A home-care visit can be completed on time, recorded correctly and counted within a program’s activity data while still telling decision-makers remarkably little about whether the person receiving support is living better. An older person may receive more hours of assistance but become less independent. A family caregiver may remain connected to services while becoming increasingly exhausted. A municipality may expand coverage while people with the greatest needs still struggle to gain access. Measurement becomes meaningful only when it can distinguish activity from change.
This is becoming a strategic issue across the Chile Aging, Long-Term Care & Community Support Knowledge Hub because Chile is moving from a collection of care-related programs toward a more explicit national system. Law No. 21.805, which created the Sistema Nacional de Apoyos y Cuidados (SNAC), does more than recognize the right to care. It establishes monitoring, evaluation, information and outcome requirements that can progressively make the effectiveness of support more visible.
The opportunity is significant. Chile already has population evidence through instruments such as the Encuesta de Caracterización Socioeconómica Nacional (Casen), administrative information across health and social programs, local assessment data and an emerging Sistema de Gestión de Información de Apoyos y Cuidados (SGIC). The challenge is turning those different information sources into an evidence architecture capable of answering the questions that matter.
Are people maintaining greater autonomy? Are caregivers experiencing less overload? Is access becoming more equitable between territories? Are services responding to changing dependency? Are public resources producing meaningful improvements rather than additional transactions? And when the evidence shows that outcomes are weaker in one place or population, does anything change?
Chile’s care legislation places outcomes inside the system architecture
Law No. 21.805 provides an unusually important starting point because measurement is embedded within the design of SNAC rather than left entirely to individual programs.
The law establishes principles of effectiveness, efficiency and quality. Effectiveness is connected to satisfying the needs of people receiving care and improving quality of life and wellbeing. Efficiency concerns achieving intended results through appropriate use of available resources. Quality is linked explicitly to the generation of indicators capable of supporting continuous improvement.
Those distinctions matter operationally. A service can be busy without being effective. It can meet a numerical target without producing the intended outcome. It can expand coverage while remaining inaccessible to particular populations.
The legislation also requires the Secretaría de Apoyos y Cuidados to request, register, monitor and administer information concerning implementation, including indicators used to monitor, report and verify progress. Programs entering the system are expected to have identifiable objectives, target populations, services and expected-result indicators.
Importantly, those indicators are intended to be verifiable and capable of being disaggregated by factors including age, gender, level of dependency and territory. This creates the foundations for a much more sophisticated approach to outcomes frameworks and indicators than a single national total can provide.
The strategic question is now how these statutory expectations translate into measures that are sufficiently consistent for national accountability while remaining meaningful to people, municipalities and services.
Activity, output and outcome answer different questions
Long-term care systems frequently collect what is easiest to count. Visits, assessments, referrals, hours, places, training sessions and equipment deliveries are important operational measures because they show whether planned activity occurred. They should not be mistaken for outcomes.
Three levels of evidence need to remain distinct.
- Inputs and capacity describe resources such as funding, workforce, available service hours, equipment and provider capacity.
- Outputs show what the system delivered: assessments completed, people supported, home visits, respite hours, adaptations or professional interventions.
- Outcomes examine what changed for people, caregivers, communities or the wider system as a result.
The distinction becomes clearer through an example. A municipality may report that 500 people received home support during the year. That is useful coverage information. It does not establish whether recipients maintained functional ability, experienced greater control over daily life, avoided preventable deterioration or felt that support reflected their preferences.
Similarly, providing respite to an unpaid caregiver is an output. Whether that respite reduces overload, improves wellbeing or enables the caregiver to remain in employment is an outcome question.
A mature evidence system needs all three layers because outcomes without operational context can also mislead. If wellbeing improves, leaders need to understand what intervention may have contributed. If outcomes deteriorate, they need to know whether the issue reflects service quality, inadequate intensity, delayed access or changes in the population being supported.
Casen provides a population baseline that service data cannot replace
Chile’s Casen survey is particularly valuable because it provides a population-level perspective rather than describing only people who successfully reach services. The Ministry of Social Development and Family uses Casen to understand household and population circumstances, identify needs and inequalities and support social-policy evaluation.
The 2024 dependency and care analysis illustrates why this matters. It estimated that around 770,000 people aged 15 and over were experiencing dependency, with 44.6% of that population experiencing severe functional dependency. The analysis also shows the strong relationship between dependency, population aging and unpaid care.
Service databases answer a different question. They can show who received a particular program. Casen can help illuminate the population that exists beyond program boundaries.
This prevents a common measurement error: treating the characteristics of current service users as though they represent the total population requiring support.
Suppose a region records relatively few people receiving formal care. That could mean dependency is genuinely lower. It could also mean informal family care is absorbing demand, services are difficult to reach, local supply is limited or eligible people are not identified.
Population evidence provides the denominator against which service reach can be interpreted.
This makes population needs assessment particularly important as SNAC develops. The question is not simply how many people the system serves, but what proportion of relevant need it reaches and how that differs across population groups.
Coverage should be measured against need, not celebrated in isolation
Chile Cuida has expanded significantly, including through the Red Local de Apoyos y Cuidados (RLAC). Expansion is an important implementation measure, but national coverage numbers become much more useful when connected to local need.
Consider two municipalities that each support 400 people through care-related services. The figure appears identical. Yet one municipality may have a much larger population of older people, more severe dependency, greater rural dispersion and fewer alternative services.
The same output therefore represents a different level of population coverage.
A stronger territorial evidence framework might examine the relationship between estimated need, identified eligible population, assessed demand, people entering services, waiting or prioritization, service intensity and outcomes after support begins.
It should also make unmet need visible. People who never receive a service can disappear from operational dashboards precisely because there is no service transaction to count.
This is one reason Law No. 21.805 requires annual monitoring of both the offer and demand for support and care programs. Demand cannot be inferred solely from utilization.
For Chile, measuring the gap between population need and effective access will be one of the most important tests of whether expansion is becoming equitable rather than simply larger.
Autonomy needs to become a measurable outcome
SNAC is explicitly intended to promote autonomy, self-reliance and independent living and to prevent dependency. These concepts therefore need a meaningful place in evaluation.
Autonomy is not equivalent to completing every activity without assistance. A person can require substantial physical support while retaining control over where they live, how their day is organized, who supports them and which risks they choose to take.
This means outcome measurement should avoid reducing independence to a narrow functional score.
Functional measures remain important. Changes in mobility, personal care, cognition or ability to undertake everyday activities can indicate changing support needs. But a person-centered outcomes framework should also consider control, participation, relationships, safety, accessibility and whether support enables the person’s own life goals.
Organizations developing similar evidence systems can use the Quality Dashboard Builder to structure different layers of performance information and test whether operational metrics are being connected to outcomes. It is not a Chilean national reporting instrument, but it illustrates an important discipline: a dashboard should show the relationship between delivery, quality and results rather than simply accumulate available numbers.
Caregiver outcomes belong inside the long-term care evidence model
A care system can improve outcomes for a person with dependency while transferring an unsustainable amount of work to their family. If measurement looks only at the person receiving support, that deterioration remains hidden.
This is especially important in Chile because unpaid care remains a major component of the care economy and is disproportionately provided by women. Law No. 21.805 recognizes unpaid caregivers as rights holders within the care system and explicitly addresses their access to support, training, mental health services, rest and opportunities for decent work.
Caregiver outcomes therefore need to extend beyond whether a caregiver remains available.
Useful evidence may include changes in perceived overload, time spent caring, ability to rest, participation in employment or education, mental wellbeing, confidence in providing care and access to support networks. No single measure captures the whole experience, and not every program needs to collect every indicator.
The important principle is that family capacity should not be treated as an unlimited resource.
Imagine a 52-year-old woman caring for her father following increasing functional decline. A local care plan introduces several hours of formal support each week. The service records successful delivery of all scheduled visits. From an activity perspective, performance is strong.
Six months later, however, the daughter has reduced her employment, stopped attending social activities and is sleeping poorly because her father now needs help during the night. The formal service has been delivered exactly as planned, but the household outcome is deteriorating.
If reassessment includes caregiver wellbeing as well as the father’s dependency, the evidence can trigger a different conversation about intensity, respite, other available supports and the sustainability of the care arrangement. This is why family carers and care burden should be treated as part of system performance rather than an external family matter.
Local measurement needs enough consistency to support national learning
Chile’s care architecture combines national policy with regional coordination and significant local implementation. Municipalities can develop Local Support and Care Plans, while programs such as RLAC rely heavily on territorial assessment, coordination and delivery.
This creates a familiar measurement tension.
Too little standardization makes comparison difficult. If each municipality defines successful support differently, national leaders cannot identify systematic variation. Too much standardization can strip measurement of local meaning and encourage services to optimize national indicators rather than respond to their populations.
The stronger model combines a relatively small common outcome set with additional measures appropriate to local circumstances and specific services.
National measures might establish whether people are accessing support, whether functional and wellbeing outcomes change, whether caregivers experience reduced burden, whether service quality is acceptable and whether inequalities persist. Municipalities may then add measures relating to particular rural conditions, Indigenous communities, local transport, service availability or other territorial priorities.
The statutory requirement for indicators to be disaggregated by territory provides an important safeguard against averages concealing variation.
Organizations examining how accountability travels between national, regional and local levels can use the Governance Maturity Assessment to structure questions about responsibility, oversight and evidence flow. It does not assess Chilean public bodies, but its underlying governance principle is relevant: information has limited value if nobody is clearly accountable for interpreting and acting upon it.
Territorial variation should trigger inquiry rather than simplistic ranking
Disaggregated data will inevitably reveal differences between Chilean territories. Those differences need careful interpretation.
Suppose a rural municipality reports slower completion of care-plan reviews than an urban municipality. A performance table could present the rural service as weaker. Further investigation might reveal long travel distances, a small specialist workforce and households dispersed across difficult terrain.
Those contextual factors do not make delayed reviews irrelevant. They change the improvement question.
If the delay reflects poor local organization, management action may be appropriate. If it reflects structural workforce scarcity, national or regional capacity planning may be required. If digital follow-up could safely complement some face-to-face activity, technology may help. If the national review expectation itself assumes urban operating conditions, policy design may need adjustment.
Measurement should therefore support explanation rather than simply classification.
This is particularly important for data-led equity planning. Equality of performance cannot always mean identical service processes. What matters is whether people have equitable opportunity to achieve good outcomes despite different territorial conditions.
For Chile’s long geography, regional and municipal variation should be treated as intelligence about system design, not merely a league table.
Longitudinal measurement is more useful than isolated snapshots
Long-term care is inherently dynamic. Dependency can improve, stabilize or increase. Caregiver circumstances change. Housing becomes more or less suitable. A hospitalization can alter support needs within days.
A single assessment describes a point in time. Outcomes emerge through change.
This makes longitudinal evidence particularly valuable. Repeated measures can show whether a person remains stable, deteriorates more slowly, regains function or experiences changing quality of life. At population level, longitudinal patterns can help leaders understand which interventions appear associated with better trajectories.
Care is also preventive, which creates a measurement problem. Some successful outcomes involve something not happening: a fall avoided, dependency delayed, caregiver breakdown prevented or residential admission deferred.
These outcomes require caution because causation is difficult to prove. An individual remaining at home for another year does not establish that one service caused the result.
Nevertheless, systems can examine patterns across sufficiently large groups, compare trajectories, combine quantitative evidence with professional and lived-experience information and avoid making stronger causal claims than the data supports.
The purpose is not to create false scientific certainty. It is to make improvement more evidence-informed.
Qualitative evidence explains what indicators cannot
Numbers can identify that an outcome changed without explaining why.
A person may discontinue a service because it is no longer required, because it achieved its purpose, because the family moved, because the person disliked the support or because access became too difficult. Recording each as “service exit” loses the distinction.
Chile’s legislation strengthens the case for qualitative evidence because participation by people receiving care and caregivers is embedded within SNAC. National and regional civil-society participation is also relevant to evaluation of the National Policy on Supports and Care.
Meaningful participation should therefore influence what is measured as well as provide comments after measurement has been designed.
Interviews, structured feedback, complaints, focus groups, care-plan reviews and individual outcome narratives can reveal whether people experience greater dignity, control, continuity and confidence. They can also identify unintended consequences that predefined indicators miss.
This is the value of qualitative evidence and case studies when they are used rigorously. A powerful story should not replace representative data, but data without lived experience can produce an equally incomplete picture.
The Community Impact Report Builder can help organizations structure quantitative evidence alongside lived experience and community impact. Used appropriately, this type of framework helps prevent outcome reporting from becoming either purely statistical or purely anecdotal.
Data quality determines whether apparent performance is real
As SNAC expands its information architecture, data quality will become an operational governance issue.
An indicator is only as reliable as the information underneath it. Different interpretations of definitions, missing reassessments, duplicated people, inconsistent coding and delayed entry can all create apparent performance differences that are actually measurement differences.
Consider two services recording caregiver overload. One completes the measure with almost every participating caregiver at baseline and review. Another records it only when staff believe the caregiver is struggling. The second service may appear to have a much higher prevalence of severe burden even if the populations are similar.
Without understanding completion rates and methodology, the comparison is unsafe.
Chile therefore needs common definitions, metadata, validation processes and visibility of missing information alongside headline indicators. Data-quality metrics should themselves form part of assurance.
This connects outcome measurement with data collection and data quality. More data is not automatically better evidence. Reliable, appropriately defined information collected for a clear purpose is more valuable than a large dataset whose meaning changes between services.
The SGIC can become an accountability infrastructure rather than a database
Law No. 21.805 creates the Sistema de Gestión de Información de Apoyos y Cuidados (SGIC), administered within the Ministry of Social Development and Family. Its purpose extends beyond storage: the legislation connects the system with timely information, system functioning and continuous improvement.
The SGIC is intended to draw together relevant data and information, support the administration of benefits and services and incorporate the National Register of Support and Care Service Providers. The wider legislation also creates obligations around information exchange and interoperability.
The strategic value of this infrastructure will depend on what decisions it enables.
A national information system should make it easier to understand coverage, demand, provider participation, territorial variation and implementation. At local level, appropriate information should help teams coordinate support rather than forcing people and caregivers to repeat the same circumstances to multiple parts of the system.
Yet integration needs boundaries. Long-term care information can include health, disability, family, socioeconomic and household information. Access should be proportionate to role and purpose, with personal-data protections and information-security requirements treated as fundamental rather than administrative obstacles.
Good interoperability is selective. It allows relevant information to move safely without assuming that every actor needs access to every record.
As Chile develops the SGIC, the stronger opportunity lies in connecting interoperability and data-exchange workflows with explicit accountability for data quality, access, correction and use.
Outcome evidence should change care at the individual level
National evaluation matters, but measurement creates its most immediate value when it changes an individual care decision.
Consider a 74-year-old man participating in a local support program following a stroke. At initial assessment he needs assistance with several daily activities and his wife provides substantial unpaid care. The plan includes home support, rehabilitation-related coordination and an assistive device.
Three months later, scheduled services have been delivered. If review focuses only on completion, the plan appears successful.
A meaningful outcome review looks further. His mobility has improved and he can now complete some tasks independently, but he has stopped attending a neighborhood activity because transport remains difficult. His wife reports less physical strain but remains unable to leave him alone for long enough to resume part-time work.
The evidence suggests a mixed result: functional improvement, incomplete community participation and continuing caregiver constraint.
The appropriate response may be to reduce assistance in an area where independence has increased while redirecting support toward a different barrier. Outcome measurement has therefore prevented the care plan from becoming static.
This is an important distinction. Measurement should not exist solely to report upward through institutions. It should also support reassessment, shared decision-making and adaptation around the person.
Public accountability needs interpretation as well as transparency
Law No. 21.805 creates several layers of formal evaluation. The Secretaría de Apoyos y Cuidados is required to monitor the National Policy and its plan annually. The policy itself is to be evaluated every three years, including coverage, quality, accessibility and cultural relevance. SNAC is subject to evaluation every two years, with a public report and recommendations for improvement.
This creates a potentially powerful accountability rhythm.
Publication alone, however, is not enough.
Public reporting needs to explain what indicators mean, how they were calculated, where data limitations exist and why performance may differ. Otherwise transparency can produce misleading conclusions rather than informed accountability.
Recommendations also need a route into decisions. If an evaluation repeatedly identifies weak rural access, caregiver overload or uneven quality, leaders should be able to see what corrective action followed, who owns it and whether subsequent evidence demonstrates improvement.
The accountability cycle should therefore connect:
- measurement of implementation and outcomes;
- interpretation of variation and underlying causes;
- public reporting and participation;
- clear improvement responsibilities;
- changes to programs, resources or technical guidance where justified; and
- subsequent measurement of whether the response worked.
This turns reporting into evidence of implementation rather than an annual exercise in producing statistics.
Outcome measurement should influence resource decisions without becoming crude payment logic
As Chile’s care system matures, outcome information will inevitably become relevant to decisions about investment and resource allocation.
That can improve policy. Evidence may show that some forms of early support preserve independence, that particular territories have large unmet needs or that caregiver interventions reduce unsustainable pressure on households.
But using outcomes for funding requires caution.
Services working with people who have severe dependency, complex disability or difficult socioeconomic circumstances may achieve smaller observable functional gains than services supporting populations with lower needs. A crude funding model that rewards only improvement could therefore penalize the organizations working with people who require the most support.
Maintenance can itself be a meaningful outcome. Slowing deterioration may be valuable. Supporting a person to remain at home according to their wishes may represent success even when functional dependency increases.
Risk adjustment, population context and the nature of the service therefore matter whenever outcomes influence funding decisions.
Chile’s emerging system has an opportunity to use evidence to inform resource allocation without turning every care outcome into a financial incentive. The purpose of measurement should remain better decisions about people and populations, not the creation of metrics that encourage services to select easier cases.
Continuous improvement closes the distance between evidence and action
The quality principle in Law No. 21.805 explicitly connects indicators with continuous improvement. That connection is critical.
Imagine that national data identifies repeated deterioration in caregiver outcomes within a particular type of support pathway. The first response should not be to assume frontline failure. Analysis may reveal that reassessments occur too infrequently, respite availability is limited or support intensity does not change quickly enough as dependency increases.
The improvement response can then target the actual mechanism.
Organizations facing similar evidence-to-action challenges can use the Quality Improvement Action Plan Builder to structure identified problems, actions, ownership, timescales and follow-up evidence. It does not prescribe Chilean policy or regulatory action; its relevance lies in reinforcing the discipline that every material finding needs an accountable improvement pathway.
The same principle applies nationally. Evaluation recommendations need owners. Technical guidance may need revision. Funding may need redistribution. Workforce development may need to change. A program may require redesign rather than simply a new target.
Measurement becomes a learning system when information changes practice and the effect of that change is measured again.
International learning: measure what the care system is trying to achieve
Long-term care systems internationally have struggled with the same fundamental problem: activity is easier to measure than quality of life.
Some systems have developed extensive standardized outcome sets, provider reporting requirements and public dashboards. These approaches can strengthen transparency, but they can also generate substantial reporting burden and encourage organizations to focus on indicators that are easiest to demonstrate.
Chile does not need to reproduce another country’s measurement architecture.
Its emerging advantage is that the legal purpose of SNAC already identifies outcomes that matter: autonomy, independent living, prevention of dependency, quality, accessibility, wellbeing, participation and support for caregivers. The task is to build an evidence framework around those purposes rather than allowing available administrative data to determine what success means.
The transferable international lesson lies less in selecting a particular indicator set and more in preserving alignment between policy purpose, frontline measurement and accountability.
A useful measure should help someone make a better decision. At individual level, that may mean adjusting a care plan. At municipal level, it may mean identifying an access gap. At regional level, it may expose workforce constraints. Nationally, it may influence program design or investment.
If an indicator is collected repeatedly but cannot plausibly affect any decision, its value deserves examination.
The next stage is an outcomes architecture, not simply more data
Chile is likely to generate substantially more care data as SNAC expands, the SGIC develops and additional programs become connected to the national system. The strategic risk is assuming that increased data volume automatically produces increased intelligence.
The stronger opportunity is to establish a coherent outcomes architecture.
That means agreeing which national questions need consistent measurement while allowing service-specific and territorial evidence where appropriate. It means connecting population surveys such as Casen with administrative and service information without confusing their different purposes. It means treating caregiver outcomes as part of system performance, preserving lived-experience evidence and making inequalities visible through disaggregation.
It also requires measurement discipline. Indicators need definitions. Missing data needs visibility. Changes in methodology need documentation. Local teams need information that helps them manage care rather than reporting requirements that exist only for central administration.
Over time, this architecture could allow Chile to understand not simply how many people receive care, but which combinations of support appear to sustain autonomy, where unmet need remains concentrated, how caregiver outcomes change and whether territorial inequalities are narrowing.
That is the point at which data becomes strategic care-system intelligence.
Conclusion
Chile’s transition toward a National System of Supports and Care creates an opportunity to make outcomes part of the operating logic of long-term care rather than an evaluation exercise added after services have been delivered. Law No. 21.805 already provides important foundations through its requirements for indicators, annual monitoring, periodic public evaluation, territorial disaggregation, participation and the development of the SGIC.
The harder work is translating those requirements into evidence that remains meaningful from the household to the national level. Coverage needs to be understood against population need. Functional measures need to sit alongside autonomy and quality of life. Caregiver wellbeing needs visibility. Territorial differences need explanation rather than simplistic ranking. Administrative data needs to be combined carefully with lived experience, while information quality and privacy remain fundamental.
Most importantly, evidence needs consequences. A deteriorating outcome should prompt reassessment around a person. Persistent local variation should trigger investigation. National evaluation should influence policy, technical guidance and resource decisions where the evidence justifies change.
Chile does not need to measure everything. It needs to measure enough of the right things, consistently enough, to understand whether the right to care is becoming a better lived reality. The strongest accountability system will therefore be one that connects evidence with decisions, decisions with improvement and improvement with demonstrable changes in autonomy, wellbeing, equity and sustainable care.