A national care system can know how many people receive a service without knowing whether the right services exist in the right places. It can count workers without understanding whether workforce capacity matches changing demand. It can record expenditure without being able to show whether people's independence, wellbeing or participation have improved.
That distinction is becoming increasingly important in Uruguay. As the Sistema Nacional Integrado de Cuidados (SNIC) moves toward wider coverage and stronger quality, information is no longer simply an administrative requirement. It is part of the infrastructure needed to decide where services should grow, where inequalities persist and whether the right to care is becoming more meaningful in practice. The wider Uruguay Aging, Long-Term Care & Community Support Knowledge Hub examines how these questions connect with the country's broader long-term care reforms.
The National Care Plan 2026–2030 makes this explicit. One of its four strategic objectives is to generate and make available timely, high-quality information and knowledge for decision-making and reflection on public care policy. Proposed developments include consolidating the National Care Registry, improving interoperability, strengthening care indicators and producing evidence that can inform policy across the life course.
The central challenge is therefore changing. Uruguay needs more complete data, but completeness alone will not create intelligence. The stronger opportunity lies in connecting information about people, services, workers, territories, unpaid care, quality and outcomes so that it answers practical questions. Where is demand increasing? Who is not reaching support? Why does access vary? Which services are producing value? Where is workforce capacity becoming fragile? What needs to change if the same problem keeps recurring?
Those are questions of intelligence rather than data collection.
SNIC needs evidence because universality cannot be managed through national averages
Uruguay created SNIC through Law No. 19.353 as an integrated system intended to progressively guarantee the right to care. As that ambition expands, national totals become increasingly insufficient for management.
A figure showing growth in Personal Assistant users, for example, says little about whether eligible people in different departments can obtain workers. The number of Day Center places does not reveal whether transport prevents people from attending. The number of trained care workers does not establish whether they are located where services need them. A national average waiting period can conceal much longer delays for particular populations or territories.
This is why data-led equity planning matters to a rights-based care system.
Equality in a national policy does not mean every locality requires identical provision. Montevideo, departmental capitals, smaller towns and dispersed rural communities have different population density, workforce markets, transport conditions and service infrastructure. A community with fewer residents may need a different delivery model precisely because a conventional service is difficult to sustain at low volume.
Good intelligence therefore helps distinguish variation that reflects legitimate local adaptation from variation that produces unequal access.
That distinction cannot be made from one indicator. It requires information about population need, service availability, actual uptake, workforce capacity and people's experience.
The National Care Registry could become the spine of care intelligence
Uruguay's current National Care Plan gives renewed importance to the Registro Nacional de Cuidados. The intended direction is toward an integrated record containing information on people using care, caregivers and services, supported by links with information already held across public institutions.
The strategic importance of such a registry extends beyond administration.
If developed effectively, it could provide a more coherent view of who receives support, what type of support is provided and where. When connected appropriately with other information, it could help SNIC understand service trajectories rather than isolated transactions.
That could mean distinguishing between somebody who successfully enters a service and somebody whose approved arrangement never becomes stable. It could reveal repeated interruptions. It could identify movement between programs as needs change. It could help determine whether people with similar levels of dependency experience different access according to age or territory.
But the registry should not be mistaken for a complete intelligence system.
Administrative records primarily describe interactions with the system. People who never reach SNIC may be largely invisible within them. Unpaid care may be underrepresented. Emerging need that has not generated an application may not appear. Quality of life, autonomy and caregiver strain cannot be inferred reliably from service enrollment.
The registry is therefore best understood as a foundational dataset that must be combined with surveys, population statistics, evaluations, qualitative evidence and service-level information.
Good intelligence starts with knowing what each dataset can actually tell you
Care systems often encounter a deceptively simple problem: data collected for one purpose is subsequently used to answer a different question.
A payment system is designed to administer financial transactions. A service register records participation. A dependency assessment supports eligibility and care decisions. A population survey estimates characteristics across society. A complaint system captures concerns raised by people who choose or manage to report them.
Each can contribute to intelligence, but they are not interchangeable.
For example, an increase in the number of people receiving a care service could reflect rising need, wider eligibility, improved awareness, additional capacity or reduced barriers to application. Administrative growth alone cannot identify the cause.
Similarly, fewer complaints do not necessarily mean higher quality. They could reflect improved services, but they could also indicate inaccessible complaints processes or reluctance to raise concerns.
This makes data collection and data quality a governance issue. Decision-makers need to understand the provenance, definitions, limitations and completeness of the evidence they use.
Organizations examining similar information environments can use the Quality Dashboard Builder to structure relationships between indicators, operational signals and management decisions. It is not an official SNIC instrument; its relevance lies in helping distinguish a useful decision framework from a collection of disconnected metrics.
Time-use evidence reveals care that service systems cannot see
One of Uruguay's important evidence strengths is its measurement of unpaid work and time use.
The Encuesta de Uso del Tiempo conducted by the Instituto Nacional de Estadística (INE) and the Ministerio de Desarrollo Social (MIDES) measures how people divide time between paid and unpaid activities, including care. The 2021–2022 survey documented substantial gender differences in unpaid work and provides information that administrative service records could never capture adequately.
This matters because SNIC exists partly to change the social organization of care, not merely to administer formal services.
If public provision expands while women continue to absorb most additional care created by demographic change, service statistics alone could give an incomplete impression of progress. Conversely, formal support that reduces unpaid-care intensity may produce benefits that are not visible if performance is measured only through the number of service hours delivered.
In February 2026, INE launched an interinstitutional forum on measuring time use and care, bringing together organizations involved in care and gender equality. Its purpose includes strengthening statistical measurement for the design, monitoring and evaluation of public policy.
This creates an important connection between population statistics and care-system intelligence.
Care policy needs to know not only what the State provides but what households continue to provide around it.
Scenario: service coverage rises while family pressure remains hidden
A department records an increase in older people receiving formal support. From a service-coverage perspective, the direction appears positive.
Local conversations with families, however, suggest that adult daughters are still providing extensive evening, weekend and coordination support around formal services. Some have reduced employment hours. Others are managing appointments, medication collection, transport and supervision that are not reflected in the formal care record.
The administrative data and family experience are not contradictory. They measure different parts of the care arrangement.
A stronger intelligence approach combines them. Service data establishes what formal support has expanded. Time-use and household evidence helps show whether unpaid-care intensity is changing. Qualitative research explains which responsibilities families continue to carry and why.
The resulting policy question is more useful than asking whether service coverage increased. It becomes: has the overall distribution of care changed in the way the national policy intended?
If formal support is simply added on top of substantial unpaid work, further service redesign, respite or navigation support may be needed. If family burden decreases while the person's independence improves, that provides evidence of wider system value.
Measuring the care economy therefore requires looking beyond the boundary of publicly administered services.
Indicators need to connect activity, quality, equity and outcomes
The National Care Plan proposes developing a system of indicators capable of supporting monitoring and evaluation across the care agenda.
The design of that indicator system will matter considerably.
Activity measures remain essential. SNIC needs to know how many people receive Personal Assistants, Telecare or Day Center support, how many workers are trained and how public resources are used. Without reliable activity data, basic administration and planning become difficult.
But a mature intelligence system needs several layers of evidence:
- Need: dependency, demographic change, disability, household circumstances and unpaid-care demand.
- Access: applications, eligibility, waiting, service commencement, geography and population differences.
- Capacity: workforce, providers, places, hours, infrastructure and territorial availability.
- Quality: continuity, safety, experience, complaints, service reliability and person-centered practice.
- Outcomes: autonomy, participation, wellbeing, caregiver effects and changes in functional support needs where measurement is appropriate.
- Sustainability: expenditure, workforce stability, service utilization and whether resources are aligned with changing demand.
No single dashboard needs to contain every measure. Different levels of SNIC require different views. Operational teams need information that helps them act now. National governance needs enough aggregation to identify patterns and allocate attention without losing important local variation.
This is the purpose of outcomes frameworks and indicators: not to maximize the number of measures, but to connect evidence with the questions the system is responsible for answering.
Territorial intelligence can turn national policy into local planning
Uruguay's geography makes territorial analysis particularly important.
Population size, density, age structure and service infrastructure differ between departments. Workforce availability and travel times can also shape whether a theoretically available service is practical.
A national care strategy therefore needs a way to translate broad entitlements into realistic territorial capacity.
Suppose population projections indicate that the number of older people will rise substantially in a particular area. That information becomes more useful when considered alongside dependency patterns, current service uptake, worker availability, transport, residential capacity, Day Center provision and existing family-care intensity.
The resulting intelligence can support decisions about whether the next investment should be a physical service, additional workforce development, mobile provision, transport support, technology-enabled care or some combination.
This is different from allocating resources purely according to historic utilization. Historic utilization may partly reflect what was available rather than what people actually needed.
Population-based planning can expose that difference.
Organizations considering similar capacity questions can use the Digital Twin Scenario Modeler to explore relationships between demand, workforce and service stability. It does not model Uruguay's national system, but the underlying principle is relevant: intelligence becomes more valuable when it helps leaders test future capacity before pressure becomes an operational failure.
Scenario: national growth conceals a territorial access gap
SNIC data shows that use of a community-based service has increased nationally over three years. The headline suggests improving access.
When analysts compare coverage with population need, however, one group of interior departments appears consistently below the national pattern. The first interpretation might be lower demand.
Further analysis tests that assumption.
Dependency and demographic indicators do not suggest substantially lower need. Applications are being made, but fewer progress rapidly into stable service arrangements. Workforce information shows a smaller pool of available workers, while local interviews identify travel and scheduling difficulties.
The intelligence changes the policy response.
A national awareness campaign would be unlikely to solve the problem because people already know about the service. Changing eligibility would not address it either. The constraint sits on the supply side.
Government can then consider a territorial response: targeted training, different organization of worker hours, collective provision, travel support or alternative community models. Subsequent data can test whether the intervention changes actual service commencement and continuity.
The important point is not that data automatically determines the answer. It helps decision-makers ask a better question, locate the constraint and monitor whether the response changes it.
Workforce intelligence needs to connect people, skills and geography
Uruguay's care workforce agenda is becoming more structured through training, competency recognition and the emerging qualifications framework. Information about workers needs to mature alongside it.
Knowing the total number of trained workers is useful but insufficient.
Planning needs to understand where workers are located, whether they are active in care employment, which roles they perform, what skills they hold, whether employment is stable and where recruitment or retention problems are affecting service continuity.
This is particularly important because care capacity is not perfectly transferable. A worker in Montevideo cannot solve a shortage hundreds of kilometers away. A person with foundational dependency-care training may still require additional support for complex cognitive, behavioral or health-related needs. Headcount therefore differs from usable capacity.
Over time, stronger links between workforce, training and service information could support workforce data and capacity planning.
Such analysis also needs care around interpretation. High turnover may indicate poor employment conditions, but it can also reflect local labor-market opportunities or the design of particular roles. Data can identify the pattern; workforce research and worker participation help explain it.
The strongest intelligence model therefore combines quantitative workforce measures with knowledge from the people doing the work.
Evaluation needs to ask whether services change people's lives
Monitoring and evaluation are related but different functions.
Monitoring asks whether a service is operating as expected. Evaluation asks whether it is producing the intended effects, for whom, under what conditions and at what cost.
Uruguay has already generated useful service-specific evidence. Evaluation of Day Centers, for example, has examined participant and family experience and reported very high satisfaction alongside perceived effects on quality of life and dependency. Evidence of this kind adds meaning to administrative counts of centers and participants.
Future evaluation can go further by examining variation between people and territories, changes over time and the interaction between formal services and family support.
A Personal Assistant arrangement may be successful because it enables somebody to study, work or participate in community life. Telecare may provide value through confidence and continued independence even when emergency activations are infrequent. A Day Center may affect both the participant's social connection and the family caregiver's ability to maintain employment.
These effects are not captured adequately through utilization alone.
The Community Impact Report Builder can help organizations working on comparable services structure evidence about outcomes, community effects and qualitative experience. It is not a Uruguayan government evaluation tool, but it illustrates an important principle: service evidence becomes stronger when operational activity is connected with the changes people experience.
Person-reported evidence protects intelligence from becoming purely administrative
National information systems naturally favor data that institutions can collect consistently. That creates a risk that the system becomes very good at measuring itself while knowing less about how people experience care.
Person-reported and family evidence provides a necessary counterweight.
A service can meet its scheduled hours while the person experiences frequent changes of worker. A Day Center can record attendance while activities feel poorly matched to participants' interests. A technically successful application process can still be difficult for somebody with sensory or cognitive impairment.
These are not soft issues sitting outside performance management. They are evidence about whether the service model is functioning as intended.
Uruguay's care governance already contains mechanisms for participation through structures including the Comité Consultivo de Cuidados, while individual services can generate user and family feedback. The analytical challenge is connecting those voices with decisions rather than treating consultation as a parallel process.
Qualitative evidence can help explain unexpected quantitative patterns, identify issues before they become visible in formal metrics and test whether national definitions of quality match lived experience.
This is why case studies and qualitative evidence have a legitimate place alongside administrative statistics.
Scenario: the performance indicator is green, but people's experience is deteriorating
A home-support arrangement is reviewed using conventional operational indicators. Scheduled support hours are being delivered, payments are correct and there are no major recorded incidents. The service therefore appears stable.
Interviews with people using the service tell a different story.
Several people report frequent worker changes. Visits occur, but workers sometimes arrive without knowing established preferences. Families increasingly leave written instructions because continuity has weakened. Nobody has experienced a serious safety event, so the deterioration is not visible in incident statistics.
A broader evidence set changes the interpretation.
Continuity of worker, unplanned substitutions, user-reported confidence and repeated complaints can be considered alongside delivered hours. Workforce data may then show that turnover has increased locally.
The problem is no longer invisible simply because the service technically occurred.
Managers can examine why continuity is deteriorating and whether scheduling, employment conditions, recruitment or supervision are contributing. National teams can determine whether the same pattern appears elsewhere.
If it does, what first looked like a local experience issue may require a wider workforce or service-design response.
The scenario demonstrates why complaints and other user signals are valuable when incorporated into intelligence rather than considered only as individual cases to be closed.
Data quality becomes more consequential when data drives resources
The more important information becomes to planning, the greater the consequences of poor data.
Missing records, inconsistent definitions, duplicate people, outdated service status or different coding between institutions can distort conclusions. If resource decisions subsequently rely on those conclusions, technical data problems become distributional problems.
Suppose one department records service interruption consistently while another records only permanent service closure. The first may appear less stable even if actual performance is similar. Comparing the figures without understanding the definitions would create misleading intelligence.
Data-quality governance therefore needs to include more than correcting errors after publication.
Definitions need ownership. Validation should occur close enough to data entry that obvious problems can be resolved. Changes to measures need version control so trends remain interpretable. Users of dashboards need enough metadata to understand what the numbers represent.
Where several institutions contribute information, responsibility for resolving inconsistencies must also be clear.
The development of SNIC's information architecture offers an opportunity to establish these controls as the system grows rather than attempting to repair fragmented datasets later.
Accountability requires transparency without simplistic league tables
Better information can strengthen public accountability, but publication needs thoughtful design.
Citizens should be able to understand whether the care system is expanding, where resources are going and whether major policy commitments are being implemented. Researchers, civil society, workers and the Comité Consultivo can contribute more effectively when relevant evidence is accessible.
Transparency also creates discipline inside government. Public measures make it harder for persistent gaps to remain invisible.
But accountability should not be reduced to ranking territories or services using indicators that ignore context.
A department serving a dispersed population may face different delivery conditions from Montevideo. A service supporting people with more complex needs may show different patterns from one serving a less intensive population. Raw comparisons can create false conclusions unless population and service differences are considered.
The stronger approach is explanatory transparency: publish what is measured, show relevant variation, explain definitions and provide enough context for the figures to be interpreted responsibly.
This connects directly with data governance and information accountability. Public information should increase understanding rather than merely increase the volume of numbers available.
Intelligence only matters when it changes a decision
Care systems can invest heavily in dashboards and reports while leaving operational decision-making almost unchanged.
The test of intelligence is therefore not whether information is produced. It is whether there is a route from evidence to action.
Different evidence should reach different decision levels.
A local service manager may need to act immediately on repeated missed support. A territorial team may need to respond to growing waiting times or workforce shortages. National SNIC leadership may need to change funding, service design or training capacity if the same pattern appears across several departments.
That requires an operating rhythm in which information is reviewed, interpreted, assigned and followed through.
Organizations examining similar arrangements can use the Governance Maturity Assessment to test whether information, responsibility and oversight are connected. It does not replicate Uruguay's governance structure, but the underlying question is directly relevant: when evidence shows a persistent problem, who is responsible for deciding what happens next?
If no answer exists, the system has reporting rather than intelligence.
Funding decisions need evidence about both cost and value
Uruguay's ambition to widen the right to care operates within real fiscal constraints. The National Care Plan itself recognizes sustainability as an important challenge.
Data can improve the quality of funding decisions, but only if cost is interpreted alongside outcomes and system effects.
A community service may appear expensive when assessed solely against its operating budget. Its wider value may include sustaining independence, supporting family employment, reducing isolation or delaying demand for more intensive support. Conversely, a low-cost service that people cannot access reliably may offer limited value despite modest expenditure.
This does not mean every wider benefit can be converted credibly into a monetary saving. Care has intrinsic social and rights-based value as well as economic consequences.
It does mean funding analysis should avoid treating unit cost as the complete measure of performance.
Over time, linking expenditure with service utilization, population need, outcomes and territorial variation could support more informed discussion of cost and outcomes.
Such evidence will become increasingly important as demographic change increases demand and policymakers decide how rapidly different parts of SNIC can expand.
Forecasting should support preparedness rather than claim certainty
Better historical data creates opportunities for forecasting.
Population projections can indicate how age structures may change. Service trends can help estimate future demand. Workforce data can identify where supply may become constrained. Scenario modeling can examine what happens if eligibility, utilization or staffing assumptions change.
These tools can improve preparedness, but care forecasting contains unavoidable uncertainty.
Future demand depends not only on population aging but on disability prevalence, health, housing, family structures, prevention, technology, service expectations and policy choices. Increasing the number of older people does not translate mechanically into an identical percentage increase in formal care.
Forecasts should therefore expose assumptions rather than conceal them.
A useful planning model might present several plausible demand scenarios and show what each would mean for workforce, funding and service capacity. Leaders can then identify decisions that remain sensible across several futures.
This approach is more valuable than producing a single apparently precise prediction that becomes obsolete as circumstances change.
Intelligence supports judgment; it does not remove uncertainty from long-term care planning.
Research partnerships can strengthen the national learning system
SNIC does not need to generate every piece of knowledge internally.
Uruguay has a strong academic and statistical environment, and the participation of academia within the Comité Consultivo creates opportunities to connect research with policy questions. INE provides population and household evidence. Universities and research organizations can undertake deeper evaluation. International organizations can contribute comparative methods and technical cooperation.
The important governance issue is whether research questions reflect genuine policy and operational uncertainties.
If government needs to understand why a service reaches one population less effectively, evaluation can be designed around that question. If workforce reforms are introduced, longitudinal research can examine whether training and employment stability improve. If a new community model is piloted, evidence can be collected early enough to inform decisions about expansion.
This creates a learning system rather than a sequence of disconnected studies.
Research independence also matters. Evidence is more credible when evaluation is capable of identifying limitations as well as successes. A rights-based national system should be able to learn from results that challenge its assumptions.
Four questions can keep Uruguay’s intelligence agenda focused
As the information architecture develops, complexity will increase. New systems can generate hundreds of potential indicators. The discipline lies in retaining focus on the questions that matter.
At national and territorial levels, much of the intelligence agenda can ultimately be tested against four broad questions:
- Are people who need care able to obtain appropriate support?
- Is the available service and workforce capacity sufficient, equitable and sustainable?
- Does support improve autonomy, wellbeing, participation, safety and the distribution of care?
- When evidence shows that the answer is no, does governance change policy, resources or delivery?
These questions connect rights with operations.
They also prevent information strategy from becoming an end in itself. The National Care Registry, surveys, dashboards, evaluations and interoperability all become means of understanding whether the care system is achieving its purpose.
International learning is about building intelligence alongside the care system
Uruguay's experience has relevance beyond its institutional structure.
Many established long-term care systems accumulated information separately across health, social protection, local government, insurers and providers. Attempts to create integrated intelligence later can be difficult because definitions, incentives and technology have already diverged.
Uruguay has an opportunity to strengthen its information architecture while SNIC is still developing toward greater universality.
The transferable lesson is not that other countries need a Uruguayan-style National Care Registry. Administrative structures, privacy law and funding systems differ too substantially for direct replication.
The stronger lesson is that care-system development and intelligence-system development should occur together.
Eligibility rules should generate information that can be evaluated. New services should be designed with outcome measures rather than adding them later. Workforce expansion should be visible territorially. Public reporting should evolve alongside national commitments. People using services should contribute evidence about whether formal policy is producing meaningful support.
In this sense, information is not simply the record left behind by care. It is part of how a care system learns.
Conclusion
Uruguay's next care challenge is not simply to collect more information. It is to create a dependable line of sight between population need, unpaid care, formal services, workforce capacity, quality, outcomes, expenditure and the decisions made in response.
The National Care Plan 2026–2030 provides an important foundation by placing information and knowledge among its four strategic objectives. Consolidating the National Care Registry, strengthening indicators, improving interoperability and developing statistical measurement can give SNIC a much richer understanding of the system it is trying to expand. But administrative information alone will never be sufficient. Surveys, evaluation, qualitative evidence and the voices of people using services and caregivers remain essential for revealing what service records cannot.
The strongest future model is therefore an intelligence system rather than a reporting system: one that identifies territorial inequality before it becomes entrenched, connects workforce capacity with demand, tests whether services improve people's lives and makes persistent variation visible to those with authority to respond.
As Uruguay moves toward a more universal right to care, evidence will increasingly determine how effectively national ambition becomes local reality. Data earns its value when it leads to better questions, clearer accountability and better decisions for the people whose lives the care system exists to support.