What Can Other Countries Learn from Taiwan’s Long-Term Care Reforms?

Long-term care reform is often described through legislation, funding announcements or the number of services created. Taiwan’s experience suggests a more demanding test. Reform becomes meaningful when a person who develops care needs can find an understandable route into support, when services exist close enough to be usable, when families are not expected to carry every responsibility alone and when health care, rehabilitation and long-term support become more continuous rather than operating as separate worlds.

Taiwan has spent almost two decades developing that architecture. Long-Term Care 2.0, introduced in 2017, substantially expanded eligibility, service types, community infrastructure and public investment. Long-Term Care 3.0, launched in 2026 after Taiwan entered a super-aged demographic phase, is now attempting to connect those foundations more closely with health care, prevention, rehabilitation, smart technology, residential support and end-of-life continuity. The wider evolution is explored throughout the Taiwan Ageing, Long-Term Care & Community Support Knowledge Hub.

There is no single “Taiwan model” that another country can simply import. Taiwan’s tax arrangements, National Health Insurance system, local government structure, labour market, migrant-care economy, family expectations and provider landscape are specific to its own institutional context. The stronger international value lies elsewhere: in understanding how reform was sequenced, where infrastructure mattered as much as entitlement, why implementation capacity became increasingly important as coverage expanded and why the next generation of reform is focusing less on adding isolated services and more on connecting them.

The first lesson is that long-term care reform is cumulative

Taiwan’s present system did not emerge through one reform. The National Ten-Year Long-Term Care Plan established an earlier foundation before LTC 2.0 significantly widened the ambition from 2017. LTC 3.0 now builds on that enlarged platform rather than replacing it.

This sequencing matters internationally.

Governments sometimes design care reform as though a new policy can reset the system. In reality, long-term care capacity is accumulated through workforce development, provider relationships, physical infrastructure, local administrative knowledge, public awareness and repeated operational learning. Each stage creates assets and constraints for the next.

LTC 2.0 expanded the range and reach of home and community support, strengthened care management and developed the community-based A-B-C network. By 2026, Taiwan had more than 15,000 long-term care service points and more than 100,000 formal care workers within the expanded system. LTC 3.0 could therefore begin from a much larger delivery base than existed when LTC 2.0 started.

Its current challenge is correspondingly different. Expansion remains necessary, but the policy question has shifted towards continuity, quality and integration: whether a larger system can operate as a coherent pathway rather than an accumulation of programmes.

The transferable lesson is that reform should establish what the previous phase has genuinely built before defining the next one. Expansion, consolidation and integration are different policy tasks and require different evidence.

Political commitment matters most when it becomes durable infrastructure

Long-term care competes with health care, pensions, housing, education and other public priorities for fiscal and political attention. Taiwan’s experience demonstrates the importance of converting policy commitment into sustained system infrastructure.

The country’s Long-Term Care Development Fund provides a dedicated mechanism through which long-term care expenditure is supported from designated tax revenues and government resources. LTC 3.0 entered its first year with public expenditure on a substantially greater scale than existed when LTC 2.0 began.

Funding alone does not guarantee effective care. It does, however, make long-term implementation possible in a way that short-lived project funding cannot.

Workforce training takes time. Community facilities need predictable utilisation. Providers need confidence that services will continue. Municipalities and counties need administrative capability. Digital systems require maintenance rather than one-off purchase. Families need to know that an entitlement available this year will not disappear when a pilot ends.

This creates a broader lesson about organisational responsibility and accountability. Sustainable reform needs clarity about which functions belong nationally, which depend on municipalities and counties and which sit with individual service organisations.

National investment creates possibility. Delivery architecture converts that possibility into support.

Community infrastructure changed the practical meaning of access

One of the most internationally interesting features of Taiwan’s LTC 2.0 period was the deliberate development of community infrastructure.

The A-B-C community care network was designed to create different layers of local support: integrated care coordination, service provision and neighbourhood-level resources. The precise mix has evolved as the system has expanded, but the underlying policy principle remains important.

An entitlement has limited value if the nearest appropriate service is too distant, if a family cannot understand how to reach it or if individual providers operate without coordination.

Taiwan therefore treated geographical service development as part of system reform rather than assuming that demand would automatically produce an adequate provider market.

This approach is particularly significant in a country containing dense metropolitan areas, rural townships, mountainous communities, Indigenous communities and offshore islands. Equal national rules cannot by themselves create equal practical access.

The lesson is not that every country needs an A-B-C structure. Administrative systems differ too greatly for that. The transferable principle is that long-term care reform requires a spatial strategy. Governments need to understand where services physically exist, whether people can reach them and what happens in communities where conventional provider economics cannot sustain sufficient capacity.

This connects directly with wider questions of inequality, prevention and access. Coverage statistics should therefore be interpreted alongside travel, waiting time, transport, workforce availability and actual service use.

Scenario: the same national entitlement produces two different local realities

Consider two older people assessed as having comparable long-term care needs. One lives in a densely populated municipality where home care, day care, rehabilitation, transport and community services operate within a relatively mature network. The other lives in a sparsely populated township where providers cover long distances and specialist services are less frequent.

Formally, both people enter the same national long-term care framework. Operationally, their choices may differ substantially.

The first person can combine home support with day care and rehabilitation while a family member continues working. The second may technically qualify for similar services but find that travel time, transport availability and workforce capacity narrow what can realistically be arranged.

A weak evaluation of reform records two eligible people. A stronger evaluation asks whether both could obtain timely support capable of meeting their assessed needs.

For the local government, the second case is not merely an individual care-planning problem. Repeated cases reveal a resource-distribution issue. The appropriate response may involve differentiated funding, workforce incentives, transport, mobile or hybrid services, stronger community infrastructure or a different configuration from that used in an urban area.

Taiwan’s experience therefore demonstrates why national equity sometimes requires local variation in delivery. Standardising entitlement does not mean every locality must organise support identically.

Expansion should be judged by outcomes as well as utilisation

LTC 2.0 demonstrated that rapid expansion in service availability and utilisation is possible when funding, eligibility and local infrastructure are developed together. That achievement creates the next governance question: what should count as success once services become established?

Activity measures remain important. Governments need to know how many people receive support, where services operate, what is being purchased and how public money is used.

But mature long-term care systems need to look beyond volume.

Relevant questions include whether people maintain function, remain safely at home where that is their preference, avoid preventable deterioration, experience continuity, participate in their communities and receive support that reduces rather than merely redistributes family burden.

The shift from counting provision towards evaluating its effect is reflected in the broader discipline of quality data and performance measurement.

Organisations exploring similar questions can use the Quality Dashboard Builder to structure relationships between activity, quality, risk and outcomes. It is not a Taiwanese performance framework, but it illustrates the broader analytical transition required as systems mature.

The international lesson is straightforward but consequential: the evidence needed to prove expansion is not the same evidence needed to prove effectiveness.

Workforce policy cannot be separated from service expansion

Taiwan’s expansion of long-term care has been accompanied by substantial growth in its formal care workforce. That growth matters because new entitlements cannot become real services without people able to deliver them.

Yet Taiwan also illustrates the limits of treating workforce policy as a recruitment exercise.

The country is ageing while its working-age population is under pressure. Care work competes with other sectors for labour. Home care requires travel and flexible deployment. Residential services need continuous staffing. Community services require workers with different capabilities, while increasing health and long-term care integration creates demand for stronger collaboration between care workers, nurses, therapists, physicians, care managers and other professionals.

Migration adds another dimension. Household-employed migrant caregivers have become deeply embedded in Taiwan’s care economy, particularly where families require intensive live-in support. They do not simply form another branch of the formal publicly supported LTC workforce, and the distinction is important. Families may combine migrant caregiving with public long-term care services, creating interfaces that policy increasingly needs to recognise.

Taiwan’s experience therefore reinforces several workforce principles:

• service expansion should be accompanied by explicit workforce capacity planning;
• retention and working conditions matter alongside recruitment;
• training should follow changing care complexity rather than remain static;
• geographic distribution can be as important as national headcount;
• migration policy and care policy need to understand their mutual dependence; and
• technology should improve the productivity and reach of workers rather than be assumed to replace human care.

The wider workforce-planning lesson is particularly relevant for countries expanding formal care from a small base. Capacity assumptions should be tested before entitlement grows faster than deliverable supply.

The Predictive Workforce Risk Module provides one transferable way of examining how vacancy, turnover and workforce instability can translate into continuity risk. It does not replicate Taiwan’s labour arrangements, but the analytical principle applies across systems.

Families need support, not policy assumptions about unlimited availability

Taiwan’s reforms have expanded formal services within a society where families have historically carried substantial responsibility for long-term support. This relationship remains central.

Formal care does not make families irrelevant. Relatives continue to coordinate appointments, provide companionship, make practical arrangements and frequently deliver substantial unpaid support. Migrant live-in caregivers may also work within family households.

The policy risk is assuming that these resources are infinitely elastic.

Very low fertility, smaller families, geographic mobility, women’s labour-force participation and longer periods of later-life dependency change what households can realistically provide. An ageing society also produces more older carers supporting spouses or relatives while managing their own health needs.

Taiwan’s expansion of respite, day care, caregiver support and community services therefore represents more than an additional service category. It reflects recognition that the sustainability of the formal system and the sustainability of family care are connected.

Recent changes allowing eligible households employing foreign family caregivers to use specified community-based long-term care services within their approved allocation reinforce the same direction. The boundary between household care and formal services can be made more permeable rather than forcing families to choose one model exclusively.

This has wider relevance to family partnership and carer support. Governments should measure not only whether family care exists but whether it remains voluntary, manageable and compatible with the wellbeing and economic participation of the people providing it.

Scenario: reform works when support changes before the family reaches exhaustion

An older woman with progressive mobility limitations lives with her daughter. The daughter initially manages most support while working full time, with occasional assistance from other relatives. As her mother’s needs increase, the family begins using publicly supported home care.

A narrow service model might simply add more home-care hours whenever the daughter reports difficulty. A broader long-term care response examines what is actually creating pressure.

The care manager identifies that the older woman would benefit from regular community participation and rehabilitation, while the daughter needs predictable periods in which she can work without remaining continuously available. Day care and appropriate transport are incorporated alongside home support. Assistive equipment reduces the physical demands of transfers. The plan is reviewed as function changes.

The outcome is not that the family stops providing support. Instead, formal services absorb the tasks that are least sustainable for the household while preserving the relationships the family wants to maintain.

For system leaders, repeated cases of this kind should influence service development. If families consistently request help only after reaching exhaustion, the problem may be poor awareness, restrictive access, inadequate respite or care planning that identifies the care recipient’s tasks but not the sustainability of the wider household.

Taiwan’s reforms show why family caregiver policy should therefore sit inside long-term care design rather than at its margins.

Health care and long-term care require connection without institutional confusion

Taiwan has a mature National Health Insurance system alongside a separately financed long-term care system. That distinction is fundamental.

National Health Insurance finances medical care. Long-term care is supported through its own public financing and service arrangements. People, however, experience needs that cross the boundary.

An older person recovering from a stroke may need medical treatment, rehabilitation, assistive equipment, personal care and family support at different points in the same recovery. Organisationally, these functions can sit within different systems. From the person’s perspective, they form one life.

LTC 3.0 therefore places greater emphasis on connecting medical care and long-term care, including hospital discharge, post-acute recovery, home and community services and ongoing support.

The important international lesson is not that health and long-term care need to become one organisation or one funding stream. Taiwan demonstrates a different possibility: retain institutional distinctions while governing the interfaces more deliberately.

That requires information exchange, clear referral responsibilities, timely assessment and shared understanding of what happens when a person moves between settings. Interoperability and system integration therefore concern operational accountability as much as technology.

The distinction matters because structural merger can consume enormous policy effort without necessarily improving the experience of transition. Systems should define the problem they are trying to solve: fragmented information, delayed referrals, conflicting plans, unclear responsibility or gaps in service availability may require different interventions.

Scenario: the hospital discharge is complete, but recovery is not

An older man is admitted to hospital following a fall and fracture. Acute treatment is successful and he becomes medically ready to leave hospital. Before admission he managed most daily activities independently; afterwards he requires assistance with mobility, bathing and household routines.

If the health system defines success solely as safe discharge, responsibility can become fragmented at the point when the man is most vulnerable.

Under a more connected pathway, discharge preparation identifies long-term care needs before he returns home. Rehabilitation goals are linked with the support available in the home environment. Equipment and accessibility requirements are considered, the family understands what assistance it is expected to provide and longer-term care is adjusted as recovery progresses.

The important measure is not whether every service continues indefinitely. Effective recovery may reduce the level of formal care required.

For governance, this creates a cross-system outcome: whether the transition restored function and sustained life at home rather than simply whether the hospital completed a referral.

Taiwan’s LTC 3.0 direction illustrates why mature systems increasingly need to govern these interfaces. Hospital performance, rehabilitation performance and long-term care activity can each appear satisfactory while the person still experiences a discontinuous pathway.

Prevention changes the purpose of long-term care

Another important feature of Taiwan’s policy direction is the connection between long-term care and healthy ageing.

Traditional long-term care systems can become organised around compensating for established dependency. People qualify because they cannot perform certain activities, and services then perform those activities for them.

Support will always be necessary for people with substantial or progressive needs. But an ageing society also has an interest in maintaining function for as long as possible.

Taiwan’s emphasis on community activity, rehabilitation, health promotion, frailty prevention and active ageing reflects this wider objective. LTC 3.0 strengthens the direction by linking care more closely with medical and preventive support.

The operational implication is significant. Care plans should ask not only what assistance somebody needs today, but what capability can realistically be maintained or regained.

This aligns with prevention and early intervention, but prevention should not become a mechanism for blaming people whose health deteriorates or withholding necessary support. Some conditions are progressive; some disability is permanent; some people will require increasing assistance despite excellent care.

The stronger principle is that long-term care should avoid creating unnecessary dependency while remaining dependable when support is genuinely required.

Technology works best when it solves a care-system problem

Taiwan’s strong technology and manufacturing base creates obvious opportunities for smart care. LTC 3.0 is increasing the role of assistive and digital technology, including an expansion of support for eligible smart assistive devices during 2026.

The international lesson, however, is not that technologically capable countries should place more devices into care.

Technology creates value when it solves a defined problem.

A transfer device can reduce physical strain. Remote monitoring may provide useful information between visits. Digital records can improve continuity. Better scheduling can reduce wasted travel. Accessible communication technology can support participation. Smart equipment may help somebody remain safely at home.

But each technology also creates operational requirements: assessment, installation, training, connectivity, maintenance, data protection, review and eventual replacement.

This is why person-centred technology matters more than technology adoption as an end in itself.

The strongest technology strategy asks whether an intervention improves independence, safety, continuity, workforce productivity or caregiver sustainability. It also identifies who responds when a device generates an alert, who maintains it and what happens if it stops working.

Organisations examining similar transformation can use the Digital Transformation Readiness Assessment to structure questions about strategy, infrastructure, workforce adoption and digital resilience. The framework is transferable; decisions about Taiwanese eligibility, funding and regulation remain governed by Taiwan’s own system.

Local implementation is where national reform becomes unequal or effective

National long-term care policy in Taiwan is led by the Ministry of Health and Welfare, while municipalities and counties play central roles in organising access, developing resources and overseeing local implementation.

This creates necessary flexibility because local conditions differ. It also creates the possibility of variation.

A municipality with dense provider networks and strong administrative capacity faces different implementation challenges from a county containing remote communities and long travel distances. Local provider markets, workforce availability, transport and community organisations all influence what national policy becomes in practice.

The central governance task is therefore to distinguish useful adaptation from inequitable variation.

Variation may be appropriate when localities use different service models to achieve comparable outcomes. It becomes more problematic when people with similar needs experience materially different access because capacity, administration or information is weaker in one area.

National oversight consequently needs sufficiently granular evidence to identify where policy is not translating into practical access. Local government needs enough flexibility to respond without every operational adjustment requiring central redesign.

This balance is difficult in every decentralised care system. Taiwan’s experience reinforces the value of treating local implementation capability as part of reform design rather than assuming that policy automatically becomes practice after national approval.

Scenario: local data changes the question from demand to system design

A municipality observes continuing growth in long-term care utilisation. At first, the trend appears to demonstrate successful expansion. More residents are receiving home care, day services and other support.

A deeper review shows a more complicated pattern. Some neighbourhoods have strong service choice while others depend heavily on a small number of providers. Hospital referrals in one district convert quickly into community services, while another experiences repeated delays. Staff turnover is concentrated among organisations covering the longest travel routes.

None of these problems is visible from the headline number of people served.

The municipality combines utilisation, waiting, workforce and pathway information. It engages providers and community organisations to understand what the data cannot explain by itself. The response is then targeted: service development in weak areas, attention to transport and workforce deployment and closer monitoring of transition delays.

National policy has not changed, but local implementation becomes more effective because evidence is used diagnostically rather than descriptively.

This is an important lesson from a system of Taiwan’s scale and diversity. Data should help decision-makers understand why outcomes differ, not simply demonstrate that activity occurred.

Quality assurance has to evolve as the system grows

Rapid service expansion creates an assurance challenge. The more providers, workers and service points a system contains, the less realistic it becomes to rely primarily on episodic inspection or administrative compliance.

Taiwan’s Long-Term Care Services Act provides a statutory framework for long-term care institutions, personnel, service provision and oversight. Different categories of care remain subject to their relevant legal and administrative arrangements, while municipalities and counties carry important implementation and monitoring responsibilities.

As LTC 3.0 increases integration, quality increasingly needs to be understood across pathways as well as within organisations.

A home-care provider can meet its own standards while a person still experiences poor continuity because information does not reach another service. A residential institution can have appropriate internal controls while repeated transfers to hospital indicate a wider interface problem. A community programme can record high attendance without demonstrating whether people maintain function or social connection.

The next generation of assurance therefore needs to combine standards, service-user experience, workforce information, incidents, outcomes and pathway evidence.

The Governance Maturity Assessment can help organisations outside Taiwan consider how responsibility, evidence and escalation connect. The transferable lesson is that expansion requires governance capacity to grow alongside service capacity.

Reform should preserve choice as systems become more organised

Large care systems naturally develop categories, assessment instruments, payment rules and standard processes. These are necessary for equitable administration and financial control.

They can also unintentionally make the system more responsive to service categories than to individual lives.

Taiwan’s next phase therefore needs to maintain the connection between system efficiency and personal outcomes. An older person may value attending a community activity more than receiving another task-focused visit. A disabled person may need support organised around employment. A family may prefer a combination of formal services and its existing caregiver arrangements rather than substitution of one for the other.

Choice is meaningful only where services are available, information is understandable and care planning can respond to changing circumstances.

This is why outcomes-focused support provides a useful wider lens. Public systems need standard rules, but those rules should enable individually meaningful outcomes rather than make uniformity itself the objective.

Taiwan’s experience also shows what remains unresolved

International learning becomes less useful if it focuses only on achievements.

Taiwan still faces substantial long-term care challenges. Rapid ageing will increase demand while the working-age population contracts. Geographic variation will remain difficult to eliminate. Family care and migrant labour continue to carry significant parts of the care burden. Residential capacity, workforce development, affordability, integration and quality all require continuing attention.

LTC 3.0 itself should be understood as an active reform programme rather than evidence that these issues have already been resolved.

Some measures are being introduced in phases. The effects of new technology, wider eligibility and stronger health-care integration will depend on implementation over time. Expansion can also expose new bottlenecks: increasing entitlement may reveal insufficient workforce, while better hospital referral can reveal inadequate community capacity.

This is not evidence against reform. It demonstrates that successful reform changes the problems a system needs to solve.

LTC 2.0 was substantially concerned with creating access and infrastructure. LTC 3.0 increasingly has to address the consequences of scale: integration, complexity, workforce sustainability, quality, evidence and continuity.

The transferable lessons are principles rather than institutions

Taiwan’s long-term care arrangements cannot be detached from National Health Insurance, its tax system, municipalities and counties, family structures, labour migration or wider social policy. Another country adopting identical organisational labels would not reproduce the same system.

Several underlying principles are nevertheless internationally relevant:

• treat long-term care reform as a sustained programme rather than a short political initiative;
• build local delivery infrastructure alongside formal entitlement;
• plan workforce capacity at the same time as service expansion;
• support families without assuming that unpaid care can absorb unlimited demand;
• govern the interfaces between health and long-term care rather than expecting organisational boundaries to disappear;
• measure practical access and human outcomes as well as expenditure and activity; and
• use technology, prevention and community support to strengthen care rather than substitute for capacity that people still need.

These principles do not prescribe a financing mechanism or administrative structure. They describe capabilities that any sustainable long-term care system eventually needs to develop.

From service expansion to system stewardship

The most significant lesson from Taiwan may ultimately concern the changing role of government as long-term care matures.

During early expansion, the central tasks are often visible: increase funding, widen eligibility, create services and recruit workers.

A larger system requires a different form of stewardship.

Government needs to understand whether resources are distributed effectively, whether local variation is acceptable, whether workforce supply matches demand, whether providers remain sustainable, whether people experience continuity and whether public expenditure produces outcomes that justify continued investment.

This requires stronger analytical capability as well as stronger services.

It also requires feedback. Experience from care recipients, families, workers, providers and local government needs to influence national policy rather than flow only downwards through implementation instructions.

Taiwan’s transition from LTC 2.0 to LTC 3.0 can therefore be read as a shift from building a larger care system towards governing a more complex one.

That transition is relevant to many countries approaching population ageing from different institutional starting points. Scale solves some access problems, but it also makes coordination, evidence and accountability more important.

Conclusion

Taiwan’s long-term care reforms offer international systems something more useful than a model to copy. They show how long-term care capacity can be built progressively through sustained investment, expanded entitlement, community infrastructure, workforce development and clearer public responsibility, and how successful expansion then creates a new set of challenges around integration, quality and sustainability.

LTC 2.0 demonstrated the importance of making services visible and locally accessible. LTC 3.0 is testing whether those services can become more continuous across homes, communities, institutions and health care while supporting prevention, families, technology and ageing in place. The distinction matters: a system can become larger without automatically becoming more coherent.

For other countries, the strongest lessons lie in the architecture of reform. Funding needs delivery capacity. Entitlement needs local infrastructure. Workforce planning needs to accompany demand growth. Family support must be treated as a sustainability issue. Technology needs operational purpose. National standards require local intelligence, while quality governance must mature as the system becomes more complex.

Taiwan’s experience also demonstrates that long-term care reform has no final state. Demography, workforce supply, technology, family structures and public expectations continue to change. The strategic task is therefore not to design a perfect system once, but to create one capable of learning and adapting without losing sight of the people it exists to support.

That may be the most transferable lesson of all: durable reform depends as much on the capacity to improve the system as on the policies used to build it.