A new one‐stop interdisciplinary cognitive clinic model tackles rural health inequality and halves the time to diagnosis: Benchmarked against a national dementia registry

Unequal access to cognitive assessments is a major barrier to timely diagnosis, especially for those living in rural or remote areas. ‘One‐stop’ cognitive clinic models are a proposed solution, but few such clinics exist. We evaluate the implementation of a new one‐stop State‐wide clinic model in Tasmania, Australia, where 27% of people live in rural/remote areas.


| INTRODUCTION
Dementia affects more than 50 million people worldwide and, with ageing populations, 150 million cases are predicted by 2050. 1 In Australia, almost half a million people have dementia and it is the greatest cause of disability in older adults. To reduce the prevalence and impact of dementia, we need to assess and diagnose cognitive symptoms earlier and modify dementia risk factors. 2 Timely diagnosis substantially reduces healthcare costs as individuals with undiagnosed dementia have higher morbidity, lower quality of life, more hospital attendances, longer admissions and three times higher associated costs. 3 Despite the widely acknowledged importance of timely diagnosis, less than half of those with dementia receive a diagnosis. 4 A major barrier is the complexity of the diagnostic process which usually requires a comprehensive assessment involving multiple visits to a series of health professionals and ancillary tests. 5 Primary care, and most hospital clinics are simply not able to provide this. In response, 'Memory Clinics' were developed in the USA 40 years ago and, since then, various cognitive clinic models have been developed around the world, providing a comprehensive service at no extra cost 6 and better satisfaction with care. 7 Similar to most other countries, in Australia, there is considerable variation in pathways to a diagnosis, and there are insufficient Memory Clinics to meet demand. 8

Australian Clinical Practice
Guidelines recommend that people experiencing cognitive symptoms should be assessed by comprehensive memory services 9

but most
Australians have to navigate a sequence of visits to private clinics or public hospitals. 10 This is particularly problematic for the 30% of the Australian population who live in rural or remote areas, geographically distant from the metropolitan specialist centres. 11 In 2020, the Australian Dementia Network (ADNeT) brought together leading researchers, clinicians, and consumers to create a network for dementia prevention, treatment and care. A recent survey of 60 Australian cognitive clinics highlighted the need for clinics to adopt a more holistic interdisciplinary assessment. 8 In addition, qualitative research exploring the views of GPs, people with dementia, and other community members in Australia, suggested a 'one-stop' approach would be preferable-in terms of speeding up the time to diagnosis and to reduce inequity of access, especially for those living outside metropolitan areas. 12 Of the seven Australian States, Tasmania has the oldest demographic profile and the highest rates of modifiable dementia risk factors. 13 Dementia prevalence in Tasmania has increased by 30% over the last decade, putting substantial pressure on health and social services. Furthermore, the whole of Tasmania is classified as 'regional to remote' based on the Australian Statistical Geography

| Participants
Adults who live in Tasmania, and referred by a GP to the Clinic with at least 3 months of cognitive symptoms, were eligible. Exclusions include prior diagnosis of dementia, acute behavioural disturbance and active substance abuse. During the development phase, Clinic staff engaged Tasmanian GPs via webinars, a new website and by written information to practice managers for distribution to practices. The template was designed by Clinic clinicians together with Primary Health Tasmania, then posted on the Clinic website (https:// islandclinic.utas.edu.au).

| GP referral
Referring GPs complete a template (Appendix 1) which includes mandatory data on cognitive screening and pathology results. GPs are advised to correct any reversible causes of cognitive symptoms and reassess after 3 months. The Clinic GP assesses referrals for eligibility, engages with the referring GP to ensure that all clinic prerequisites are completed before applying a triage category and informing the patient and GP of estimated waiting time.

| MRI brain scan
An MRI brain scan is arranged with the local radiology provider (I-

| Screening health questionnaires
We post a 'Welcome pack' to patients 2 weeks before their appointment. This outlines Clinic processes, requests a support person accompanies the patient, and includes questionnaires on mood, sleep and activities of daily living to be completed at home (Appendix 3).

| Capacity and consent
Upon arrival, the Clinic GP or psychologist meets with the patient and support person to assess capacity and obtain written consent for: a) clinical assessments for diagnostic purposes and, b) participation in research; Appendix 4. This process is video recorded. Clinical consent follows local legislation and guidance. 16 If the patient gives verbal consent to research, we use an adapted version of the University of California, Brief Assessment of Capacity to Consent to formally assess capacity.

| Medical consultation
The patient rotates through three one-hour assessments in the morning ( Figure 1

| Neuropsychology assessment
A major challenge of one-stop clinic models is administering a neuropsychological assessment that is comprehensive in the context of the Clinic's time constraints. This has been achieved in the ISLAND Clinic using the one-hour protocol outlined in Table 1 and by developing an automatic scoring system to score tests and produce normed standard scores. A neuropsychologist, or psychologist under the guidance of a neuropsychologist, administers a battery of tests, that align with ADNeT recommendations (Table 1). Neuropsychological profiles contribute to the diagnosis and facilitate prognosis, 17 including predicting risk of mild cognitive impairment (MCI) conversion to AD, 18 and help identify non-degenerative factors such as functional cognitive disorders and mood disorders.

| Physiotherapy and movement assessments
In recognition of the interconnection between measures of frailty, falls and motor function with dementia risk and cognitive decline, a physiotherapist, or researcher under supervision, conducts a movement assessment including objective measures relevant to diagnosis, ALTY ET AL.

F I G U R E 1
The ISLAND Clinic 'one-stop' Cognitive Clinic flow. The patient moves through a series of assessments. Steps 3-11 are performed during a single day. Usually-Step 2 (MRI scan) can be offered for the afternoon/evening before Clinic for patients who live remote to the Clinic so all assessments are completed over a 24-h period. Patients rotate through steps three to five in different orders during the morning. The patient has a lunch break while the interdisciplinary team are in the consensus diagnosis meeting. Steps 8-11 are undertaken in the afternoon.
prognosis, functional abilities and research. 19 Tests include grip strength, single and dual-task gait assessments, balance measures and computerised upper limb motor function tests (e.g. TAS Test 20 ).
We also collect data on pain, falls and physical activity. After the patient has completed all 3 assessments, they have a lunchbreak and then return in the afternoon.

| Blood sample for biomarker and genetic analysis
After lunch, patients are invited to provide blood samples (collected by a phlebotomist, 3 � 8.5 mL Vacutainer® tubes) for research purposes, particularly genetic and protein biomarker studies. Plasma and serum samples are retained indefinitely so current, and future, biomarkers of brain ageing and neurodegeneration can be analysed.
DNA is also extracted from the samples to investigate genetic changes related to dementia risk.

| Interdisciplinary team (IDT) meeting and consensus diagnosis
The IDT meets over the lunchbreak where the three sets of assessments are discussed, MRI scans are presented online (by a Radiologist remotely) and a consensus diagnosis and management plan are formed. Diagnoses are subjective cognitive decline (SCD), MCI or dementia (all cause); see Appendix 6 for definitions. Diagnoses are sub-categorised according to clinical pattern (e.g amnestic MCI, frontotemporal dementia) and likely pathological cause (e.g. vascular, Alzheimer's). The risk (high/low) of progression for cases of SCD and MCI, based on criteria, 18,21 is also documented.

| Delivery of diagnosis
The patient and their support person meet with the doctor or neuropsychologist to discuss the consensus diagnosis and management plan. Verbal information is supplemented with information sheets (https://www.dementia.org.au/resources/help-sheets). Signposting to sources of information on modification of risk factors, ISLAND Project and resources from My Aged Care or the National Disability Insurance Scheme (NDIS) are provided, including referral information for government funded community supports.

| Post diagnosis support meeting
All patients, regardless of diagnosis, are invited to meet with a Dementia Australia (national peak body for dementia) support worker who attends the Clinic -for post-diagnosis counselling, registration for courses that provide education and peer support, and to arrange further review through their local offices.

| National clinical quality registry
The Clinic is part of the ADNeT Registry, a national clinical quality registry for people newly diagnosed with dementia or MCI 22 ; the primary aim is to collect data to monitor and enhance quality of care and patient outcomes. We upload demographic and clinical data of

| Management plan
Each member of the IDT populates an online template management plan letter, that includes information on all the assessments undertaken, diagnosis, general and specific recommendations, recommendations for medications (if indicated) and driving safety. This is sent to the GP 2-3 weeks later. The Clinic GP phones the patient to advise them that the report will be sent on that day, obtain feedback, answer questions, confirm that they have an appointment with the referring GP and may obtain a copy of the report from the GP once discussed.
T A B L E 1 Battery of neuropsychological tests used in ISLAND Clinic (<1 h protocol).

Duration (minutes) Domain
Test of premorbid functioning a,b Word list of words with irregular spelling, read aloud to establish baseline educational status and scored according to correct pronunciation

5-10 Premorbid function and reading
Rey auditory verbal learning test c A list of 15 nouns is read to the participant 5 times.
After each reading patients recall as many items as they can. A distractor list is then read (followed by participant recall), and then the participant is asked to recall the original list. After approximately 20 min the participant is asked to recall the original list. They are then asked to identify the items from a written list Consists of brief narratives that patients are asked to listen to and then recount, with wording as close as possible to the original. It has immediate and delayed recall, and a delayed recognition component.

5-10 Semantic/structured verbal memory
Trail making test c The task resembles a dot-to-dot puzzle, in which patients connect circles (numbered 1-25) in ascending order In part B half the circles contain numbers and the other half contain letters. The participant is instructed to connect circles switching between numbers and letters e,g, 1-A-2-B-3-C etc Scored primarily for duration, but pen lifts, sequence, switch and inability to finish are typically recorded

5-10
Visual scanning and attention, visuo-motor, and speed of cognitive processing Attentional switching/Executive function Part A Part B

WAIS-IV digit span subtest b,e
Patients are asked to listen to and then recite strings of digits. The assessment is in three parts-string recitation: 1) in original order, 2) in reverse order, 3) in numerical sequence.

3-6
Simple attention and working memory WAIS-IV similarities subtest b,e Patients are asked to explain how two words are alike. Initial items are relatively concrete, final items are substantially more abstract.

2-5 Abstract verbal reasoning
Letter fluency c Patients are asked to name as many words as possible, in 60 s, that begin with a specified letter 10 Orthographic fluency Category fluency c Patients are asked to name as many words as possible, in 60 s, that belong to a specific category.

Semantic fluency
Note: The validated neuropsychological tests used in the 1 h protocol of the ISLAND Clinic assess all the major cognitive domains and superscript numbers refer to the original publications of the tests.

| Data analysis
Electronic data collected via the MAB, questionnaires, cognitive and movement assessments is uploaded to a secure encrypted database.
For this evaluation, we extracted parameters from the database and ADNeT Registry and used descriptive statistics. We used the RE-AIM (measuring Reach, Effectiveness, Adoption, Implementation and Maintenance) framework to evaluate the Clinic implementation, 23 where Reach referred to the number, and demographics, of patients assessed; Effectiveness to the proportion who received a same-day cognitive diagnosis; Adoption to the proportion of GPs who referred patients; Implementation to the timeliness of appointments and patient feedback. As the Clinic is new, maintenance was not evaluated. 15.9% each from the North (urban and rural), South (rural and remote) and Midlands (rural), 46.8% from Greater Hobart (regional urban) and 5% from the East Coast (rural). 40 (36.8%) of all patients assessed were from rural or remote areas.

| Effectiveness
98.5% (128/130) received their diagnosis on the same day as Clinic assessments. The diagnoses and demographic details are outlined in

| Adoption
129 different GPs referred to the Clinic over the first year, representing 19.0% (129/680) of GPs in Tasmania. All patients were referred by GPs but five were initially identified by secondary care physicians who asked GPs to refer.

| Implementation
Of 89 patients diagnosed with dementia or MCI during the study period, 83 were eligible for the Registry and 81 were recruited. The most recent ADNeT Registry site report (September 2022), showed that our patients had high participation rates (96.6%) in the Registry, constituting 5.7% of participants. Compared to other ADNeT Registry sites, our Clinic had: a larger proportion of patients diagnosed with MCI (53.6% vs. 29.8%; See Table 3) and patients were younger (77 vs. 80 years for dementia, 70 vs. 78 years for MCI) and better educated (38.1% vs. 21.9% having tertiary education). The Clinic's wait time between referral and appointment was comparable to other sites (median: 78 vs. 71 days) but the wait time between appointment and diagnosis was shorter (0 vs. 41 days). As a result, our patients received their diagnosis 2 months quicker than other sites, which was equivalent to half the waiting time from referral to diagnosis (78 vs. 133 days after referral).
Our clinic has a higher patient survey response rate (78.8% vs. 50.9%). Of the 41 (out of 52) patients who returned the survey, two thirds (65.9%) and half (46.3%) rated their health and well-being as "Good" or "Very Good", respectively (see Table 4). Over 90% of survey respondents rated their overall clinic experience as "Good" or "Very Good" and agreed the experience met their expectations; Table 4. Most cognitive clinics can be described as interdisciplinary centres specialising in the assessment and diagnosis of dementia 8 but there is no gold standard recommendation for the composition of the clinical team nor the model of service delivery. 25 The ISLAND Clinic's model can be described as a one-stop model and fulfils many, but not all, reported preferences of GPs, older people, community representatives, local government, and healthcare professionals in a multidisciplinary "one-stop-shop" cognitive clinic. 12 We recognise that many memory clinics are run by old age psychiatrists and future development of the clinic would seek to include this speciality as well as occupational therapy expertise.

| DISCUSSION
Experience from our Clinic suggests that participation in a quality registry should be encouraged as it helps better understand patient characteristics and inform the development of local quality improvement initiatives. For example, compared to other ADNeT participating Registry sites, our clinic has a higher proportion of patients diagnosed with MCI which presents a unique opportunity to implement evidence-based dementia prevention programs. 26 Participation in the ADNeT Registry also enables us to have independently collected patient-reported outcome and experience data which is valuable to inform patient-centred care.
The research-focused ISLAND Clinic is well placed to contribute to the development of seamless models of care cutting across health and aged care systems, given the inadequate state of services in Australia. 27 Primary care partnerships could expand to include clinicians from a range of allied health professions, supporting ongoing future care needs and institutionalisation 24 but future work should include formal health economics analysis and integration with the local public health service. This is particularly important as the number of participants assessed each week was relatively low, especially considering the number of staff involved, and the ongoing successful administration of the clinic will likely require more streamlining and refinement.

| CONCLUSION
In conclusion, the ISLAND one-stop interdisciplinary cognitive clinic model provides a successful approach to a timely diagnosis of cognitive symptoms. This model may help inform service development in other parts of the world, particularly in regional, rural and remote areas. Wider availability of timely dementia diagnosis will prepare a foundation for ongoing work to improve post-diagnostic support and enable people living with dementia to participate in clinical trials at an earlier stage.