EDITIONEN·FR·ქართ
ASF Logo

Accréditation Sans Frontières

International Accreditation of Healthcare Facilities

Start accreditation  
ASF Standards · Long-Term Care · Standard 12

Standard 12 — Digital Care and Artificial Intelligence Systems for Care

7 criteria · 2 core · 5 standard-level · Version 1.0 · Aligned to ISQua EEA Principle 7, 6th Edition

Criteria in this standard

12.1 — Digital Systems Are Genuinely Evaluated Before Adoption
12.2 — Digital Care Never Disadvantages a Resident or Family Who Cannot Use It
12.3 — Genuine Technical Support Is Available for Digital Systems
12.4 — AI Systems Are Introduced in Line With Law and Best-Practice Guidance
12.5 — AI Systems Are Genuinely Monitored for Unintended Consequences
12.6 — Staff Are Genuinely Consulted Before an AI System Is Introduced
12.7 — Accountability for AI-Assisted Care Is Explicitly Defined
12.8 — Residents and Families Are Genuinely Informed When Care Involves AI
12.1

Digital Systems Are Genuinely Evaluated Before Adoption

Standard

Before adopting any new digital system — care management software, family communication portal, fall-detection or monitoring technology — the facility genuinely evaluates its cost, benefit, and compatibility with existing systems, not adopted on vendor assurance with no real internal evaluation.

In plain terms: Before the facility adopts a new digital tool for resident care or family communication, someone has actually checked it’s worth it and will genuinely work — not adopted because a vendor made it sound good.

Facility category Crisis Transition Small Standard
Applicability Adapted Full Full Full

Why this matters

Long-term care facilities are increasingly offered fall-detection systems, family communication apps, and care management platforms, each promising real benefit but each requiring genuine evaluation before adoption — not assumption that a vendor’s claims translate directly into actual operational fit for this specific facility’s residents and routines.

What good looks like

  • A genuine cost/benefit evaluation happens before adoption.
  • Compatibility with existing resident record systems is genuinely checked.
  • Unintended consequences for resident care are genuinely considered.

Common failure modes

  • A system is adopted on a vendor demonstration with no independent evaluation.

Worked example

In practice
A facility considering a family communication app.
BeforeA vendor’s free trial for a family update app was activated with no check of whether it would genuinely integrate with the facility’s existing care documentation process.
ActionA brief evaluation during the trial confirmed genuine workflow fit and identified a real training need for less digitally confident staff before full rollout.
AfterThe Monitor reviewed the evaluation notes and the resulting training plan. Verified.

If you are starting from zero — do this first

  1. Build a simple evaluation checklist before adopting any digital system.
The most common mistake: Treating a vendor trial as low-stakes and skipping evaluation entirely.

Self-assessment questions

1. Is a genuine cost/benefit evaluation conducted before adoption? — A real, documented evaluation.
Evidence: Evaluation checklist
2. Is compatibility with existing resident record systems genuinely checked? — Real, verified compatibility.
Evidence: Compatibility check record
3. Are unintended consequences for resident care genuinely considered? — Real, proactive consideration.
Evidence: Pre-launch risk review

Common reasons for a PARTIAL answer

  • Evaluation happens for major systems but is skipped for smaller add-ons.

Implementation plan

When What
Week 1 Build an evaluation checklist.

How the Monitor verifies this

Method What Detail
DOCUMENT Evaluation review Reviews the evaluation record for a recently adopted digital system.

Supervisor tips

  • Ask about the most recently adopted digital tool.

Evidence base

World Health Organization. Global Strategy on Digital Health 2020-2025. Geneva: WHO; 2021.

ASF training courses on GMJ Academy →

Foundation courses A-00 to A-03 are live. Criterion-specific modules are being developed and will link here when published.

12.2

Digital Care Never Disadvantages a Resident or Family Who Cannot Use It

Core

A genuine process ensures that residents and family members who cannot use digital devices, lack internet access, or are otherwise unable to engage digitally are never disadvantaged in care communication, family updates, or resident engagement — not digital convenience purchased at the cost of real connection for the residents and families least able to adapt.

In plain terms: A family member who can’t use the facility’s app still gets the same real updates about their loved one — a genuine alternative, not a quietly worse option.

Facility category Crisis Transition Small Standard
Applicability Adapted Full Full Full

Why this matters

Family members of long-term care residents are often themselves older adults, and residents receiving care here, by definition, are an elderly or disabled population — making this criterion genuinely central rather than peripheral in this setting. This is marked Core because a family communication app that leaves a less digitally confident family member genuinely out of the loop about their loved one’s care isn’t a minor inconvenience — it’s a real, significant loss of connection to someone they love.

What good looks like

  • A genuine, equally functional alternative exists for families and residents.
  • New digital services are genuinely pre-tested with representative families.
  • Real evidence shows the alternative delivers equivalent engagement.

Common failure modes

  • Phone updates become deprioritised once an app is introduced.

Worked example

In practice
A facility that introduced a family update app.
BeforeAfter the app launched, staff began posting daily updates there exclusively, and families without the app or with limited comfort using it stopped receiving the same frequency of updates by phone.
ActionA policy was set requiring the same update frequency by phone for any family that preferred it, with staff time specifically allocated for these calls.
AfterThe Monitor interviewed a family member using the phone option who confirmed receiving updates at the same frequency as app users. Verified.

If you are starting from zero — do this first

  1. Confirm non-digital families receive genuinely equivalent update frequency.
The most common mistake: A digital update channel quietly becoming the real primary channel, with the stated alternative receiving less frequent, lower-quality attention.

Self-assessment questions

1. Is there a genuine, equally functional alternative for families who cannot use digital channels? — A real, equivalent alternative.
Evidence: Documented alternative channel
2. Was a new digital service genuinely pre-tested with representative families? — Real, prior testing.
Evidence: Pre-launch testing record
3. Is there real evidence the alternative delivers equivalent engagement? — A real, demonstrated instance.
Evidence: Family interview or comparison data

Common reasons for a PARTIAL answer

  • The alternative exists but receives genuinely less frequent updates in practice.

Implementation plan

When What
Week 1 Audit update frequency for non-app families against app users.

How the Monitor verifies this

Method What Detail
ASK Family interview Asks a non-digital family member about their actual update frequency.

Supervisor tips

  • Ask a family member who doesn’t use the app directly.

Evidence base

World Health Organization. Ethics and Governance of Artificial Intelligence for Health. Geneva: WHO; 2021.

ASF training courses on GMJ Academy →

Foundation courses A-00 to A-03 are live. Criterion-specific modules are being developed and will link here when published.

12.3

Genuine Technical Support Is Available for Digital Systems

Standard

The facility has genuine access to technical support — in-house, vendor-provided, or contracted — for its digital systems, with testing before full implementation, not systems left running with no real support behind them, which carries particular risk where fall-detection or monitoring technology is involved.

In plain terms: When a monitoring system breaks, there’s a real person to call — especially urgent here, since these systems can be directly tied to resident safety.

Facility category Crisis Transition Small Standard
Applicability Adapted Full Full Full

Why this matters

A fall-detection or monitoring system failing overnight, with no clear support path, is genuinely different in consequence from a booking system failing — a malfunctioning safety system could mean a resident fall goes undetected for longer than it should. Real, accessible support, especially covering overnight hours when the facility is least staffed, is what this criterion is actually protecting.

What good looks like

  • Genuine, accessible support exists, including for safety-related systems.
  • Testing genuinely occurs before implementation.
  • A real, known escalation path covers night shifts.

Common failure modes

  • Support exists only during business hours, with no overnight coverage.

Worked example

In practice
A facility with a monitoring system that failed overnight.
BeforeA fall-detection sensor malfunctioned at 2am with no night-shift knowledge of who to call, leaving it offline until the day administrator arrived.
ActionA 24-hour vendor support line was confirmed and posted at the night nursing station, with a documented fallback (increased manual rounding) for any monitoring outage.
AfterThe Monitor asked a night-shift nurse who could correctly describe both the support line and the manual fallback protocol. Verified.

If you are starting from zero — do this first

  1. Confirm support coverage for safety-related systems genuinely extends to overnight hours.
The most common mistake: Support arrangements that only genuinely function during daytime business hours, leaving overnight gaps for safety-critical systems.

Self-assessment questions

1. Does the facility have genuine, accessible support, including for safety-related systems? — Real, available support with particular urgency for safety systems.
Evidence: Support contract
2. Were genuine testing and quality checks performed before implementation? — A real pre-launch check.
Evidence: Testing record
3. Is there a real, known route to support overnight? — A genuine, known path covering night and weekend shifts.
Evidence: Night-shift staff interview

Common reasons for a PARTIAL answer

  • Support exists but night staff don’t actually know how to reach it.

Implementation plan

When What
Week 1 Confirm and post 24-hour support contacts for safety-related systems.

How the Monitor verifies this

Method What Detail
ASK Night-shift interview Asks a night-shift staff member about the support and fallback process.

Supervisor tips

  • Ask specifically about overnight coverage, not just daytime.

Evidence base

NHS Digital. Technology Assessment Criteria for Digital Health Products. London: NHS Digital; 2022.

ASF training courses on GMJ Academy →

Foundation courses A-00 to A-03 are live. Criterion-specific modules are being developed and will link here when published.

12.4

AI Systems Are Introduced in Line With Law and Best-Practice Guidance

Standard

Any artificial intelligence tool used to support fall-risk prediction, behavioural monitoring, or care planning is introduced in accordance with applicable law, or where none exists, with available best-practice guidance such as WHO’s AI ethics and governance guidance — not adopted with no reference to any external standard of responsible use.

In plain terms: Before using an AI monitoring tool, the facility has actually checked what the law requires and how it handles resident consent — not made it up as it went along.

Facility category Crisis Transition Small Standard
Applicability Adapted Full Full Full

Why this matters

AI-assisted monitoring in long-term care carries a genuine, specific sensitivity: it often involves continuous observation of residents who may lack full capacity to consent, making the governance basis and the facility’s approach to consent genuinely important considerations that a formal review is designed to surface.

What good looks like

  • AI systems are genuinely introduced in line with applicable law.
  • Where no law exists, introduction is genuinely informed by recognized guidance.
  • The facility can identify the specific basis, including consent considerations.

Common failure modes

  • Resident or family consent for monitoring was never explicitly addressed.

Worked example

In practice
A facility using an AI-assisted fall-risk monitoring system.
BeforeThe system had been installed with no formal consideration of resident consent or any check of applicable regulation around continuous monitoring.
ActionA governance review confirmed applicable regulation and formally documented the facility’s consent process, involving family or legal representatives where residents lacked capacity.
AfterThe Monitor reviewed the governance note and the consent documentation. Verified.

If you are starting from zero — do this first

  1. Document the governance basis and consent approach for any AI monitoring tool.
The most common mistake: Installing AI monitoring technology with no formal consideration of resident consent.

Self-assessment questions

1. Is any AI system genuinely introduced in line with applicable law? — Real, verified compliance.
Evidence: Regulatory check record
2. Where no law exists, is introduction genuinely informed by recognized guidance? — A real, referenced source.
Evidence: Documented guidance reference
3. Can the facility identify the specific governance basis, including consent? — A real, specific answer covering resident consent.
Evidence: Staff interview

Common reasons for a PARTIAL answer

  • A guidance source is named but consent handling was never explicitly documented.

Implementation plan

When What
Week 1-2 Document governance basis and consent process for AI monitoring.

How the Monitor verifies this

Method What Detail
DOCUMENT Governance note review Reviews the documented governance and consent basis.

Supervisor tips

  • Ask specifically how resident consent was handled.

Evidence base

World Health Organization. Ethics and Governance of Artificial Intelligence for Health: Guidance on Large Multi-Modal Models. Geneva: WHO; 2024.

ASF training courses on GMJ Academy →

Foundation courses A-00 to A-03 are live. Criterion-specific modules are being developed and will link here when published.

12.5

AI Systems Are Genuinely Monitored for Unintended Consequences

Standard

The facility genuinely monitors any AI-assisted fall-risk or behavioural tool for false alarms, missed events, and other unintended consequences — not deploying a tool and assuming it works as intended with no real, ongoing scrutiny, given the genuine risk of alarm fatigue undermining its actual protective purpose.

In plain terms: The facility actually keeps checking whether its monitoring tools are working as expected — and specifically watches for staff becoming numb to constant false alarms, which can undermine the whole point of the system.

Facility category Crisis Transition Small Standard
Applicability Adapted Full Full Full

Why this matters

Fall-detection and behavioural monitoring systems carry a genuinely specific failure mode: a high rate of false alarms can lead staff to develop real alarm fatigue, responding more slowly or dismissively over time — which means a system intended to improve resident safety can, if unmonitored, actually degrade it. Genuine monitoring has to watch for this specific pattern, not just raw accuracy.

What good looks like

  • AI output is genuinely, periodically audited.
  • Alarm fatigue among staff is genuinely monitored as a real risk.
  • A documented instance exists of monitoring catching a real issue.

Common failure modes

  • A high false-alarm rate is never actually tracked or addressed.

Worked example

In practice
A facility with a fall-detection system generating frequent false alarms.
BeforeThe system’s false-alarm rate had never been tracked, and staff had begun responding more slowly to alerts, a pattern nobody had formally noticed.
ActionA monthly review of alert accuracy was introduced, revealing a genuinely high false-alarm rate, which was reported to the vendor for sensitivity recalibration.
AfterThe Monitor reviewed the monthly review record and the vendor’s subsequent recalibration. Verified.

If you are starting from zero — do this first

  1. Establish a monthly review of alert accuracy specifically tracking false-alarm rate.
The most common mistake: Treating any monitoring technology as inherently effective, never checking whether a high false-alarm rate is quietly undermining staff responsiveness.

Self-assessment questions

1. Is AI-assisted output genuinely audited for false alarms or missed events? — Real, ongoing audit.
Evidence: Audit record
2. Is alarm fatigue among staff genuinely monitored? — A real, specific concern genuinely tracked.
Evidence: Staff response time data
3. Is there a documented instance of monitoring identifying a genuine issue? — A real, concrete example.
Evidence: Issue response record

Common reasons for a PARTIAL answer

  • Accuracy is reviewed but alarm fatigue specifically isn’t tracked.

Implementation plan

When What
Week 1-2 Establish a monthly alert-accuracy and fatigue review.

How the Monitor verifies this

Method What Detail
DOCUMENT Audit review Reviews the monitoring record for genuine, specific alarm-fatigue tracking.

Supervisor tips

  • Ask staff directly whether they ever feel numb to the alerts.

Evidence base

US Food and Drug Administration. Artificial Intelligence and Machine Learning in Software as a Medical Device. Silver Spring: FDA; 2023.

ASF training courses on GMJ Academy →

Foundation courses A-00 to A-03 are live. Criterion-specific modules are being developed and will link here when published.

12.6

Staff Are Genuinely Consulted Before an AI System Is Introduced

Standard

Direct care staff who will actually respond to an AI tool’s alerts are genuinely consulted before its introduction, with real training needs identified and addressed — not a tool rolled out to care staff with no engagement beyond a brief notification.

In plain terms: Before a new monitoring tool goes live, the care staff who will actually respond to its alerts have had a real say — including those on night shifts.

Facility category Crisis Transition Small Standard
Applicability Adapted Full Full Full

Why this matters

Care staff across all three shifts actually respond to monitoring alerts, and their genuine, practical input — especially from night staff, who are often most affected by a new alert system and most easily overlooked in daytime-scheduled consultations — is what identifies real workflow issues before they become patient safety concerns.

What good looks like

  • Staff across all shifts are genuinely consulted.
  • Consultation genuinely identifies training needs.
  • Night staff specifically can describe genuine consultation.

Common failure modes

  • Consultation happens only with day-shift staff.

Worked example

In practice
A facility introducing a new fall-detection system.
BeforeA consultation session was held during daytime hours only, meaning night staff — who would be the primary responders to overnight alerts — never had input.
ActionA separate consultation session was scheduled specifically for night-shift staff, surfacing a practical concern about alert audibility that was addressed before go-live.
AfterThe Monitor interviewed a night-shift staff member who described the consultation and the resulting change. Verified.

If you are starting from zero — do this first

  1. Hold a consultation session specifically scheduled to reach night-shift staff.
The most common mistake: Consultation scheduled only during daytime hours, systematically excluding the staff most affected by overnight monitoring technology.

Self-assessment questions

1. Are staff across all shifts genuinely consulted before introduction? — Real, prior consultation reaching night staff.
Evidence: Consultation record
2. Does consultation genuinely identify training needs? — Real needs matched by real training.
Evidence: Training delivery record
3. Can night staff specifically describe having been genuinely consulted? — Tests whether consultation actually reached night staff.
Evidence: Night-shift staff interview

Common reasons for a PARTIAL answer

  • Day staff were consulted but night staff were not.

Implementation plan

When What
Before any launch Hold consultation sessions reaching every shift.

How the Monitor verifies this

Method What Detail
ASK Night-shift interview Asks night-shift staff whether they felt genuinely consulted.

Supervisor tips

  • Ask a night-shift worker specifically, not just day staff.

Evidence base

Topol EJ. High-Performance Medicine: The Convergence of Human and Artificial Intelligence. Nature Medicine. 2019;25(1):44-56.

ASF training courses on GMJ Academy →

Foundation courses A-00 to A-03 are live. Criterion-specific modules are being developed and will link here when published.

12.7

Accountability for AI-Assisted Care Is Explicitly Defined

Core

The facility has genuinely considered and documented accountability for care decisions made with AI support — who is responsible when a fall-risk prediction or behavioural alert is wrong or missed — not leaving this as an unexamined question until an actual resident is harmed.

In plain terms: Everyone knows, in advance, who is actually responsible if a monitoring alert fails or a prediction is wrong — not a question nobody has thought through.

Facility category Crisis Transition Small Standard
Applicability Adapted Full Full Full

Why this matters

This is marked Core because of a genuine, specific risk in this setting: staff can begin to over-rely on monitoring technology, implicitly treating it as a substitute for direct observation rather than a support to it. Explicit accountability — making clear that the technology assists but does not replace actual care staff vigilance — protects against exactly this drift, which otherwise emerges gradually and invisibly.

What good looks like

  • Clinical accountability is explicitly documented.
  • Staff genuinely understand monitoring technology does not replace direct observation.
  • A defined process exists for reviewing accountability after a missed-alert incident.

Common failure modes

  • Staff gradually come to rely on the monitoring system as a substitute for rounds.

Worked example

In practice
A facility where staff had reduced manual rounding after installing fall-detection sensors.
BeforeStaff had informally reduced the frequency of manual room checks since the sensors were installed, effectively treating the technology as a substitute for direct observation.
ActionA clear policy was reissued: sensors support but never replace scheduled manual rounds, with accountability for resident safety explicitly remaining with care staff regardless of sensor status.
AfterThe Monitor interviewed a care worker who could clearly state that rounding frequency had not changed and that accountability remained with staff. Verified.

If you are starting from zero — do this first

  1. Explicitly document that monitoring technology supports but never replaces direct observation.
The most common mistake: Staff gradually, informally reducing direct observation once monitoring technology is introduced, without this drift ever being explicitly addressed.

Self-assessment questions

1. Is clinical accountability for an AI-supported decision explicitly documented? — A real, clear answer.
Evidence: Documented accountability policy
2. Do care staff genuinely understand monitoring does not replace direct observation? — Real, demonstrated understanding.
Evidence: Staff interview
3. Is there a defined process for reviewing accountability after a missed-alert incident? — A real, usable process.
Evidence: Incident review protocol

Common reasons for a PARTIAL answer

  • A policy exists but rounding frequency has quietly declined in practice anyway.

Implementation plan

When What
Week 1 Reissue and reinforce the direct-observation policy.

How the Monitor verifies this

Method What Detail
ASK Staff interview Asks a care worker to describe rounding frequency and accountability.

Supervisor tips

  • Check rounding frequency records against the stated policy.

Evidence base

European Commission. Ethics Guidelines for Trustworthy Artificial Intelligence. Brussels: European Commission; 2019.

ASF training courses on GMJ Academy →

Foundation courses A-00 to A-03 are live. Criterion-specific modules are being developed and will link here when published.

12.8

Residents and Families Are Genuinely Informed When Care Involves AI

Standard

A resident (or their family, where the resident cannot meaningfully participate) whose care involves an aspect delivered with AI system support is genuinely informed of this — not left to assume every decision was made by a human staff member alone, with disclosure treated as optional rather than a genuine, standard part of informed care.

In plain terms: If AI is actually involved in part of a resident’s care — a fall-detection sensor, for instance — the resident or their family genuinely knows this, not left to assume everything is purely human-monitored.

Facility category Crisis Transition Small Standard
Applicability Adapted Full Full Full

Why this matters

A resident’s or family’s genuine right to understand care includes knowing, in plain terms, when a monitoring or care decision involved an AI system rather than staff judgement alone. This is especially relevant in long-term care, where AI-assisted monitoring (fall detection, for instance) often runs continuously in the background — easy for a resident or family to never genuinely realise is there at all unless actively disclosed.

What good looks like

  • Residents and families are genuinely informed when AI is involved in care.
  • Disclosure is genuinely understandable, in plain language.
  • A resident or family member can genuinely confirm they were told.

Common failure modes

  • AI-assisted monitoring runs in the background with no genuine disclosure ever made to the resident or family.

Worked example

In practice
A facility using an AI-assisted fall-detection sensor.
BeforeThe sensor’s role was mentioned only in a lengthy general admission document, which families signed without genuine awareness that an AI tool was actually monitoring their relative.
ActionA brief, plain-language conversation at admission now explains: “A sensor helps alert staff to a possible fall; a staff member always checks before responding.”
AfterThe Monitor interviewed a family member who could genuinely confirm they understood AI was involved. Verified.

If you are starting from zero — do this first

  1. Add a brief, plain-language disclosure point at admission for any AI-assisted monitoring in use.
The most common mistake: AI-assisted monitoring that runs continuously with no genuine disclosure ever actually reaching the resident or family.

Self-assessment questions

1. Are residents and families genuinely informed when AI is involved in care? — Real, standard disclosure.
Evidence: Disclosure protocol
2. Is this disclosure genuinely understandable, not buried in technical language? — A real, plain-language explanation.
Evidence: Admission materials
3. Can a family member asked directly confirm they were genuinely told? — A real, concrete confirmation.
Evidence: Family interview

Common reasons for a PARTIAL answer

  • Disclosure exists in admission paperwork but families genuinely cannot recall or explain it when asked directly.

Implementation plan

When What
Week 1-2 Draft plain-language disclosure wording for admission conversations and train staff to deliver it.

How the Monitor verifies this

Method What Detail
ASK Family interview Asks a family member whether they were genuinely told AI was involved in their relative’s care.

Supervisor tips

  • Ask a family member directly rather than relying on the admission form’s existence alone.

Evidence base

World Health Organization. Ethics and Governance of Artificial Intelligence for Health: Guidance on Large Multi-Modal Models. Geneva: WHO; 2024.

ASF training courses on GMJ Academy →

Foundation courses A-00 to A-03 are live. Criterion-specific modules are being developed and will link here when published.

← LTC Standards — all standards
Self-assess against these criteria →
← Standard 11 — Sustainable Care
Standard 13 — Supporting the Care Workforce →
© 2026 Accréditation Sans Frontières · PHIG · Sheni Network