Measuring Quality, Outcomes and Accountability in French Long-Term Care: From Compliance to Evidence of Impact

A French EHPAD can receive a strong external evaluation and still face questions that the evaluation alone cannot answer. Are residents maintaining mobility for longer? Are avoidable hospital transfers reducing? Do people feel listened to when their care changes? Are staffing pressures beginning to affect continuity? Has a recurring medication problem actually been solved, or has an action merely been recorded as complete?

These questions illustrate the next stage of France’s quality agenda. The country now has a more standardised national framework for evaluating établissements et services sociaux et médico-sociaux, or ESSMS, alongside growing public transparency through Qualiscope. Within the wider France Ageing, Long-Term Care & Community Support system, the strategic challenge is increasingly to connect formal evaluation with evidence of what happens to people between evaluation visits.

This does not diminish the importance of compliance. Rights, safeguarding, risk management, workforce competence and governance controls remain fundamental. But the strongest long-term care assurance system needs to answer a wider question: not simply whether a service has the expected processes, but whether those processes produce safer, more personalised and more sustainable support.

France is now in a stronger position to make that shift. The national HAS framework creates common expectations. Qualiscope makes evaluation results visible. Open data increases analytical potential. Départements, agences régionales de santé, CNSA and providers already hold large volumes of operational information. The opportunity is to connect these components so quality becomes a continuous management discipline rather than an event associated mainly with inspection or evaluation.

France Has Built a More Unified National Quality Framework

The quality of social and medico-social services in France has long been subject to statutory evaluation, but the current system represents a significant consolidation.

The Haute Autorité de santé developed a common evaluation framework for ESSMS that began to be implemented from 2023. It applies across a very diverse social and medico-social sector encompassing older people, disabled people and other populations, while allowing evaluation methods to be applied in ways appropriate to different services.

The first national cycle runs across 2023–2027. By April 2026, approximately 37% of France’s 47,700 ESSMS had been evaluated under the renewed framework.

Evaluation is carried out by independent accredited organisations using the national HAS methodology. Results are normally transmitted every five years according to the programme established by the authority or authorities responsible for the service’s authorisation.

The framework contains 139 standard criteria and 18 critères impératifs, the latter reflecting requirements where weak performance demands particular attention and corrective action.

Its subject matter is considerably broader than procedural compliance. It addresses areas including:

• the rights and participation of the person receiving support;
• co-construction and personalisation of the support plan;
• support for autonomy and health;
• continuity and fluidity of people’s pathways;
• workforce policy;
• bientraitance and ethics;
• continuous quality improvement and risk management.

This breadth matters because it gives France a common language for quality and governance in services for older people without reducing quality to a single clinical or administrative measure.

The HAS Evaluation Looks Beyond Documents

A major strength of the current evaluation approach is that it is not intended to be a desk-based review of policies.

The methodology draws evidence from the experiences of people receiving support, professional practice and organisational governance. The accompagné traceur method gives people using services an opportunity to describe their own experience directly to an evaluator. Other methods examine professional practice and the organisation’s systems.

This triangulation is important.

A provider may possess a policy describing how residents participate in their care plan. Records may show that reviews occurred. But the person’s own account may reveal whether they understood the review, influenced decisions or felt that the agreed changes happened.

Equally, a manager may believe an incident-learning system is functioning because investigations are completed. Staff interviews and practice observation may reveal whether learning has actually reached frontline teams.

The distinction is fundamental to mature quality monitoring systems: evidence becomes stronger when records, practice and lived experience point towards the same conclusion.

Organisations considering comparable assurance questions can use the Governance Maturity Assessment to examine whether responsibility, evidence, challenge and escalation are connected rather than operating as separate management activities.

Qualiscope Has Changed the Transparency of Social and Medico-Social Quality

The expansion of Qualiscope to ESSMS represents another important development.

Since September 2025, results from the renewed ESSMS evaluations have been progressively made available to the public through the HAS platform. From January 2026, the underlying ESSMS evaluation data also became available as open data.

For each evaluated establishment or service, users can see information about the structure, its quality level and detailed thematic results. Services must also make their principal evaluation results accessible locally.

The four-level quality scale provides a more understandable public summary than a long technical report alone.

Transparency can strengthen accountability in several ways. Families considering an EHPAD can see independent evaluation information. Provider organisations can compare themes across their own services. Départements and ARS can identify wider patterns. Researchers and public bodies can analyse national data at a scale that was previously difficult.

Yet transparency also creates a new governance responsibility: ensuring numbers are interpreted properly.

HAS has explicitly warned against using the data to create simplistic rankings between services. ESSMS differ substantially in the populations they support, territorial conditions and operational context. An evaluation also captures quality at a particular point in time; it does not automatically show improvements made afterwards.

The objective should therefore be intelligent comparison, not league tables.

A Quality Rating Is a Starting Point, Not the Outcome

The visibility created by Qualiscope can tempt organisations to make the external score the principal objective.

That would be too narrow.

An evaluation can establish whether the organisation has met defined quality expectations at the point of assessment. It can expose weaknesses and prompt improvement. But it cannot substitute for continuous knowledge of what is happening in the service.

Consider an EHPAD whose evaluation identifies strong resident participation and appropriate governance. Six months later, nursing vacancies increase, agency usage rises and several experienced aides-soignants leave. Medication incidents then increase and families begin reporting inconsistency.

The historic evaluation remains valid as evidence of the quality observed at the time. It is no longer sufficient evidence of current service stability.

This is why external evaluation needs to sit inside a wider cycle:

evaluation should generate learning; learning should generate action; action should change practice; and subsequent operational evidence should show whether the change worked.

That final stage is the difference between compliance and impact.

Operational scenario: A strong evaluation is followed by workforce deterioration

An EHPAD receives a positive external evaluation. Its processes for personalised support, resident rights and governance are well established.

Eight months later, the service experiences several departures from its night team. Vacancies are temporarily covered, but continuity deteriorates. Managers initially see the issue as a recruitment problem rather than a quality problem.

Over the following weeks, other signals emerge. Falls increase during evening transfers. Families report that newer staff do not always know residents’ routines. Documentation is completed, but the timing of some care becomes less consistent.

A mature quality system brings these signals together. Workforce turnover, agency usage, falls, complaints and resident experience are reviewed as connected evidence rather than separate departmental metrics.

The response includes recruitment, but it also examines deployment, induction, supervision and the preservation of resident-specific knowledge during staff changes.

The external evaluation has not become irrelevant. Instead, the service uses it as a baseline. The important question becomes whether the conditions supporting that earlier quality remain present.

This is where workforce assurance becomes part of quality assurance rather than a parallel human-resources activity.

France Needs to Distinguish Process Measures From Outcome Measures

Long-term care measurement becomes stronger when organisations distinguish clearly between what they do and what happens as a result.

A process measure might show that every resident has a falls risk assessment. An outcome measure asks whether falls, injury or fear of falling are changing.

A process measure may show that personalised plans are reviewed on schedule. An outcome question asks whether the person’s choices are reflected in daily life.

A service may record that every complaint received a written response. The more important longer-term evidence is whether the same concern keeps recurring.

Both types of measure are necessary.

Process evidence is particularly important where a control exists to prevent serious harm. Services need to know that medication checks, safeguarding escalation, staff competency assessments and emergency procedures actually occur.

But outcomes give those controls meaning.

The strongest measurement systems therefore connect three levels of evidence:

• inputs and capacity — staffing, competencies, funding, equipment and service availability;
• process and practice — whether agreed support and controls are delivered;
• outcomes and experience — what changes for the person, family and service system.

This creates a more useful form of quality data and performance measurement because leaders can understand not only that an indicator moved, but what may have caused the change.

Autonomy Is an Outcome, but It Is Difficult to Measure Well

One of the central purposes of French long-term care is to support autonomy. Yet autonomy cannot be reduced to a single numerical score.

For some people, a positive outcome may mean recovering enough strength after hospitalisation to resume preparing meals. For someone with progressive dementia, success may instead mean remaining safely in familiar surroundings, maintaining meaningful relationships and avoiding unnecessary distress.

For an EHPAD resident with advanced frailty, maintaining function rather than improving it may represent a significant achievement.

Outcome systems that reward improvement alone can therefore create misleading incentives. They may unintentionally make services supporting people with more complex or deteriorating conditions appear less effective.

Measurement has to account for starting point, prognosis, personal goals and the type of service.

Useful autonomy evidence may include mobility, participation, ability to undertake valued activities, stability of support, avoidable deterioration and the person’s own perception of control.

It should also recognise the importance of outcomes-focused support in which goals are meaningful to the individual rather than chosen simply because they are easy to count.

Resident and Family Experience Must Carry More Weight

France’s evaluation framework already gives the experience of people receiving support a formal place through methods such as the accompagné traceur.

The next opportunity is to strengthen this principle between evaluation visits.

EHPADs have conseils de la vie sociale, while other participation mechanisms operate across the wider ESSMS sector. Complaints, compliments, surveys, informal feedback and family meetings can all provide evidence about quality.

But collecting feedback is not the same as using it.

A service can achieve a high survey-response rate while learning very little if questions are generic or results are aggregated too heavily. Conversely, a small number of detailed comments can reveal an important pattern.

Experience data becomes operationally valuable when it is connected to decisions.

If residents repeatedly say evenings feel rushed, management should compare that feedback with staffing, call-bell response, medication rounds and incident patterns. If families report frequent changes in home-care workers, the issue should be reviewed alongside rota continuity and vacancy levels.

This is where service-user feedback and co-production become governance evidence rather than consultation activity.

Operational scenario: Residents describe a problem the dashboard misses

An EHPAD’s conventional indicators appear stable. Falls have not risen significantly, medication incidents remain low and staffing hours are within the planned level.

However, several residents tell the conseil de la vie sociale that evenings have become less pleasant. Staff appear hurried and people feel they have less choice about when they prepare for bed.

There is no single major incident to investigate.

Management nevertheless treats the feedback as an early quality signal. A closer review shows that staffing numbers are technically sufficient, but activity is concentrated around the same period because medication, personal care and documentation routines overlap.

The service redesigns workflow and changes some responsibilities. It then seeks feedback again rather than assuming the action has succeeded.

Resident experience improves without increasing overall staffing hours.

The important lesson is that a dashboard built only from incidents and statutory measures would have missed the problem. Lived experience identified deterioration before it became a conventional safety event.

Workforce Indicators Are Often Leading Quality Indicators

Long-term care quality cannot be understood without workforce data.

The relationship is rarely as simple as a fixed staff-to-resident ratio. Skill mix, continuity, sickness, turnover, vacancy duration, use of temporary workers, supervision and access to training all affect what people experience.

Different service models also require different workforce configurations.

A rural service autonomie à domicile may have enough contracted staff in total but still fail to cover distant communes efficiently. An EHPAD may appear adequately staffed overall but have insufficient nursing capacity on particular shifts. A team may complete mandatory training yet lack confidence in dementia-related distress or palliative care.

The most useful workforce measures therefore connect capacity with consequence.

For example, rising turnover becomes particularly significant when it coincides with increasing missed or delayed home visits. A training gap becomes more important if incidents reveal the same weakness in practice.

This form of analysis also prevents quality teams from treating workforce pressure as outside their remit.

The person receiving support experiences one service. They do not experience separate human resources, quality and operational departments.

Incidents Should Produce System Learning, Not Just Closure

Incident reporting is necessary, but the number of closed incidents says little about whether the organisation is becoming safer.

A mature learning system looks for recurrence, common causes and weak controls.

Three medication errors occurring in different units may initially appear unrelated. If each occurred during temporary staffing changes, they may point towards a common induction or medicines-handover issue.

Repeated falls may indicate individual frailty, but a cluster at the same time of day could expose a deployment or environmental problem.

Quality teams therefore need thematic analysis alongside case-by-case review.

This connects directly to learning from incidents and continuous improvement. The objective is not simply to demonstrate that action followed an event. It is to establish whether the action reduced the underlying risk.

Organisations developing comparable systems can use the Quality Dashboard Builder to structure operational, workforce, experience and safety indicators into a more coherent assurance view.

Home Care Requires Different Quality Evidence From Residential Care

The renewed HAS framework covers services delivered in the home as well as residential establishments, but the operational evidence required to understand quality differs.

In an EHPAD, many interactions occur within one organisation and one building. Managers can observe practice, review incidents and compare units relatively directly.

Home support is dispersed across thousands of private homes.

A service autonomie à domicile therefore needs visibility over travel, punctuality, missed visits, continuity of worker, rapid changes in need, communication with nurses and other professionals, lone-working risks and the reliability of digital scheduling.

Raw visit-completion data is insufficient. A worker may technically arrive within the permitted window while the person experiences frequent changes of staff or badly sequenced visits.

Outcome evidence also needs to reflect the purpose of home support. Important questions include whether the person remains able to live at home safely, whether support adapts when needs change, whether carers remain sustainable and whether deterioration is identified early.

These factors make supervision and quality assurance in homecare particularly dependent on information quality and frontline judgement.

Operational scenario: Every visit is completed, but continuity is deteriorating

A SAD reports a very high proportion of commissioned visits completed. On the surface, performance appears strong.

A deeper analysis identifies a different pattern. People with the most complex support needs are seeing an increasing number of different workers each month. Families begin contacting the service more frequently because they are having to explain routines repeatedly.

No major safety failure has occurred, and conventional completion data remains positive.

The service therefore adds continuity measures to its quality review. It examines the number of different workers supporting each high-need person, short-notice rota changes and complaints linked to unfamiliar staff.

Managers then redesign small teams around defined geographic groups and complex cases. Recruitment remains necessary, but deployment also changes.

The result illustrates why a technically successful activity measure can coexist with deterioration in the person’s experience. Quality becomes visible only when the service measures what continuity means to the person rather than simply whether a visit occurred.

Territorial Authorities Need More Than Provider-Level Scores

Départements and ARS need to understand the performance of individual services, but they also need a territorial view.

A collection of satisfactory provider evaluations does not necessarily mean the wider system is functioning well.

People may still experience long waits, poor transitions, shortages in rural areas or repeated hospital admissions because gaps exist between organisations.

Territorial accountability therefore requires measures that cross organisational boundaries.

Examples include delays in accessing home support, emergency EHPAD placements, continuity following hospital discharge, geographical availability, unmet support needs and the sustainability of informal carers.

These indicators help distinguish service quality from system accessibility.

A provider may be delivering excellent support to everyone it accepts while the territory still lacks enough capacity to meet demand.

That distinction becomes increasingly important as the SPDA develops a more coherent departmental autonomy pathway. Better navigation should eventually be assessed not only through the existence of new organisational arrangements but through whether people experience faster, clearer and more continuous support.

Public Transparency Creates Accountability but Also Requires Context

The publication of evaluation results changes the relationship between services and the public.

Families can ask more informed questions. Provider groups can see variation between establishments. Journalists, researchers and representative organisations can examine broader patterns.

This should strengthen accountability.

However, public reporting needs sufficient explanation to prevent crude interpretation.

A service working with a particularly complex population should not automatically be judged against another organisation as though the populations, resources and territorial environments were identical.

Nor should a five-yearly evaluation score become a permanent label.

Quality can improve after a weak evaluation and deteriorate after a strong one.

Transparency therefore works best when the public score is treated as one part of an evidence narrative. Providers should be able to explain what the evaluation found, what changed afterwards and what current evidence shows.

This approach makes accountability more dynamic. It shifts the conversation from defending a rating towards demonstrating improvement.

Digital Systems Can Strengthen Assurance but Can Also Produce False Confidence

French long-term care increasingly generates digital information through care records, scheduling systems, medication tools, workforce platforms and national reporting.

Used well, this can create far stronger quality visibility.

Managers can identify deteriorating trends earlier, compare services and monitor whether actions are producing change. Territorial bodies can combine provider and population data to understand demand and capacity.

But digitisation does not automatically improve quality.

A digital record can still contain poor-quality information. A dashboard can display precisely calculated indicators that do not measure anything meaningful. Automated alerts can increase workload if thresholds are badly designed.

Digital governance therefore needs three tests: is the data accurate, is the measure meaningful and does someone have responsibility for acting on it?

This is why digital records and information governance are inseparable from quality assurance.

For organisations examining how technology supports evidence and operational oversight, the Digital Transformation Readiness Assessment provides a practical framework for considering data, workforce adoption, governance and digital capability together.

Quality Improvement Must Be Visible Between External Evaluations

The five-year evaluation cycle makes continuous internal assurance essential.

French law already requires ESSMS to identify quality-improvement actions in their annual activity reporting. The principle is clear: external evaluation should sit within a continuing improvement process.

In practice, the risk is that providers concentrate activity around the evaluation date and then allow assurance to become less visible afterwards.

A stronger model maintains a continuous line of evidence.

Evaluation findings should be translated into specific priorities. Responsibility should be assigned. Progress should be reviewed. Relevant metrics should be monitored. People receiving support and professionals should be asked whether practice has changed.

Where improvement is not occurring, the organisation should understand why.

Completion of an action is not the same as resolution of the problem.

If a training course is delivered following repeated safeguarding concerns, leaders need to know whether staff practice subsequently changes. If a new falls protocol is introduced, the service should assess incidents and resident confidence afterwards.

This is the difference between an action plan and continuous improvement.

Operational scenario: Training is complete but the outcome does not improve

A provider group identifies repeated documentation weaknesses following several internal incidents. The agreed response is refresher training, which every relevant employee completes.

The improvement tracker therefore shows 100% completion.

Three months later, audit results have improved only marginally. Frontline discussions reveal that the problem is not primarily knowledge. Staff are recording information late because the digital workflow is cumbersome during busy periods.

The provider changes its approach. It redesigns parts of the workflow, provides targeted coaching and reviews staffing pressure at the busiest times.

Subsequent audits improve substantially.

The original action had been completed, but the original problem had not been solved. Outcome-focused governance identified the difference.

This is the type of evidence that should sit behind quality reporting: not only what the organisation did, but whether its intervention changed practice.

A Strong French Quality Dashboard Should Connect Different Forms of Evidence

No single indicator can summarise quality in long-term care.

A useful dashboard should therefore combine a limited set of measures that allow leaders to understand quality from different perspectives.

For an EHPAD, that may include resident experience, falls with harm, medication incidents, hospital transfers, pressure damage, workforce turnover, sickness, vacancies, complaints, continuity of key staff and progress against improvement priorities.

For a SAD, measures may instead emphasise missed or late visits, continuity of worker, changes in need, hospital admission, complaints, staff travel pressure and capacity to respond to urgent deterioration.

The purpose is not to maximise the number of metrics.

Too many indicators create noise. The strongest dashboards focus on information that supports decisions.

Each metric should therefore answer four questions:

• What aspect of quality does this measure?
• What change would require investigation?
• Who is responsible for responding?
• What additional evidence would confirm whether the change matters?

This turns data into governance.

National Data Can Reveal System-Level Quality Problems

The growing availability of standardised evaluation data creates a significant opportunity for France.

HAS’s 2025 results, published in 2026, already showed that the quality observed across evaluated ESSMS was generally satisfactory while highlighting weaker areas around quality systems and risk management.

It also showed that criteria examined through the experience of people receiving support tended to perform somewhat better than areas relating to professionals and governance.

These patterns matter because they show where national improvement effort may need to focus.

As the first evaluation cycle progresses, France will accumulate a far richer dataset across service types and territories. That should allow analysis of recurring weaknesses, geographical variation and differences between parts of the medico-social sector.

The opportunity is not to centralise every operational decision.

It is to use national evidence to identify common problems and then support territorial and organisational learning.

If risk-management weaknesses repeatedly appear across many services, national guidance, training or digital support may be justified. If a particular region shows persistent access or workforce problems, the response may need to be territorial.

Quality measurement therefore becomes part of policy design, not merely provider oversight.

Accountability Should Include Whether People Can Access Quality Support at All

A high-quality service cannot benefit someone who cannot obtain access to it.

This is an important limitation in conventional quality measurement.

France can assess the standards within authorised services while people simultaneously experience waiting, workforce shortages or limited provision in particular territories.

Accountability therefore needs two dimensions:

quality within services and quality of the wider autonomy pathway.

The latter includes whether assessment occurs promptly, whether an appropriate service is available, whether hospital discharge can be supported and whether people can remain close to their community.

This is especially important in rural areas, where formal entitlement may coexist with limited practical availability.

The same principle applies to families. A care plan may look adequate on paper because relatives continue absorbing unmet need. Unless carer strain is measured, system capacity can appear stronger than it really is.

Measurement therefore needs to expose hidden pressure rather than only count formal service activity.

From Compliance to Evidence of Impact

France does not need to abandon compliance-based assurance to become more outcome focused.

The two approaches should reinforce each other.

External evaluation establishes whether fundamental expectations are being met. Quality and risk systems protect against preventable harm. Public transparency creates accountability.

Outcome evidence then answers the next question: whether those structures are improving people’s lives.

This requires a shift in management culture.

Leaders should become less interested in whether a dashboard is entirely green and more interested in whether it tells the truth. A deteriorating indicator is useful if it allows early intervention. Resident criticism is valuable if it reveals a problem that would otherwise remain invisible. A weak external evaluation can become productive if it generates sustained improvement.

The strongest evidence culture therefore does not punish every negative signal. It makes those signals safe enough to surface, rigorous enough to investigate and important enough to act upon.

That principle is internationally transferable even where regulatory structures differ substantially from France.

International Learning: Do Not Confuse Standardisation With Uniformity

France’s experience offers an important lesson for other long-term care systems.

A common national framework can create consistency without requiring every service to produce identical outcomes.

Standardised expectations around rights, participation, safety, workforce and quality improvement are valuable because they establish a common floor. Standardised publication also makes the sector more transparent.

But different populations and service models require different operational indicators.

The outcome expected from a reablement-focused home service cannot be identical to that of an EHPAD supporting people with advanced dementia. A rural SAD cannot be interpreted without considering geography. A service supporting people with very high acuity should not be penalised simply because its raw incident rate exceeds that of a lower-risk population.

The transferable principle lies in combining national standards with context-sensitive measurement.

Other systems can adapt that idea without replicating France’s HAS, département or ARS architecture.

The Next Stage Is Continuous, Connected Accountability

France’s quality infrastructure is becoming significantly more transparent and analytically useful.

The next stage is to connect it.

External HAS evaluation, provider quality systems, workforce information, resident feedback, incidents, territorial capacity data and public reporting should not operate as separate evidence streams.

Each describes part of the same system.

At provider level, this means understanding relationships between workforce stability, practice, safety and experience. At département and ARS level, it means connecting provider quality with access and territorial capacity. At national level, it means using accumulated evaluation data to identify structural patterns rather than simply producing aggregate scores.

The result would be a stronger form of accountability: one capable of showing not only whether organisations meet expectations but whether public investment, professional effort and service reform are producing better support for older people.

Conclusion

France has moved decisively towards a more visible and standardised model of quality assurance in its social and medico-social sector. The renewed HAS evaluation framework, five-year assessment cycle, public Qualiscope reporting and expanding open data provide a stronger national infrastructure for understanding how ESSMS perform.

But external evaluation can only show part of the picture.

The central strategic challenge is now to connect formal assessment with continuous evidence of resident experience, autonomy, safety, workforce stability, service continuity and territorial access. A service should be able to demonstrate not simply that policies exist or actions were completed, but that practice changed and people benefited.

This requires better measurement rather than simply more measurement. Providers need a focused view of the indicators that genuinely reveal risk and outcomes. Départements and ARS need evidence that crosses organisational boundaries. National bodies need to distinguish meaningful variation from differences in population and context. People using services and families must remain part of the evidence itself rather than merely recipients of published information.

The strongest direction for French long-term care is therefore continuous, connected accountability. Compliance remains essential, but it becomes the foundation rather than the destination. Quality is demonstrated most convincingly when rights, governance, workforce and operational controls can be traced through to the experience and outcomes of the people the system exists to support.