Taiwan

Routine cognitive assessments use the Mini-Mental State Examination (MMSE) test and the Montreal Cognitive Assessment Taiwanese (MoCA-T). Recent studies validated MoCA-T for the Taiwanese linguistic and educational context, showing higher accuracy compared with MMSE and clinical utility for patients living with mild cognitive impairment (MCI) in Taiwan.

Computerised Tomography (CT) and Magnetic Resonance Imaging (MRI) are available in tertiary and many regional hospitals. Positron Emission Tomography (PET) imaging, including amyloid PET and, more recently, tau PET, is concentrated in university centres and available mainly for complex differential diagnosis, younger-onset dementia, or research cohorts.

There is no evidence that genetic testing for APOE and monogenic mutations (APP, PSEN1, PSEN2) is part of the routinely reimbursed pathway. However, it is possible to have the test done in private clinics, such as Genetrack Taiwan. This is consistent with regional practice, where diagnosis is primarily the result of clinical criteria, imaging, and biomarkers rather than by APOE status.

Large tertiary centres provide CSF biomarker testing (Aβ42/40 ratio, t-tau, p-tau) in selected cases. The Taiwan Dementia Society released, in 2024, recommendations outlining principles for adopting blood-based biomarkers (including plasma p-tau181 and p-tau217, and plasma Aβ42/40) as clinical screening tools with confirmatory testing. Moreover, a recent study using prospective plasma samples from the Taiwan Alzheimer’s Disease Neuroimaging Initiative (Taiwan-ADNI) integrated plasma p-tau217 screening with follow-up amyloid PET for diagnostic confirmation into prediction models of Alzheimer’s disease and tau burden in the Taiwanese population.

Taiwan’s dementia care ecosystem is anchored in major university-affiliated medical centres such as National Taiwan University Hospital (NTUH), Taipei Veterans General Hospital (TVGH), the Chang Gung Memorial Hospital (CGMH) network, and Taipei Medical University hospitals (TMU/Wanfang, TMU/Shuang-Ho). These institutions provide comprehensive diagnostic services, including neurology, psychiatry, neuropsychology, advanced imaging (MRI, PET), and CSF/blood biomarkers, and manage complex or atypical dementia presentations. Large private hospitals operate with similar capabilities and often offer shorter appointment waiting times for self-funded services.

At the community level, Long-Term Care Plan 2.0 (LTC 2.0) organises support through the A–B–C network:

A-level integrated service centres act as regional case-management hubs, handling assessment, navigation, and referral (including dementia pathways).
B-level community service stations provide day care, rehabilitation, caregiver support programmes, respite services, and dementia-friendly education.
C-level neighbourhood “points” anchor hyperlocal activities such as memory cafés, screening and referral assistance, prevention workshops, and volunteer-led home visits.

This model aims to keep people at home as long as possible by coordinating transport, personal care workers, assistive devices, and respite. Taiwan is also expanding palliative and end-of-life care for dementia patients, with hospital-based palliative teams increasingly linked to community nursing services through national plans overseen by the Ministry of Health and Welfare (MOHW).

Under NHI, outpatient visits to primary care doctors and specialists are reimbursed with modest copayments that vary by facility level, lower in community clinics and higher in medical centres, with exemptions for low-income and certain chronic disease categories. Standard symptomatic AD medications (donepezil, rivastigmine, galantamine, and memantine) are covered under the national formulary; while patients pay a regulated copay for the initial month, subsequent second and third refills within a 90-day chronic illness prescription are exempt from copayment.

However, brand-name products, certain dispensing preferences, and convenience services can produce out-of-pocket expenses. Importantly, new anti-amyloid therapies (e.g., lecanemab, donanemab) are not broadly reimbursed in Taiwan. Since amyloid-targeting therapies require regular infusions with MRI monitoring, and specialist follow-up, the indirect costs — transportation, imaging fees not fully covered, and caregiver time — add to the financial load when these interventions fall outside the NHI benefit scope.

Taiwan does not offer dementia-specific cash allowances, but LTC 2.0 funds a wide suite of in-kind services that support caregivers directly or indirectly. These include case management, in-home personal care, day-care centres, community rehabilitation, short-stay and respite services, and formal caregiver training. Taiwan has also implemented measures to improve workforce retention, including raising wages for long-term care personnel under LTC 2.0, with average salaries reaching around NT$38,000 per month. Non-governmental organisations like the Taiwan Alzheimer’s Disease Association (TADA) provide additional support, such as caregiver helplines, psychoeducation, support groups, memory cafés, safety-awareness programmes, and nationwide dementia-friendly community initiatives. MOHW’s policy documents describe these expansions and integration efforts across health and social care.

Taiwan’s overarching framework for dementia is the Dementia Prevention and Care Policy & Action Plan 2.0 (2018–2025), issued by the MOHW. The strategy adopts a dementia-friendly society approach, structured around six pillars:

Public awareness and risk reduction, including large-scale health promotion campaigns, dementia-friendly community activities, and workplace education.
Early detection and timely referral, integrating opportunistic cognitive checks into chronic-disease management and linking primary care to LTC 2.0 navigation hubs.
Standardised diagnosis and treatment, promoting guideline-based cognitive testing, imaging pathways, and access to tertiary hospital expertise.
Workforce development, including dementia-care training for physicians, nurses, social workers, community case managers, and long-term care personnel.
Support for family caregivers, providing training, respite, day care, and psychological support through the LTC 2.0 A–B–C community network.
Dementia-friendly communities and environments, expanding memory cafés, volunteer programmes, safe-wandering initiatives, and city/county certification programmes.

The plan is explicitly linked to Long-Term Care Plan 2.0, which provides the community infrastructure (A/B/C) needed to operationalise dementia policy at the local level. Together, they aim to create a system where early detection at clinics seamlessly connects to long-term home and community care.

Emerging priorities for the remainder of the 2018–2025 cycle, outlined in ongoing updates from MOHW and NHI, focus on scaling and standardising implementation rather than introducing new policy layers. The main forthcoming directions include:

Expansion and strengthening of LTC 2.0’s A–B–C network: The government aims to enhance case-management capacity, increase coverage in rural and indigenous communities, and raise the quality and consistency of home-care and respite services. This includes more structured training for community workers and improved data sharing across localities.

Embedding new diagnostic technologies in tertiary pathways: Major academic hospitals and MOHW taskforces are developing consensus guidance on the use of blood-based biomarkers (p-tau, Aβ ratios) as first-line signals, with PET imaging reserved for confirmatory or complex cases. Standardised protocols are being piloted through Taiwan-ADNI and tertiary neurology departments to ensure consistent interpretation and referral logic.

Continued optimisation of Universal Health Coverage (UHC): Following the 2023 copay reforms, NHI is strengthening tiered usage of primary/community providers to reduce crowding in medical centres. For dementia, this means streamlining referral pathways, reducing unnecessary repeat imaging, and improving financial protection for long-term home care.

Cross-sector guidance on diagnosis and care: Government–academic working groups are preparing new professional consensus statements to reduce regional variation in cognitive testing, biomarker use, and follow-up schedules, and to align primary-care providers with tertiary standards.

According to the latest announcements, LTC Plan 3.0 is to be launched in 2026. The new plan builds on the foundation of LTC 2.0 and responds to the country’s transition into a super-aged society. It is expected to further advance a community-based, person-centred system by strengthening continuity of care and pursuing key goals such as healthy ageing, ageing in place, and dignified end-of-life care.

A landmark 2025 TW-ADNI study proposed one of Taiwan’s most influential diagnostic innovations to date: a two-step plasma-to-PET triage workflow designed to make Alzheimer’s disease diagnosis both more accessible and more cost-effective. The model begins with Step 1, a plasma p-tau217 screening test, which TW-ADNI researchers showed could reliably predict amyloid pathology with high sensitivity and specificity. Because plasma testing is accessible, minimally invasive, and quick to process, it can be deployed at scale in busy outpatient clinics, allowing memory specialists to screen far more patients than previously possible using CSF or PET alone. This first step is particularly well suited to Taiwan’s volume medical centres, where waiting times for advanced imaging and specialist evaluation can slow diagnostic timelines.

Step 2 of the workflow reserves amyloid PET imaging for only those patients who screen positive or borderline on plasma biomarkers. This selective use of PET allows clinicians to confirm amyloid status when it is most clinically meaningful, while dramatically reducing the number of unnecessary scans. TW-ADNI’s modelling demonstrated that such an approach can cut PET utilisation by a substantial margin without compromising diagnostic accuracy, essentially redirecting high-resource testing to the subgroup most likely to benefit.

The broader implication of this workflow is its alignment with Taiwan’s NHI priorities, which emphasise efficient resource allocation and precision diagnostics. By shifting the initial diagnostic burden onto scalable blood tests, the system can reduce costs, improve access for patients in both urban and rural regions, and speed up the pathway from suspicion to confirmed diagnosis. The TW-ADNI plasma-to-PET workflow is therefore emerging as a template for how middle-income and high-income Asian health systems can incorporate biomarker innovation into routine practice while maintaining financial sustainability.

The Taiwan Dementia Society released, in 2024, one of the first professional recommendations in Asia outlining how blood-based biomarkers should be incorporated into clinical Alzheimer’s disease pathways. This guidance represents a major shift toward scalable, cost-effective diagnostics, reflecting both international evidence and Taiwan’s own research capacity. The document emphasises that plasma markers can enhance earlier detection, streamline referral decisions, and reduce dependence on more invasive or expensive tests.

Central to the recommendations is the clinical use of several key biomarkers: p-tau181, p-tau217, the plasma Aβ42/40 ratio, and NfL (neurofilament light). These markers have shown high diagnostic accuracy in distinguishing Alzheimer’s disease from other dementias and in predicting amyloid and tau PET positivity. Other studies increasingly support their reliability in local populations, reinforcing the rationale for formal clinical adoption.