Integrating Health and Social Care in Latvia: Closing the Gaps Between Systems

An older person can be medically stable enough to leave hospital and still be unable to manage safely at home. A social-care worker can recognise that someone's condition is worsening but have no authority to diagnose or treat it. A family doctor may understand a person's health needs while knowing little about whether they are actually coping with eating, washing, mobility or medication routines between appointments. These are not unusual exceptions. They are the everyday interface between healthcare and social care.

In Latvia, that interface matters increasingly because population ageing, disability, chronic illness and community-based long-term care mean that many people need both systems at the same time. The wider structure is explored throughout the Latvia Ageing, Long-Term Care & Community Support Knowledge Hub, where national health responsibilities, municipal social services, registered providers, families and community infrastructure intersect around the same person.

The two systems remain distinct for legitimate reasons. Healthcare is responsible for diagnosis, treatment, clinical monitoring and rehabilitation delivered through health-service structures. Social care is concerned with support for daily living, social functioning, rehabilitation, community participation and maintaining quality of life where age, disability or other circumstances limit independence. Problems arise not because those roles are different, but because people's needs do not respect organisational boundaries.

Latvia's current policy direction increasingly recognises this. The Ministry of Health explicitly frames integrated healthcare as a process that should connect different levels of health service and work across boundaries with other ministries and municipalities, including social care and disability. The 2026–2027 Social Services Improvement and Development Plan similarly emphasises stronger cooperation between social-service providers, healthcare, education and municipal structures. Most significantly, a new integrated home-care pilot began in Riga in August 2026, testing a model in which health and social care are coordinated around one person at home.

Integration begins with recognising that health and social need are different but inseparable

Integration does not mean merging healthcare and social care into one indistinguishable service.

Clinical care requires its own professional standards, treatment responsibilities and funding arrangements. Social services operate through their own legal framework, assessment processes, municipal responsibilities and provider infrastructure. Those differences protect important functions and accountabilities.

The integration problem emerges when the distinction becomes a gap.

An older person recovering from a stroke may need wound care, nursing or rehabilitation while also requiring help with bathing, food preparation and safe movement around the home. A person with dementia may need primary healthcare for chronic conditions while depending on social-care workers to observe daily changes that indicate deterioration. Someone with severe physical disability may rely on personal assistance and technical aids while also receiving specialist clinical treatment.

No single professional can legitimately perform every function. Yet somebody needs to see the whole picture.

This is why interoperability and system integration should be understood as an operating principle rather than solely a technology issue. Integration requires processes, information and responsibilities that allow separate services to function around one person without forcing that person or their family to become the system coordinator.

Latvia’s institutional structure creates predictable coordination points

The Ministry of Health develops national health policy, while the National Health Service administers important elements of state-funded healthcare and contracts healthcare providers. Municipal social services operate within the national social-service framework and have substantial responsibility for assessing needs and organising social support for residents. The Ministry of Welfare provides the national policy and legal framework for social services and social assistance.

These responsibilities create a system in which coordination is required vertically and horizontally.

Vertically, national policy has to translate into local service arrangements. Horizontally, healthcare and social-care organisations need to coordinate around people whose needs cross both systems.

Operationally, several interfaces matter repeatedly:

• hospital discharge into home or community support;
• home healthcare operating alongside municipal home care;
• long-term social-care institutions managing residents with significant healthcare needs;
• primary care interacting with social workers and care providers;
• rehabilitation connecting clinical recovery with everyday functioning; and
• palliative care combining health, social and family support.

Each interface has different funding, professional and organisational arrangements. Integration therefore cannot rely on one generic coordination procedure.

Home care exposes the divide most clearly

Latvia already provides state-funded healthcare at home to defined groups of patients, including people with chronic conditions and movement limitations who cannot attend a medical institution, people discharged after certain hospital treatment and people requiring specified forms of rehabilitation at home.

This service is healthcare. It can involve nurses, physician assistants and rehabilitation professionals according to the person's needs and eligibility.

Municipal home care is different. It supports daily living where a person cannot adequately care for themselves because of age, functional impairment or other circumstances. Municipal social services assess the person's situation and the contribution that family members or others in the household can realistically make.

Many people need both.

The risk is that each system operates correctly in isolation while the person experiences fragmentation. A nurse may visit to provide a clinical intervention without knowing that the person's care worker has observed increasing confusion. A home-care worker may repeatedly assist with meals while being unaware that the person's diet has changed after hospital treatment. A family member may separately contact several providers because no one professional coordinates the complete plan.

The distinction between services remains necessary. The absence of a coordination mechanism does not.

Scenario: two services are present but the person still experiences a gap

An 84-year-old man in Riga has heart failure, reduced mobility and increasing difficulty managing personal care. He receives state-funded healthcare at home following a deterioration and municipal home-care support for everyday activities.

Both services are formally appropriate. The nurse monitors his clinical condition and provides healthcare interventions. Care workers assist with washing, dressing, meals and the home environment.

Over several days the care workers notice that he is becoming more breathless when walking from his bedroom to the bathroom. They record the change within their own service system but are uncertain whether the pattern requires urgent clinical attention. At the same time, the healthcare team does not routinely see the social-care records.

His daughter becomes the practical link between the two services, telephoning one provider to explain what the other has observed.

A more integrated model does not ask the care worker to make a clinical judgement. It establishes a clear route through which an observed change can be communicated quickly to the professional able to assess it. The nurse remains responsible for clinical assessment; the care worker contributes valuable real-world observation.

The improvement comes from connecting responsibilities, not blurring them.

The Riga integrated-care-at-home pilot is an important operational experiment

From 3 August 2026 to 31 July 2027, the National Health Service and Riga municipality are testing an integrated-care-at-home model for eligible adults who are declared and resident in Riga and have complex health and social-care needs.

The pilot is significant because it addresses precisely the fragmentation experienced by people who currently need to contact several providers separately.

Its stated objective is to create and test a model in which coordinated health and social-care support is delivered at the person's residence through one service arrangement. Assessment is linked to specified clinical pathways, while participating providers need the relevant contractual relationships for both state-funded healthcare and Riga municipal social care at home.

This is not yet a nationwide integrated-care system. It is a time-limited pilot in Riga. That distinction should remain clear.

But operationally it represents an important shift in design. Instead of expecting separate healthcare and social-care services to coordinate informally around the margins of their existing processes, the pilot tests integration as part of the service model itself.

The value will depend on what is learned. The key question is not merely whether participating organisations can deliver both forms of support. It is whether the model reduces duplication, improves continuity, clarifies responsibility and produces a better experience for people with complex needs.

Hospital discharge is where fragmented responsibility becomes highly visible

Hospital discharge sits at one of the most consequential health-social interfaces.

A hospital's central responsibility is healthcare. Once inpatient treatment is no longer necessary, keeping a person in hospital simply because community support has not been organised is neither an effective use of hospital capacity nor necessarily the best outcome for the person.

Yet discharge based solely on medical stability can transfer risk into the home.

A person may be physically weaker than before admission. Their cognitive state may have changed. Family members who previously provided limited help may now be expected to deliver substantially more. The home environment may need equipment. Existing municipal services may no longer provide enough support.

Good discharge therefore depends on understanding functional and social need alongside medical status.

This connects Latvia's system with the broader hospital and home-care interface. The principle is simple but operationally demanding: the receiving environment needs to be capable of supporting the person who is actually leaving hospital, not the person described in an older social-care assessment.

Scenario: the previous care package no longer fits after hospital treatment

A woman in her late seventies from a municipality outside Riga is admitted to hospital following pneumonia and a fall. Before admission she received limited home care and her son visited several times a week.

She is medically ready for discharge after treatment, but she now requires help transferring, tires rapidly and is unable to prepare meals independently.

If discharge planning simply reinstates her previous social-care arrangement, the formal service resumes but the package no longer reflects her needs. Her son may try to fill the gap temporarily, increasing the risk that an unsustainable informal arrangement becomes the default.

A stronger pathway requires changed functional needs to become visible to the municipal social service before or immediately after discharge. The service can then reassess what level and form of support is required, while healthcare professionals remain responsible for ongoing clinical needs and any home healthcare that is indicated.

The quality measure is not only the date she leaves hospital. It includes whether she remains safely at home, whether deterioration is recognised and whether unplanned readmission is avoided where possible.

This kind of transition illustrates why integration is fundamentally about continuity across a change in setting.

Assessment remains divided even when the person is one

Health and social care assess different things for legitimate reasons.

Healthcare professionals assess diagnosis, treatment needs, symptoms, rehabilitation potential and clinical risk. Municipal social services assess social circumstances, functional ability, existing support and the need for social-care or rehabilitation services.

The problem is duplication without connection.

A person and family may repeatedly explain the same history to different professionals while each system generates separate information. Important details can be lost because one assessment does not automatically inform another.

Integration therefore does not necessarily require one universal assessment replacing all specialist assessment. A more realistic goal is a common core of information combined with profession-specific analysis.

That core might include the person's goals, functional status, living situation, significant risks, informal support, current services and agreed coordination arrangements. Clinical information can then sit alongside social-service information without forcing either profession to abandon its expertise.

This approach also reduces the burden on people with complex needs, who are often least able to navigate repeated assessments.

Long-term institutions sit permanently at the health-social boundary

Latvia's long-term social care and social rehabilitation institutions provide accommodation, social care and social rehabilitation for people whose needs exceed what can be provided through home care, day services or other community arrangements.

Residents may also have substantial healthcare needs.

Latvian law recognises this interface. Long-term institutions may implement treatment plans determined by healthcare professionals and may establish organisational units for the provision of healthcare services.

This is a practical acknowledgement that social-care institutions cannot be clinically isolated from the healthcare system.

But the distinction in responsibility remains important. A care institution does not become a substitute hospital simply because residents have complex health needs. Healthcare professionals remain responsible for clinical diagnosis and treatment. Social-care workers need competence to observe change, follow authorised plans within their role and escalate when clinical input is required.

The quality of integration is therefore visible in how quickly the institution can access healthcare, whether relevant information is shared and whether avoidable transfers to hospital can be prevented without creating unsafe attempts to manage clinical deterioration inside a social-care setting.

Primary care can become a critical continuity point

Family doctors occupy an important position because many people with long-term-care needs have chronic health conditions managed outside hospital.

In Latvia, family doctors can refer or contribute to home healthcare pathways and continue clinical oversight for many community-based patients. They may therefore become one of the few healthcare professionals with a longitudinal view of the person's health.

Yet continuity of health information does not automatically equal continuity of whole-person care.

A family doctor may know the patient's diagnoses and medicines but not know that a spouse providing daily care is becoming exhausted. A municipal social worker may know that the home situation is deteriorating but not know that the person's clinical condition has changed substantially.

Integration benefits from clearer routes through which relevant social information can reach primary care and relevant health information can influence social-service planning.

This is particularly important for people with dementia, frailty and multiple chronic conditions, where the boundary between a social problem and a health problem is often blurred in early stages.

A decline in eating may reflect depression, swallowing difficulty, dental problems, cognitive deterioration, poverty or several factors together. The correct response depends on which part of the picture is visible.

Care coordination is a function, not merely a job title

Integrated systems often respond to fragmentation by creating a coordinator role. That can be valuable, but coordination should not depend entirely on one individual.

Someone still needs responsibility for ensuring that the overall plan makes sense. But the surrounding system must support that function.

A coordinator cannot compensate indefinitely for inaccessible records, unclear eligibility rules, incompatible provider contracts or professionals who do not know when to share information.

Latvia's emerging integrated-care models therefore need to define coordination operationally.

At minimum, somebody should know:

• which health and social-care services are involved;
• what each service is responsible for;
• what changes require escalation;
• how information moves between professionals;
• who updates the overall plan when needs change; and
• who the person or family contacts when the system no longer appears to fit together.

This connects strongly with care coordination and continuity. Although the linked theme is rooted in mental-health services, the operational principle applies widely: complex support becomes safer when responsibility for coordination is explicit rather than assumed.

Scenario: complexity creates duplication before it creates failure

A 59-year-old man with a physical disability, diabetes and recurring depression receives municipal social support, primary healthcare and occasional specialist treatment. His sister helps informally but does not live nearby.

Over time, three different professionals independently ask about his daily functioning. Each records useful information, but the records do not produce one coherent plan.

His social worker knows that he has recently stopped attending community activities. His family doctor knows his diabetes control has deteriorated. A mental-health professional knows he has become increasingly withdrawn. No single piece of information appears urgent in isolation.

Together, the pattern suggests significant deterioration.

An integrated response brings those observations into one review. The aim is not to make every professional responsible for everything. It is to ensure that relevant information changes the collective understanding of risk and support.

The result may involve adjustment to social support, closer clinical follow-up and practical action to restore routine and community participation.

This is where integration adds value before a crisis. It reveals relationships between information that would otherwise remain technically accurate but operationally disconnected.

Funding fragmentation can reinforce organisational boundaries

Health and social care are funded through different routes in Latvia.

State-funded healthcare operates through national health financing and provider arrangements. Social services are significantly shaped by municipal responsibility, with state funding or co-financing applying to particular services and population groups.

Different funding streams are not inherently problematic. Problems emerge when organisations have incentives to protect their own budgets by shifting cost elsewhere.

For example, insufficient community social care can contribute to prolonged hospital stays or readmissions. Inadequate healthcare access within residential services can lead to avoidable emergency transfers. Family carers may absorb gaps between both systems at no direct cost to either formal budget.

This makes integration partly a financial-governance issue.

Decision-makers need to understand where expenditure in one sector may reduce pressure in another. That does not mean every preventive social service will produce immediate cash savings for healthcare. It means system value should not be assessed solely through the budget of the organisation paying for the intervention.

Organisations examining cross-system evidence can use the Commissioner Evidence Builder as a general framework for structuring evidence expectations across service relationships. It is not a Latvian purchasing instrument, but its relevance lies in making outcomes, responsibilities and evidence visible where several organisations contribute to one pathway.

Family carers remain the hidden integration layer

Where formal systems do not connect smoothly, families often compensate.

A daughter carries a discharge summary to a municipal office. A spouse explains medication changes to care workers. A son telephones several providers to reconcile appointment times. Families become information brokers, transport organisers, interpreters of professional advice and out-of-hours monitors.

This work can make fragmented systems appear more functional than they are.

It also creates inequality. People with confident, available relatives may navigate complexity more successfully than those who live alone, have family abroad or have relationships that cannot safely provide support.

Integration should therefore reduce rather than institutionalise dependence on family coordination.

Family knowledge remains valuable. Relatives often understand the person's history, routines and early signs of deterioration better than any one professional. But involving family should mean partnership, not transferring responsibility for making the system work.

This is especially important in Latvia because informal care continues to support a substantial proportion of long-term need. As demographic change reduces the number of available family carers in some households, weaknesses in formal coordination may become more visible.

Workforce integration requires role clarity, not role dilution

Integrated care needs professionals to work across boundaries without losing the expertise that made those boundaries necessary.

Social workers do not become nurses because a person has health needs. Care workers should not make clinical decisions beyond their competence. Healthcare professionals should not assume that functional and social problems can be solved through medical treatment alone.

The workforce challenge is therefore collaborative competence.

Workers need to understand the contribution of other professions, recognise the limits of their own role and know how to escalate when a need moves beyond their competence.

Training should include not only technical skills but understanding of pathways and interfaces.

A home-care worker should know how to report a clinically concerning change. A hospital professional involved in discharge should understand that social-care capacity is not automatically available because a need has been identified. A social worker coordinating complex support should understand which issues require healthcare assessment rather than attempting to solve them through social services.

These are simple principles, but integration often succeeds or fails through exactly these everyday decisions.

Information sharing is the infrastructure underneath coordination

Professionals cannot coordinate effectively if relevant information remains inaccessible.

Latvia's health and social-care systems have different records, purposes and information-governance responsibilities. Complete access to every record is neither necessary nor appropriate.

What matters is whether the information required for safe continuity can move.

A hospital discharge needs enough information for social services to understand changed functional needs. A care provider needs current instructions relevant to the support it is expected to deliver. Healthcare professionals may need to know that the person's home circumstances have deteriorated or that carers are no longer able to cope.

This is the practical value of digital records and data. The goal is not one enormous record containing every detail about a person's life. It is a coherent information architecture in which authorised professionals can access the information necessary for their responsibilities.

Data minimisation and integration are not opposites. Good systems share enough to support care while preserving privacy and accountability.

Digital integration needs workflow redesign as well as technology

Interoperability is often described as a technical problem: can one system exchange information with another?

That is only the first layer.

Even technically interoperable systems can fail operationally if professionals do not know which information they should review, alerts have no clear recipient or records are duplicated because organisations do not trust each other's data.

Latvia's future integration therefore needs workflow design alongside digital infrastructure.

The Digital Transformation Readiness Assessment can help organisations examine whether strategy, workforce, governance and digital capability are aligned. It is not specific to Latvia, but its core principle is highly relevant: technology cannot repair an unclear operating model by itself.

For integrated care, digital readiness includes agreed data standards, role-based access, clear responsibility for updating information, staff confidence in digital tools and contingency arrangements when systems fail.

The best digital interface is one that reduces repeated explanation for the person while improving rather than increasing professional workload.

The Riga pilot should be judged on experience as well as activity

Because the 2026–2027 Riga integrated-care-at-home pilot is testing a new model, evaluation will be especially important.

Counting the number of people enrolled, visits delivered or providers participating will establish scale. Those measures will not show whether integration worked.

Useful evaluation should examine whether people experience fewer duplicated contacts, whether one coordinated plan is genuinely used, whether changes in need are acted upon more quickly and whether families spend less time mediating between services.

Other relevant measures may include unplanned hospital use, continuity, workforce time, service delays, complaints, staff experience and the sustainability of the model's funding and contractual structure.

The Quality Dashboard Builder provides a general way of structuring this type of multidimensional evidence. It does not define the evaluation requirements of Latvia's pilot, but it illustrates why activity, quality, workforce and outcomes should be considered together.

The most valuable lesson may also be what does not work. Pilots are useful precisely because they allow operational assumptions to be tested before wider implementation.

Scenario: one-provider integration reduces navigation but reveals a new capacity risk

A participant in the Riga pilot has complex mobility limitations, chronic disease and significant personal-care needs. Under previous arrangements, her daughter coordinated separate healthcare-at-home and municipal home-care providers.

Within the pilot, one provider coordinates both components. Information exchange improves, appointments are easier to manage and the daughter has one clearer point of contact.

Several months later, however, the integrated provider experiences staffing pressure. Because the organisation now carries a wider part of the pathway, workforce disruption affects both health and social components simultaneously.

This does not mean the model has failed. It reveals a different governance question: integration can reduce fragmentation while also concentrating dependency.

The provider and public bodies therefore need continuity arrangements, clear escalation and sufficient workforce resilience. Evaluation should recognise both the improvement in coordination and the new operational exposure.

This is an important principle for wider reform. Integration changes risk; it does not eliminate it.

Regional variation will make national expansion more difficult than urban piloting

Riga provides a logical environment in which to test integrated home care because it has a large population, substantial health infrastructure and a comparatively dense provider market.

Scaling the same principles across Latvia would raise different questions.

Rural municipalities face longer travel distances, smaller labour markets and potentially fewer providers able to deliver both health and social-care components. Specialist services may be concentrated farther from the person's home.

A nationally consistent objective therefore may require regionally different operating models.

One area might sustain a provider delivering both components directly. Another may require formal partnerships between separate health and social-care organisations. Remote professional advice could extend specialist reach, while local workers provide hands-on support.

The principle should be continuity, not identical organisational form.

This distinction is important because integrated-care reform can become overly focused on structures. The purpose is not to reproduce the Riga pilot mechanically in every municipality. It is to identify which elements of coordination are essential and then determine how they can be delivered in different local contexts.

Governance needs to own the gaps between organisations

Integration creates a familiar governance paradox. Every organisation can have clear internal responsibility while the shared boundary remains weak.

A hospital is responsible for discharge planning within its remit. A municipality is responsible for organising social services. A home-healthcare provider is responsible for clinical care. A social-care provider is responsible for its assigned support.

Yet who is accountable when the overall pathway does not make sense?

This is where cross-system governance becomes essential.

Organisations using the Governance Maturity Assessment can examine whether accountability and escalation remain clear across complex arrangements. It is not designed around Latvian administrative law, but the underlying issue applies directly: shared care needs explicit governance of shared interfaces.

Good governance asks:

• who has operational responsibility for coordination;
• which organisation acts when needs change;
• how unresolved disagreements are escalated;
• what evidence shows that the pathway is working;
• where repeated failures are reviewed; and
• which level has authority to change funding, service design or contractual arrangements when the problem exceeds frontline control.

The final point matters especially. A worker cannot solve a structural funding gap through better communication. A provider cannot create regional specialist capacity through goodwill. Integration needs escalation routes that reach the level capable of changing the underlying constraint.

Quality should be assessed across transitions, not only within services

Latvia's quality-assurance architecture can become stronger by examining transitions as outcomes in their own right.

Traditional assurance often measures each provider separately. Yet people experience the gaps between providers most acutely.

Useful pathway indicators can include:

Whether relevant information arrives before or with the person. Whether social support is ready at discharge. Whether healthcare teams receive important observations from care workers. Whether people are repeatedly assessed for the same information. Whether responsibility becomes unclear after deterioration. Whether family carers understand whom to contact.

This connects integrated care with decision-making and escalation. Integration is not demonstrated by frequent meetings. It is demonstrated when the right decision is made by the right person at the right time, using information from the relevant parts of the pathway.

Integration should reduce complexity for the person, not simply move it behind the scenes

One of the strongest tests of integration is whether the system becomes easier to navigate.

People with complex needs should not need a detailed understanding of ministerial responsibilities, municipal funding rules, provider contracts and professional boundaries in order to receive coherent support.

Those distinctions may remain necessary internally. Integration should manage them behind the service experience.

This does not necessarily mean one telephone number or one organisation in every case. It means the person should know what happens next, who is coordinating the current plan and where to go when something changes.

Communication therefore becomes part of integration.

Accessible information is particularly important for people with cognitive impairments, communication difficulties or limited digital confidence. Families may support navigation where the person wants their involvement, but formal systems should not assume that every person has somebody able to act as an unpaid case manager.

The next phase of reform should focus on repeatable coordination, not exceptional collaboration

Healthcare and social-care professionals in Latvia already collaborate successfully in many individual cases. Strong personal relationships between workers can overcome organisational barriers.

But systems cannot depend on exceptional individuals knowing whom to telephone.

Mature integration turns good informal practice into repeatable infrastructure.

That means defined pathways, agreed information exchange, clear escalation, service specifications that recognise coordination time and workforce development that prepares professionals for cross-system working.

The 2026–2027 Social Services Improvement and Development Plan provides an important policy signal by explicitly strengthening cooperation between social-service providers and health, education and municipal structures. The Riga home-care pilot provides a practical opportunity to test a more integrated model.

The strategic challenge is to learn from both without describing integration as already complete.

The stronger direction is incremental but structural: identify high-risk interfaces, test models, measure outcomes, retain what works and build stronger coordination into standard service architecture over time.

What Latvia’s experience offers internationally

Latvia's health financing, municipal social-service responsibilities and provider arrangements are specific to its own institutional system. Other countries may integrate care through insurers, regional governments, pooled budgets or dedicated long-term-care legislation.

The transferable lessons lie in the underlying operating problems.

First, separate professional systems are not inherently fragmented. Fragmentation occurs when their interfaces are weak. Clear role boundaries can coexist with strong coordination.

Second, home-based care makes integration visible because clinical and everyday support occur in the same physical environment. Designing around the home can expose organisational divisions that institutional settings previously concealed.

Third, pilots should test the operating model rather than merely demonstrate activity. Integration should improve continuity, experience and decision-making, not simply place multiple services under one administrative label.

Fourth, families often hide fragmentation by coordinating care themselves. Systems should measure this burden rather than interpreting family success as proof that formal coordination is adequate.

Finally, integration requires governance of the spaces between organisations. Shared responsibility is valuable only when it does not become ambiguous responsibility.

Conclusion

Latvia's health and social-care systems have different purposes, legal frameworks and funding routes, but an increasing number of people need them simultaneously. Population ageing, chronic illness, disability and the expansion of community-based support make the quality of the interface between those systems increasingly important.

The country's current direction is significant. The Ministry of Health explicitly recognises integrated healthcare and cross-sector cooperation, the 2026–2027 social-services plan strengthens collaboration across health and social structures, and the Riga integrated-care-at-home pilot is testing a more coordinated service model for adults with complex needs. These developments should be understood as steps in integration rather than evidence that a fully integrated national system already exists.

The next challenge is operational: align assessment without erasing professional expertise, move relevant information safely, clarify coordination, connect hospital discharge with municipal capacity, protect families from becoming default system navigators and ensure that funding arrangements do not reward cost-shifting between sectors.

Integration ultimately succeeds when organisational complexity becomes less visible to the person. Latvia does not need every health and social-care organisation to become one institution. It needs the boundaries between them to become sufficiently well governed that people experience continuity rather than fragmentation. That is the stronger route from policy cooperation towards an integrated long-term-care system that works where it matters most: in the daily lives of people with complex and changing needs.