Evaluations must be comprehensive and multi-modal, and diagnosis must be established with interdisciplinary consultation. Four professional disciplines must be available through the memory clinic: geriatrics, geriatric psychiatry, neurology, and neuropsychology. In addition, memory clinics must comply with national and international recommendations and engage in public awareness and network-building with the community and other related services. 12 process-oriented quality standards have been developed to ensure that memory clinics meet standards of membership.
Danish Memory Clinic Guidelines
Recently published guidelines from the National Health Board of Denmark are directed at the changes to existing memory clinics the five administrative regions of the country are expected to make to qualify for additional financing for memory clinics under the National Plan. The guidelines recommend that each memory clinic serve a population of at least 300 000 people and conduct a minimum of 400, but preferably at least 500, diagnostic evaluations each year. Memory clinics should have neurologists, psychiatrists, geriatricians, neuropsychologists, and nurses on staff. Add more once guidelines are out. The guidelines are partially based on the experiences developing memory clinics in the capitol region of Denmark.
2.2.2. Services at memory clinics can vary widely, though many have begun to take on roles beyond diagnosis
In recent years, diagnostic memory clinics have expanded beyond their initial role as a location for detailed diagnostic assessment, and increasingly provide post-diagnostic support services and care for people with dementia throughout their lives. In addition to ‘traditional’ diagnostic memory clinics, new models of clinic-based care have begun to emerge in many OECD countries.
After receiving a diagnosis of dementia, people living with dementia and their families and carers must adjust to what can be a crushing diagnosis. A range of resources – from health services, social services, and the nongovernmental sector –are often available to help them with daily living activities, emotional and social support, information and training, advanced care planning, and other needs. Yet in many cases, these available services are fragmented, and people with dementia may not receive adequate information about what these services are, their relevance, and how to access them. In the worst case, a person with dementia may be sent home with a diagnosis and no information related to what this means or what to do moving forward. To avoid this outcome, many countries have begun to pay attention to strengthening the co-ordination of care that occurs after a person has received a diagnosis. Post-diagnostic support refers to the kind of support – in terms of care, services, resources, and co-ordination – that a person with dementia receives in the aftermath of a diagnosis.
Some comprehensive memory clinics emphasise the role of other health professionals, including primary care physicians and nurses. Others do not identify as memory clinics but offer many of the same services, often with a greater emphasis on holistically caring for the person with dementia and their families. As interest in the role of memory clinics has increased, the services provided by some clinics have expanded beyond diagnosis, to increasingly include post-diagnostic treatment, case management, support for families of people with dementia, and education and training services for health professionals or the broader community. In England, for example, memory clinics work closely with the Alzheimer’s Society. Dementia support workers and dementia care advisors may be present at the memory clinic, or are otherwise available to signpost people with dementia and their families to post-diagnostic support.
Comprehensive clinic models have moved away from diagnostic services to focus more broadly on dementia care
In the past years, memory clinics have increasingly moved away from a focus on diagnostics to also play an active role in dementia care management, through a model that can be conceptualised more broadly as comprehensive memory clinics. One such model has seen memory clinics expand to the primary care level. In Ontario, Canada, a shortage of geriatricians has created serious capacity constraints and high waiting times for specialist consultation. Across Canada, there were just 261 registered geriatricians in 2015, a rate of 0.7 per 100 000 population, compared with nearly 3 000 paediatricians (Canadian Medical Association, 2015, Canadian Medical Association, 2016). As one response, physicians developed a model of primary care-based dementia detection and management (Lee et al., 2014). Part-time memory clinics were established within existing family health teams (FHT), a collaborative model that consists of family physicians, nurses, and social workers delivering primary care in local communities. Memory clinics are led by family physicians who have received additional training in diagnosing and caring for dementia, with the explicit aim of reducing reliance on geriatrics and other specialty services in not only care, but also diagnosis (Lee et al., 2011). Geriatricians or geriatric psychiatrists provide additional consultative support to the family physicians as needed. In an evaluation of 16 family health teams operating memory clinics, a review by geriatricians of diagnoses made by primary care practitioners found high agreement with their evaluations and subsequent care management plans (Lee et al., 2014). Just 8.9% of patients were referred to a specialist, compared with referral rates of over 80% among regular primary care physicians (Lee et al., 2014).
A similar approach to primary care-led memory clinics has been taken in the England. Beginning in 2006, the Gnosall Primary Care Clinic in Staffordshire has run a monthly memory clinic in its primary care facilities with an aim to increase diagnosis rates and reduce waiting times between the identification of symptoms and diagnosis (Benbow et al., 2013). As in Ontario, the memory clinic operates part-time and is located within a broader primary care clinic. Unlike the Ontario model, where the primary care physicians lead diagnosis, the Gnosall Primary Care Memory Clinic incorporates a geriatric psychiatrist into the practice staff at the memory clinic. Following the identification of suspected cases of dementia by the general practitioner, consultant geriatric psychiatrists review health care records and see the patient and their family members during the memory clinic visit. By situating the memory clinic within the existing primary care practice, proponents of the model argue that the consultant specialists can more thoroughly review the patient’s medical history before the consultation (Benbow et al., 2013). In addition, the familiar location of the memory clinic is seen as advantageous in facilitating a lower-stress environment in which to receive the diagnosis (Benbow et al., 2013). The Gnosall Primary Care Memory Clinic has further developed the role of an ‘eldercare facilitator’ to help coordinate between the staff of the memory clinic and external agencies and services (Greaves et al., 2015). A review of the Gnosall model found that, compared with similar practices in the region, the primary care-based memory clinic resulted in higher satisfaction among people with their dementia and their families, lower utilisation of other health care services, and substantial cost-savings for the health system (Clark et al., 2013).
It is possible that primary care-led memory clinics are most appropriate where staffing shortages or other capacity constraints lead to long waiting times or otherwise hamper access to specialist services, including specialist memory clinics. In Ontario, at least one family health team closed their memory clinic due to the availability of easily accessible specialist services in their area (Lee et al., 2014). The occasional nature of the memory clinics, which largely operate once or twice per month, and need for minimal additional investments in staff or supplies suggests this model may be well suited to address these resource constraints with minimal disruption to the regular operations of the health system.
In Israel, people with dementia frequently receive care at geriatric clinics, rather than diagnostic memory clinics. Memory clinics are normally specialist-led and primarily located within acute hospitals and other tertiary care facilities. These largely diagnostic memory clinics are outside the public health insurance system, and require pre-approval by insurance funds for visits to be reimbursed. For this reason, most primary care physicians do not refer suspected cases of dementia to diagnostic memory clinics, but rather to in-network geriatric clinics. Often located in the same physical building as the primary care clinic, geriatric clinics are responsible for a range of elderly care services and do not specialise in memory services. Their primary staff includes geriatricians but not psychiatrists or neurologists. While they can conduct diagnostic evaluations for memory disorders, they are less likely to have access to the latest imaging technology and may be less familiar with dementia diagnoses than specialists at dedicated memory clinics. However, as in primary care memory clinics in Canada and the UK, their location in the same building as primary care facilities can help with patient comfort and promotes communication between the person with dementia’s primary care physician and the geriatric clinic staff. In an evaluation of geriatric clinics, up to 70% of the patients were found to have been diagnosed with some form of cognitive impairment.
As an alternative to the comprehensive memory clinic, health professionals in a number of countries have developed care clinics that provide post- and sometimes pre-diagnostic support, with the actual diagnostic evaluation conducted elsewhere. In a way, these models build on the expanded notion of a comprehensive memory clinic, by reflecting the understanding that people with dementia may need integrated services that offer signposting and care beyond the clinical diagnosis. In Austria, a model of care management support known as Dementia Service Centres have been developed in the region of Upper Austria (Auer et al., 2015). The centres have three explicit goals: to promote early but timely detection of dementia, to support caregivers, and to delay premature institutionalisation. Recognising that memory clinics and other institutions may have a high physical and psychological barriers for entry, Dementia Service Centres are integrated into the community, encouraging people who may have dementia to be tested at their own pace, and helping them before, during and after the diagnostic process. There are currently eight centres in two regions of the country. The centres respond to a recognised need in Austria, as reflected in the Austrian Dementia Strategy to develop low-threshold centres that support people with dementia and their carers. Initially funded by Austria’s national insurance company, the centres are now financed publicly, and will be further rolled out in Upper Austria and possibly other areas of the country in the coming years.
A similar model has been developed in Belgium, where memory rehabilitation centres provide support to people with dementia and their carers, without providing diagnostic services. There are currently 12 memory rehabilitation clinics throughout the country, with at least one located in each province. Clinics are intended to facilitate living in the community longer and offer occupational therapy, home visits, psychoeducation, and support for families. The services of memory rehabilitation centres can be accessed once a person with dementia has received a formal diagnosis of dementia, including from a diagnostic memory clinic. Up to 20 visits to a memory rehabilitation centre are reimbursed over a lifetime.
Memory clinics may be a more appropriate setting for providing care to people with complex cases of dementia, though their cost effectiveness should be considered
The role of memory clinics in providing care beyond diagnosis has become particularly pronounced for complex or uncommon cases of dementia. In some cases, this expanded role has been formalised into guidelines or care pathways for dementia. In Denmark, the extent to which memory clinics are involved in follow-up care depends largely on the case management programmes of each administrative region. In the capital region around Copenhagen, memory clinics are responsible for diagnosis, treatment and follow-up in the first months following diagnosis, after which care management is returned to the primary care physician. Capacity constraints limit the ability of memory clinics to provide longer-term care for all people with dementia. In certain cases, however, including where people with dementia also have special needs or where cases are particularly complex, such as where patients have frontotemporal dementia, memory clinics are expected to offer lifetime follow up care.
The Healthy Aging Brain Center in the US state of Indiana is a memory clinic located in the public health care system ‘Wishard Health Services’ in Marion County, Indiana. The health system serves a primarily minority, low-income population of about 750 000 in and around the city of Indianapolis. The Healthy Aging Brain Center was launched to improve existing standards of care around diagnosis and management of people with dementia through a model that, while delivered through the Centre, also strengthens the role of primary care physicians, informal carers, and people with dementia. The staff of the Healthy Aging Brain Center includes two geriatricians, two dementia co-ordinators, and one medical assistant (Boustani et al., 2011).
As a diagnostic memory clinic, the Healthy Aging Brain Center (HABC) accepts referrals from general practitioners and specialists, as well as referrals from family members or caregivers and self-referrals. Prior to an initial in-person assessment, a person with possible cognitive impairment and their informal carer will completed a structured needs assessment by phone or mail. If the in-person diagnostic evaluation indicates a diagnosis of dementia, staff at HABC develops a tailored care management plan, the details of which are discussed with the person with dementia and their informal carer at a second in-person meeting. Components of the care plan include pharmacological and non-pharmacological interventions, support and counselling services for the person with dementia and their carers, participation in support groups, telephone support, and the opportunity for co-management in complex cases. In addition, the HABC staff engages in case-finding for other possible comorbidities and behavioural challenges, including depression and psychosis, and actively assesses both the physical and emotional well-being of informal caregivers to monitor carer stress and intervene as needed. Following the initial diagnostic and needs assessment, people with dementia and their carers are followed up by telephone and in-person visits. In-person visits can range from once per month to once per year, based on need.
Compared with people with dementia who received support through primary care centres, people with dementia who accessed the Healthy Aging Brain Center were found to have had fewer and shorter rates of hospitalisation, with lower rates of readmittance and higher health quality outcomes, including better management of high cholesterol (Boustani et al., 2011). The net cost savings from the HABC model have been estimated at up to USD 2 856 per patient, suggesting that substantial cost savings could be generated for health systems if such integrated care models were adopted (French et al., 2014).
Special provisions for complex cases are also made in Norway, where memory clinics are not included in the recommended typical diagnostic pathway. New guidelines released in 2017 advise that in complex and uncommon cases, including cases of early onset dementia, mild cognitive impairment, dementia in people with different language or cultural backgrounds, and dementia accompanied by serious comorbidities, care should be handled by memory clinics. For less complicated cases of dementia, primary care health services, including the general practitioner and municipal memory teams are considered to be sufficient for diagnosis and care. Complex cases are similarly managed by memory clinics in the Netherlands, while most people with less complicated diagnoses of dementia will be referred back to their general practitioner for care following a memory clinic assessment.
While a small number of countries have clearly articulated the expected services offered by memory clinics, in most countries the expansion of services beyond diagnosis has developed in a more ad-hoc manner. A review of memory clinics in Ireland found that of the 14 memory clinics in operation in 2013, 12 specialised in diagnosis, one offered both pre- and post-diagnostic support, one provided post-diagnostic support, and two specialised in caring for people with intellectual disabilities (Cahill et al., 2013). The range of services offered is similarly varied in Austria, Israel, Portugal, and the United States, which lack standards or definitions for memory clinics. In Greece, memory clinics provide diagnostic services and clinical follow-up but do not offer non-pharmacological treatment options or support for families and carers. However, half of memory clinics are located in day care centres associated with the Alzheimer’s Association of Greece, ensuring proximity to such services. As in Greece, many memory clinics in Portugal are associated with day care centres and sometimes offer non-pharmacological treatments, including cognitive therapy.
Despite growing interest in memory clinics, their effectiveness – and particularly cost-effectiveness – for diagnosis and care is not clear. A recent study of memory assessment services in the United Kingdom found that while the care provided through memory clinics was effective, it was not cost effective (Gomes et al., 2017). Moreover, whether memory clinics provide more cost-effective post-diagnostic care than primary care settings is far from established. Recent evidence suggests that in many cases, primary care can offer equivalent care for much lower cost. A randomised control trial of follow-up care through memory clinics in the Netherlands found that care provided through memory clinics was no more effective than post-diagnostic support provided through general practitioners (Meeuwsen et al., 2012). There is also evidence that the demographic profile of memory clinic users has changed over time, with memory clinics increasingly seeing people with higher levels of cognitive function (Azam et al., 2016). More evidence is needed to evaluate whether memory clinics are equally effective for people with higher cognitive function, or whether they should be reserved for people with serious cognitive difficulties. Given the cost pressures faced by many health and social care systems, countries should carefully consider whether memory clinics should always be included in the clinical pathway, particularly for post-diagnostic support.
2.3. Efforts must be strengthened to monitor diagnosis rates and improve measurement
In the past years, as dementia has risen on the international policy agenda, many OECD countries have developed or updated their dementia strategies, and there is broad interest in improving the quality of care that people with dementia receive. However, a lack of data on the quality of dementia care continues to hold back improvement efforts. Improving data for dementia is recognised as a key policy in a number of national action and strategic plans, including in Ireland, Switzerland, the UK, and the United States. Other countries have made improving dementia research – not only clinical research, but also population health – a key component of their national plans, including in Australia, Finland, and Mexico. Yet despite these signals, few countries have data systems that can generate the kind of information critical to developing evidence-based policies and measuring progress toward defined goals. Data challenges have further constrained the ability to engage in international comparisons in addition to hampering effective domestic policy development.
In response to this, the OECD recently undertook work to develop a small number of internationally comparable measures of the quality of dementia care. As part of this project, 21 OECD countries participated in a scoping survey aimed at identifying the most promising areas for developing indicators. 14 also participated in a feasibility study. The scoping survey and feasibility study provided important information related to data availability for dementia in OECD countries. An initial set of six pilot indicators was launched in January 2017, with the intention of providing information related to the comparative quality of dementia care across OECD countries.
2.3.1. Few countries have basic data on dementia diagnosis
Diagnosing dementia is critical to responding to dementia and remains a policy priority for many OECD countries. Understanding the success of diagnostic strategies requires an estimate of both the number of people diagnosed with dementia and the total prevalence in the population. Out of 21 reporting countries, 15 can estimate total prevalence at a national level. For the most part, country-level prevalence estimates appear to match estimates in the international literature. More than 70% of countries that submitted prevalence data as part of the 2017 OECD pilot dementia data collection reported prevalence rates that are close to a recent international review of prevalence estimates, which estimated regional prevalence rates based on published prevalence studies (ADI, 2015).
While most countries can estimate dementia prevalence, fewer than 40% of countries can estimate rates of diagnosis nationally. Given the policy priority improving diagnosis rates has received in many countries, establishing a baseline from which to improve is critical. A number of countries, including the United Kingdom (England) and Denmark, have set specific targets around improving diagnosis rates. As part of the Prime Minister’s Challenge on Dementia, a target was set to increase dementia diagnosis rates from 42% to two-thirds of people living with dementia in England. In Denmark, the National Action Plan for Dementia has set a goal to increase diagnostic assessment among people living with dementia, including securing a specific diagnosis for 80% of those assessed.
Identifying diagnosed cases of dementia is important to understanding how people with dementia navigate and use health and social care services. Without a dementia diagnosis, people with dementia cannot be tracked through the health system, and any differences in the quality or quantity of care that they might receive cannot be measured. For example, there is compelling evidence indicating that stays in hospital for people with dementia can negatively impact their well-being (Fong et al., 2012; Fick et al., 2013). Reducing both hospital admissions and length of stay in hospital are therefore important policy goals in improving the quality of life of people with dementia. A number of policies in OECD countries have been developed with these outcomes in mind. Without dementia diagnoses recorded consistently in medical records, however, measuring progress and policy effectiveness is not possible. At the moment, less than half of OECD countries can measure the utilisation of health and social care services for people with dementia at a national level (see Figure 2.2).
Source: OECD Dementia Care Survey and Interviews 2016-17.
Health data in OECD countries most regularly comes from administrative data (such as insurance datasets), population surveys, patient surveys, clinical records, and disease registries (OECD, 2015b). Many more countries can identify people with dementia in administrative datasets than can do so in registries or surveys. This is in part driven by the fact that administrative datasets are available in all countries. This data is collected to help with the administration of the system – for example, tracking the number and type of patients to determine payments from health insurers to providers – rather than explicitly for research or to measure quality. However, it is possible to derive some information about the quality of care from this type of data. Administrative data also has the advantage that it usually covers all people who interact with a certain part of the system. The most common care setting in which countries use administrative data to identify people with dementia is hospitals. Nearly 80% of surveyed OECD countries can identify people with dementia using national-level administrative hospital data. Data from other care settings is less widely available. Fewer than half of the countries surveyed can identify people with dementia in administrative data from primary care, long-term care (LTC) facilities, LTC provided at home, community health care, disability services, diagnostic services and palliative care.
2.3.2. Effective monitoring of diagnosis rates will require better primary care data or the development of dementia registries
Across the OECD, countries are at very different stages of development in terms of how health data is collected, stored, shared, and used. Countries differ in terms of how data is processed, transferred between organisations, linked across organisations and jurisdictions for federated states, and protected. These different approaches have very real consequences for what is understood about a disease in a country, and how data can be used to maximise health benefits while safeguarding privacy.
Most OECD countries could do more to ensure that primary care data for dementia is accurate. Even where dementia has been diagnosed in primary care, it may not be recorded in patient records. In one study, just 19% of patients who were screened and confirmed to have dementia in Indiana, the United States were found to have a dementia diagnosis subsequently included in their medical records (Boustani et al., 2005). In the UK, coding guidelines for general practitioners were developed through NHS England based on the belief that practitioners may be confused by dementia coding processes (NHS England, 2015).
Even where information about a dementia diagnosis is recorded in primary care data, the information can be difficult to access and share where primary care data is poorly linked to other health data systems. Across the OECD, primary care data remains one of the least regularly linked data sources in the health system. 10 of 22 OECD countries reported that they can link primary care data with other data sources, while just two reported that primary care data is regularly linked with other data sources to monitor health care quality and health systems performance (OECD, 2015b). Eight of 21 OECD countries – Austria, Canada, Finland, France, Latvia, Norway, Spain, and the United Kingdom (Northern Ireland, Scotland, and Wales) reported that primary care administrative datasets that can identify people with dementia were available at the national level. A further four countries – Australia, Japan, the Netherlands, and Sweden – could identify people using registry or survey data.
Dementia registries offer an alternative approach to improving the data landscape. Data registries are increasingly being used by countries to monitor and improve health outcomes related to specific diseases. Across different countries, registries have been shown to improve data transparency and lead to the identification of best practice, better health outcomes, and lower cost of care (Larsson et al., 2012). One study suggested that replicating a hip replacement surgery registry from Sweden could lead to cost savings of 8% in the United States by reducing second surgeries (Larsson et al., 2012). In recent years, a number of countries have developed registries focused specifically on dementia, though they remain uncommon. Six OECD countries have developed dementia registries at either the national or regional level; of those countries with regional dementia registries, at least two are in the process of scaling up to the national level. While dementia registries indicate a concerted effort to improve the data infrastructure for dementia, the depth and coverage of registries differs substantially between countries. A number of major data challenges remain even in the registry model, including incorporating data from different health care settings ∗ notably primary care ∗ and ensuring comprehensive coverage.
In recent years, the OECD has actively participated in global debates considering the key principles that should govern the use of health data. A comprehensive study of national health data systems published in 2015 identified eight key health data governance mechanisms that can help to ensure health data can be used to maximum benefits for patients and societies, while minimising risks to the privacy of patients and to the security of health data:
The health information system supports the monitoring and improvement of health care quality and system performance, as well as research innovations for better health care and outcomes.
The processing and the secondary use of data for public health, research and statistical purposes are permitted, subject to safeguards specified in the legislative framework for data protection.
The public are consulted upon and informed about the collection and processing of personal health data.
A certification or accreditation process for the processing of health data for research and statistics is implemented.
The project approval process is fair and transparent and decision making is supported by an independent, multidisciplinary project review body.
Best practices in data de-identification are applied to protect patient data privacy.
Best practices in data security and management are applied to reduce re-identification and breach risks.
Governance mechanisms are periodically reviewed at an international level to maximise societal benefits and minimise societal risks as new data sources and new technologies are introduced.
Source: OECD (2015), Health Data Governance: Privacy, Monitoring and Research, https://doi.org/10.1787/9789264244566-en..
SveDem: The Swedish Dementia Registry
The Swedish Dementia Registry SveDem is arguably the most comprehensive dementia registry currently operating in the OECD. Established in May 2007, the registry includes people receiving new diagnoses of dementia. All people included in the registry receive yearly follow-up. Demographic data, including age and gender and clinical information related to the dementia diagnosis (MMSE score, dementia sub-type, results of the diagnostic evaluation) are both included in the registry. In addition, further information, including pharmacological and non-pharmacological treatments received, available community-based support, and the time it took to receive a formal diagnosis following referral are all recorded (Religa et al., 2015). Separate indicators related to nursing care are also collected when patients reside in a long-term care facility. SveDem has increasingly considered how to incorporate patient-reported measures, including patient-reported outcome measures (PROMs) in their registry. At the moment, only the ‘Quality of Life in Late-Stage Dementia’ (QUALID) indicator, reported by carers for people with advanced dementia living in nursing homes, is included in the registry. There is recognition that including patient-reported measures for people with dementia in the early stages of the disease would also be beneficial (Religa et al., 2015).
Data is collected from 100% of memory clinics in Sweden, as well as 79% of general practices. Reporting from general practitioners has increased substantially since the launch of the registry, though full coverage may be difficult to achieve. While SveDem offers a useful tool for longitudinal follow-up, it is difficult to encourage general practitioners who see just a small number cases of dementia per year to invest the time it takes to participate in the registry. Sweden currently has 110 distinct disease registries, leading to the possibility that general practitioners may experience ‘registry fatigue’ and elect not to participate in those that cover conditions where their caseload is low. Nursing homes began to participate in the dementia registry in 2012. The number of nursing homes contributing data has risen from 63 in 2012 to 381 in 2015. Data from hospital settings is not included in the registry. Nursing homes are able to compare statistics from their home to others in the region and across Sweden, offering a useful benchmark for improving quality.
Because SveDem includes data from all memory clinics and the majority of general practices, it is considered a relatively robust measure of dementia incidence in the country. It provides a rich trove of data related both to diagnosis and types and pathways of care following diagnosis. Data from the registry has been used by researchers to further understanding of the dementia care pathway, including analyses of medications treatment by dementia sub-type, diagnostic and treatment differences by gender and age, and cost of diagnosis in primary versus secondary care (Religa et al., 2012; Johnell et al., 2013; Wimo et al., 2013). Seven indicators to measure quality of dementia care have been identified for use from the SveDem data (Religa et al., 2015). These are based on quality indicators identified by the Swedish national guidelines for dementia care and treatment, and are accompanied by target values for the registry to achieve (Religa et al., 2015).
The Danish Dementia Registry
The administrative Capital Region of Denmark initially launched a dementia registry in 2005, covering 30% of the population (Johannsen et al., 2011). The registry, which draws its data from secondary care settings, was expanded to cover the entire country beginning in 2016. Like in Sweden, the Danish registry enrols all new diagnoses of dementia in the database. All 39 memory clinics in Denmark are legally mandated to record new diagnoses of dementia with the registry (Johannsen et al., 2011). Unlike in SveDem, however, general practitioners are not obligated to participate in the registry, limiting the number of diagnoses that may be captured by the system. In Denmark, general practitioners can diagnose dementia, though because medications must be prescribed by a specialist, many dementia patients are likely referred on to specialist care and may still be captured by reporting from memory clinics. Data from nursing homes and inpatient hospital stays are excluded from the registry.
The Danish Dementia Registry was developed as a quality registry, and as such has identified a set of eight process and volume indicators focusing on the evaluative process, as well as target values for each. The Danish indicators were developed with by experts in diagnostic evaluation and are intended to be markers of quality diagnosis that can be easily recorded by clinicians in less than two minutes (Johannsen et al., 2011). In comparison to the Swedish quality indicators, the Danish indicators focus more on the diagnostic process and less on longer-term questions of care. However, the additional information collected by the Danish Dementia Registry, allows for many of the same or similar indicators to be evaluated.
Though the Danish Dementia Registry was only rolled out nationwide in 2016, results from the data collected have already been incorporated into dementia policy. Most notably, the results of the 2016 data collection indicated that many of the 39 memory clinics operating in Denmark see very few patients each year. The number of patients evaluated at memory clinics ranged from 5 to 1 200, while a large number of clinics saw fewer than 80 patients in 2016. As a result, as part of the Dementia Action Plan, the government has allocated DKK 170 million (approximately EUR 23 million) to restructuring and improving memory clinics so they are ‘bigger, better, and fewer.’ Financing will be distributed to the five administrative regions according to how well they meet the recommendations for memory clinics that were recently published by the National Board of Health.
Developing registries from insurance data: examples from Israel and France
SveDem and the Danish Dementia Registry represent a deliberate approach to build a data infrastructure that facilitates the development of a disease-specific registry based on data that was either available and not previously linked, or was not previously collected. In other countries where national insurance systems cover a large part of or close to the entire population, insurance fund data has been used to develop similar dementia registries using information collected for administrative purposes.
In Israel, the third-largest health maintenance organisation, Kupat Holim Meuhedet, developed a methodology to use their administrative data to measure service utilisation for people with dementia. Developed in concert with the Israel Centre for Disease Control, the registry collates dementia diagnoses made at the primary and acute care levels to link patient information across health care settings. In Israel, most diagnoses are made at primary health or geriatric health clinics, both of which are funded by health maintenance organisations. Prescriptions data is also included in the registry, though memory clinics, which are private entities located outside health maintenance organisations, are excluded. The registry will collect information related to health care utilisation in addition to basic demographic information. Initial analyses of the data collected by the dementia registry indicate that people with dementia use more health care resources – including more frequent doctor’s visits, more frequent hospitalisations, and longer hospitalisations – than similar populations without dementia. The dementia registry will be rolled out nationally, covering all four health maintenance organisations, by the end of 2017.
As part of the third French National Plan on Alzheimer’s Disease and Related disorders (2008-2012), a national dementia registry, the Banque nationale Alzheimer, was launched in 2009. The registry is intended to record all cases of dementia reported by memory clinics and memory research centres in France, as well as cases of dementia diagnosed by independent specialists. As of 2017, 435 memory consultation clinics, 28 memory research centres, and 74 independent specialists have contributed data to the registry, which now includes data on over 800 000 patients across mainland France. Information collected in the registry includes details of the dementia diagnosis, demographic information, and basic information about treatment (both pharmacological and non-pharmacological) and a patient’s living circumstances. However, data from other sources – including hospital data, primary care data, and medicines data, is not currently included in the registry. A recent evaluation of the Banque nationale Alzheimer raised questions over the representativeness of people who seek help at memory consultation clinics and memory research centres (Santé publique France, 2017). Among other recommendations, the review strongly encouraged linking the data available in the Banque nationale Alzheimer to other data sources. These include the national health insurance database Sniiram, which could provide information related to medicines consumption, hospitalisation, and registration of a patient with the category affectation de longue durée – a special category of reimbursement open to a limited number of long-term conditions, including dementia. The review additionally recommended linking information in the data bank with the national death registry, CépiDC (Santé publique France, 2017). The Banque nationale Alzheimer is currently undergoing a restructuring to improve the quality of its data and to expand its coverage to allow for a better analysis of care management and the continuum of care experienced by people with dementia.
Dementia surveillance registries in the United States
Three US states – South Carolina, West Virginia, and Georgia – have established databases that link a range of data sources into dedicated dementia registries. First established in 1988, the dementia registry in South Carolina incorporates data from inpatient hospitalisation, Medicaid, mental health records, home health, long-term care evaluations, emergency departments (in South Carolina), state health plans, vital records, community mental health centres and mental health resource centres, and other sources of information linked by a unique patient identifier (UPI) to develop an understanding of the current situation facing the state (Arnold School of Public Health, 2017). While the registry collects information related to the diagnosis, sociodemographic information, caregiver information, and the living circumstances of the person with dementia, it does not collect information on treatments and cannot be seen as a quality register similar to those in Scandinavia.
Dementia registries help to improve dementia care and policy, though in the longer-term countries should strengthen their health data infrastructure
The case studies discussed in this section highlight the range of forms dementia registries have taken across different countries. Other OECD countries have begun to evaluate the usefulness and feasibility of developing dementia registries, with plans for a national registry in Italy and recent calls to develop a registry in Australia. A number of key components can help registries to contribute effectively to our understanding of dementia and the quality of dementia care, and can be used to inform evidence-based policymaking.
Most importantly, dementia registries are most useful when they come close to capturing all diagnosed cases of dementia in a country. The proportion of diagnoses made in different health care settings will differ between countries, but in most of the OECD, both primary care and secondary memory clinics will be important sources of diagnosis. Efforts should be made to include these settings in dementia registries. It may be particularly challenging to include data from primary care clinics, particularly given that dementia likely represents a small proportion of all patients for most primary care physicians. Nevertheless, it is an important source of information, particularly considering that the population visiting their primary care physicians may be more representative of the population than those who continue on to specialised assessment centres. Financial or other incentives could be considered to engage primary care doctors in contributing to the registry. In Sweden, general practitioners were initially paid per diagnostic assessment, to encourage participation, though these benefits have since been lifted. Other important care settings that could be included to improve measuring quality outcomes include nursing homes, acute hospitals, and community-based day care centres.
Data generated from dementia registries can be an important tool in evaluating not only the care received by people with dementia in a given country, but also inform whether the institutions created to facilitate this care are operating as they should be. The reorganisation of memory clinics currently underway in Denmark, as a direct result of the first year of data from the Danish Dementia Registry, suggests that dementia registries are an important source of data in addressing important questions related how the care structures being developed to address dementia can themselves impact the quality of care.
To best facilitate improvements in quality, dementia registries should wherever possible encourage follow-up of individual cases be recorded within the registry. The Swedish BPSD registry provides a good example of how quality of care can be quickly improved when cases are monitored on a regular basis. Ensuring follow-up information is recorded in the registry will allow countries to understand better the full pathway of dementia care. In particular, including data from nursing homes and other intensive care settings would enable countries to use registries to develop a better understanding of living with more advanced forms of the disease.
References
Alzheimer’s Disease International (2015), World Alzheimer Report 2015: The Global Impact of Dementia – An Analysis of Prevalence, Incidence, Cost and Trends, Alzheimer’s Disease International, London.
Arnold School of Public Health, Office for the Study of Aging, University of South Carolina (2017), 2017 Annual Report: South Carolina Alzheimer’s Disease Registry, Columbia: University of South Carolina.
Auer, S. R., Span, E. and B. Reisberg (2015), “Dementia service centres in Austria: A comprehensive support and early detection model for persons with dementia and their caregivers – theoretical foundations and model description”, Dementia, Vol. 14, No. 4, pp. 513–527. https://doi.org/10.1177/1471301213502214
Azam, B. et al. (2016), “Trends in referred patient profiles in a memory clinic over 20 years”, Dementia, Vol. 15, No. 4, pp. 789–797. https://doi.org/10.1177/1471301214539691.
Barnes, D. E. et al. (2014), “Development and validation of a brief dementia screening indicator for primary care”, Alzheimer's & Dementia, Vol. 10, No. 6, pp. 656–665.e1. https://doi.org/10.1016/j.jalz.2013.11.006.
Benbow, S. M. et al. (2013), “Closing the diagnosis gap and improving care: the primary care memory clinic”, Progress in Neurology and Psychiatry, Vol. 17, No. 6, pp. 27–30.
Borson, S. et al. (2003), “The Mini-Cog as a Screen for Dementia: Validation in a Population-Based Sample”, Journal of the American Geriatrics Society, Vol. 51,, pp. 1451–1454.
Borson, S. et al. (2007), “Implementing Routine Cognitive Screening of Older Adults in Primary Care: Process and Impact on Physician Behavior”, Journal of General Internal Medicine, Vol. 22, No. 6, pp. 811–817. https://doi.org/10.1007/s11606-007-0202-8.
Borson, S. et al. (2013), “Improving dementia care: The role of screening and detection of cognitive impairment”, Alzheimer's & Dementia, Vol. 9, No. 2, pp. 151–159. https://doi.org/10.1016/j.jalz.2012.08.008
Boustani, M. et al. (2005a), “Implementing a screening and diagnosis program for dementia in primary care”, Journal of General Internal Medicine, Vol. 20, No. 7, pp. 572–577. https://doi.org/10.1007/s11606-005-0103-7
Boustani, M. A. et al. (2011), “Implementing innovative models of dementia care: The Healthy Aging Brain Center”, Aging & Mental Health, Vol. 15, No. 1, pp. 13–22. https://doi.org/10.1080/13607863.2010.496445
Bradford, A. et al. (2009), “Missed and delayed diagnosis of dementia in primary care: prevalence and contributing factors”, Alzheimer Disease & Associated Disorders, Vol. 23., No. 4, pp. 306-314.
Cahill, S. and M. Pierce (2013), “Briefing Paper on Dementia Diagnosis”, Genio Dementia Series, Briefing Paper No. 3, pp. 1–44, Mulingar, Ireland.
Cheung, G. and J. Strachan (2008), “A Survey of Memory Clinics in New Zealand”, Australasian Psychiatry, Vol. 16, No. 4, pp. 244–247. https://doi.org/10.1080/10398560701852131
Clark, M. et al. (2013), “Putting personalisation and integration into practice in primary care”, Journal of Integrated Care, Vol. 21, No. 2, pp. 105–120. https://doi.org/10.1108/14769011311316033.
De Lepeleire, J. et al. (2008), “The primary care diagnosis of dementia in Europe: An analysis using multidisciplinary, multinational expert groups”, Aging & Mental Health, Vol. 12, No. 5, pp. 568–576. https://doi.org/10.1080/13607860802343043.
Dodd, E. et al. (2014), “An evaluation of primary care led dementia diagnostic services in Bristol”, BMC Health Services Research, Vol. 14, No. 592, pp. 1–11. https://doi.org/10.1186/s12913-014-0592-3.
Dong, Y. et al. (2010), “The Montreal Cognitive Assessment (MoCA) is superior to the Mini-Mental State Examination (MMSE) for the detection of vascular cognitive impairment after acute stroke”, Journal of the Neurological Sciences, Vol. 299, No. 1-2, pp. 15–18. https://doi.org/10.1016/j.jns.2010.08.051.
Downs, M. et al. (2006), “Effectiveness of educational interventions in improving detection and management of dementia in primary care: cluster randomised controlled study”, BMJ, Vol. 332, No. 7543, pp. 692–696. https://doi.org/10.1136/bmj.332.7543.692.
Fick, D. M. et al. (2013), “Delirium superimposed on dementia is associated with prolonged length of stay and poor outcomes in hospitalized older adults”, Journal of Hospital Medicine, Vol. 8, No. 9, pp. 500–505. https://doi.org/10.1002/jhm.2077.
Fong, T. G. et al. (2012), “Adverse Outcomes After Hospitalization and Delirium in Persons With Alzheimer Disease”, Annals of Internal Medicine, Vol. 156, No. 12, pp. 848–W296. https://doi.org/10.7326/0003-4819-156-12-201206190-00005.
French, D. D. et al. (2014), “Healthy Aging Brain Center Improved Care Coordination And Produced Net Savings”, Health Affairs, Vol. 33, No. 4, pp. 613–618. https://doi.org/10.1377/hlthaff.2013.1221
Gomes, M. et al. (2017), “Cost-effectiveness of Memory Assessment Services for the diagnosis and early support of patients with dementia in England”, Journal of Health Services Research & Policy, Vol. 22, No. 4, pp. 226–235. https://doi.org/10.1177/1355819617714816.
Greaves, I. et al. (2015), “Gnosall Primary Care Memory Clinic: Eldercare facilitator role description and development”, Dementia, Vol. 14, No. 4, pp. 389–408. https://doi.org/10.1177/1471301213497737
Guideline Adaptation Committee (2016), Clinical Practice Guidelines and Principles of Care for People with Dementia, Guideline Adaptation Committee, National Health and Medical Research Council, Sydney.
Hughes, C. (2011), “What you need to know about the Medicare Preventive Services Expansion”, Family Practice Management, January/February, pp. 22–25.
Johannsen, P. et al. (2011), “Development of a dementia assessment quality database”, Aging & Mental Health, Vol. 15, No. 1, pp. 40–46. https://doi.org/10.1080/13607863.2010.508769
Johnell, K., Religa, D. and M. Eriksdotter (2013), “Differences in drug therapy between dementia disorders in the Swedish Dementia Registry: A nationwide study of over 7,000 patients”, Dementia and Geriatric Cognitive Disorders, Vol. 35, pp. 239-248. https://doi.org/10.1159/000348408.
Jolley, D., Benbow, S. M. and M. Grizzell (2006), “Memory clinics”, Postgraduate Medical Journal, Vol. 82, No. 965, pp. 199–206. https://doi.org/10.1136/pgmj.2005.040592
Khanassov, V., Vedel, I. and P. Pluye (2014), “Barriers to Implementation of Case Management for Patients With Dementia: A Systematic Mixed Studies Review”, The Annals of Family Medicine, Vol. 12, No. 5, pp. 456–465. https://doi.org/10.1370/afm.1677
Larsson, S. et al. (2012), “Use Of 13 Disease Registries In 5 Countries Demonstrates The Potential To Use Outcome Data To Improve Health Care's Value”, Health Affairs, Vol. 31, No. 1, pp. 220–227. https://doi.org/10.1377/hlthaff.2011.0762
Lathren, C. R. et al. (2013), “Improving dementia diagnosis and management in primary care: a cohort study of the impact of a training and support program on physician competency, practice patterns, and community linkages”, BMC Geriatrics, Vol. 13, No. 134.
Le Couteur, D.G. et al. (2013), “Political drive to screen for pre-dementia: not evidence based and ignores the harms of diagnosis”, BMJ, Vol. 347: f5125. doi https://doi.org/10.1136/bmj.f5125.
Lee, L. et al. (2014), “Primary Care–Based Memory Clinics: Expanding Capacity for Dementia Care”, Canadian Journal on Aging, Vol. 33, No. 3, pp. 307–319. https://doi.org/10.1017/S0714980814000233.
Lee, L., Kasperski, M.J. and W.W. Weston (2011), “Building capacity for dementia care”, Canadian Family Physician, Vol. 57,, pp. e249–e252.
Lindesay, J. et al. (2002), “The second Leicester survey of memory clinics in the British Isles”, International Journal of Geriatric Psychiatry, Vol. 17, No. 1, pp. 41–47. https://doi.org/10.1002/gps.522.
Livingston, G. et al. (2017), “Dementia prevention, intervention, and care”, The Lancet, Vol. 390, No. 10113, pp. 1–62. https://doi.org/10.1016/S0140-6736(17)31363-6.
Mate, K. E. et al. (2017), “An evaluation of the additional benefit of population screening for dementia beyond a passive case-finding approach”, International Journal of Geriatric Psychiatry, Vol. 32, No. 3, pp. 316–323. https://doi.org/10.1002/gps.4466.
Meeuwsen, E. J. et al. (2012), “Effectiveness of dementia follow-up care by memory clinics or general practitioners: randomised controlled trial”, BMJ, Vol. 344,, pp. e3086. https://doi.org/10.1136/bmj.e3086.
Mitchell, A. J., Meader, N. and M. Pentzek (2011), “Clinical recognition of dementia and cognitive impairment in primary care: a meta-analysis of physician accuracy”, Acta Psychiatrica Scandinavica, Vol. 124, No. 3, pp. 165–183. https://doi.org/10.1111/j.1600-0447.2011.01730.x
Nakanishi, M., and T. Nakashima (2014), “Features of the Japanese national dementia strategy in comparison with international dementia policies: How should a national dementia policy interact with the public health- and social-care systems?”, Alzheimer’s & Dementia, Vol. 10, No. 4, pp. 468-476. https://doi.org/10.1016/j.jalz.2013.06.005.
Nasreddine, Z. S. et al. (2005), “The Montreal Cognitive Assessment, MoCA: a Brief Screening Tool for Mild Cognitive Impairment”, Journal of the American Geriatrics Society, Vol. 53, No. 4, pp. 695–699.
National Institute for Health & Care Excellence (NICE) (2006), Dementia: supporting people with dementia and their carers in health and social care: Clinical Guideline, http://nice.org.uk/guidance/cg42.
NHS England (2015), Dementia diagnosis and management: A brief pragmatic resource for general practitioners, NHS England, Publications Gateway Reference Number 02615.
OECD (2015b), Health Data Governance: Privacy, Monitoring and Research, OECD Health Policy Studies, OECD Publishing, Paris, https://doi.org/10.1787/9789264244566-en.
OECD (2015a), OECD Reviews of Health Care Quality: Japan 2015: Raising Standards, OECD Reviews of Health Care Quality, OECD Publishing, Paris, https://doi.org/10.1787/9789264225817-en.
OECD (2017), Health at a Glance 2017: OECD Indicators, OECD Publishing, Paris, https://doi.org/10.1787/health_glance-2017-en.
Religa D. et al. (2012), “Dementia Diagnosis Differs in Men and Women and Depends on Age and Dementia Severity: Data from SveDem, the Swedish Dementia Quality Registry”, Dementia and Geriatric Cognitive Disorders, Vol. 33, pp. 90–95.
Religa, D. et al. (2015), “SveDem, the Swedish Dementia Registry – A Tool for Improving the Quality of Diagnostics, Treatment and Care of Dementia Patients in Clinical Practice”, PLoS ONE, Vol. 10, No. 2, pp. e0116538–14. https://doi.org/10.1371/journal.pone.0116538.
Robinson, L., Tang, E., and J.P. Taylor (2015), “Dementia: timely diagnosis and early intervention”, BMJ, Vol. 350, No. 14, pp. h3029–h3029. https://doi.org/10.1136/bmj.h3029
Santé publique France (2017), “La banque nationale Alzheimer, un outil pour la surveillance épidémiologique des démences en France ?” Santé publique France.
Tannebæk, K. et al. (2014), “The Danish ABC dementia: new e-learning for professional caregivers”, Alzheimer's & Dementia, Vol. 10,, pp. P180–P181. https://doi.org/10.1016/j.jalz.2014.04.200
Turner, J. J. (2013), “Proposed model of early dementia screening cannot be valid”, BMJ, Vol. 346, No. 1, pp. f505–f505. https://doi.org/10.1136/bmj.f505
Trzepacz, P. T. et al. (2015), “Relationship between the Montreal Cognitive Assessment and Mini-mental State Examination for assessment of mild cognitive impairment in older adults”, BMC Geriatrics, Vol. 15, No. 107, pp. 1–9. https://doi.org/10.1186/s12877-015-0103-3.
Tsolaki, M. et al. (2010). “Consensus statement on dementia education and training in Europe”, The Journal of Nutrition, Health & Aging, Vol. 14, No. 2, pp. 131-135.
UK National Screening Committee (2015), “UK NSC dementia screening recommendation”, https://legacyscreening.phe.org.uk/dementia.
U.S. Preventive Services Task Force (2014), “Understanding Task Force Recommendations: Screening for cognitive impairment in older adults”, Task Force Final Recommendation.
Waldorff, F.B. and S. Moller (2001), “Diagnostic evaluation of dementia in general practice in Denmark. A national survey”, Scandinavian Journal of Primary Health Care, Vol. 19, No. 2, pp. 117–121. https://doi.org/10.1080/028134301750235367.
Waldorff, F.B. et al. (2003), “Implementation of a clinical dementia guideline”, Scandinavian Journal of Primary Health Care, Vol. 21, No. 3, pp. 142-147. https://doi.org/10.1080/02813430310005136.
Wimo, A. et al. (2013), “Costs of diagnosing dementia: results from SveDem, the Swedish Dementia Registry”, International Journal of Geriatric Psychiatry, Vol. 28, No. 10, pp. 1039–1044. https://doi.org/10.1002/gps.3925
Woodward, M. C. and E. Woodward (2009), “A national survey of memory clinics in Australia”, International Psychogeriatrics, Vol. 21, No. 4, pp. 696–702. https://doi.org/10.1017/S1041610209009156.
Zadikoff, C. et al. (2008), “A comparison of the mini mental state exam to the Montreal cognitive assessment in identifying cognitive deficits in Parkinson's disease”, Movement Disorders, Vol. 23, No. 2, pp. 297–299. https://doi.org/10.1002/mds.21837.