ASF Standards · Hospital · Standard 15
Standard 15 — Psychiatry & Mental Health
Criteria in this standard
15.1 — Suicide Risk Screening and Continuous Observation for High-Risk Patients
15.2 — Environmental Ligature Risk Is Identified and Mitigated
15.3 — Restraint and Seclusion Used as Genuine Last Resort, With Strict Documentation
15.4 — Psychiatric Medication Management Follows a Defined Interaction and Monitoring Protocol
15.5 — Staff Are Trained and Drilled in De-escalation, Not Only Briefed
15.6 — Discharge Safety Planning With Follow-Up Care Genuinely Arranged
15.2 — Environmental Ligature Risk Is Identified and Mitigated
15.3 — Restraint and Seclusion Used as Genuine Last Resort, With Strict Documentation
15.4 — Psychiatric Medication Management Follows a Defined Interaction and Monitoring Protocol
15.5 — Staff Are Trained and Drilled in De-escalation, Not Only Briefed
15.6 — Discharge Safety Planning With Follow-Up Care Genuinely Arranged
15.1
Suicide Risk Screening and Continuous Observation for High-Risk Patients
Non-Negotiable
Every psychiatric admission undergoes suicide risk screening using a validated tool, with a patient identified as high-risk placed on a genuinely continuous, documented observation level — not a screening completed once at admission with no follow-through into the observation practice that should result.
In plain terms: Every patient is genuinely screened with a real tool, and a high-risk patient is actually, continuously watched at the level their risk requires — not a form completed at intake that doesn’t translate into real observation practice.
| Facility category | Crisis | Transitional | Small | Standard |
|---|---|---|---|---|
| Applicability | Where applicable | Full | Full | Full |
Why this matters
A validated screening tool catches risk considerably more reliably than clinical impression alone, particularly since a patient at genuine risk may not spontaneously disclose it, and a structured tool prompts specific questions an informal conversation might miss. A screening result that identifies high risk but doesn’t translate into genuinely continuous observation — nominally ordered but with real gaps in actual practice — provides no real protection; the entire protective value lies in the observation genuinely happening, not merely being documented as ordered.
What good looks like
- Screening uses a validated tool, applied consistently to every admission.
- A high-risk observation level is genuinely continuous with real documented intervals.
- Observation level is reassessed as risk changes during the admission.
Common failure modes
- Screening relies on clinical impression rather than a consistent validated tool.
- An ordered observation level has gaps in actual documented checks.
- Observation level is set at admission and never reassessed as risk changes.
Worked example
In practice
A patient admitted and identified as high suicide risk.
BeforeThe patient was correctly screened as high-risk and ordered for continuous observation, but a review of actual observation logs found gaps of up to forty-five minutes during busy shift periods, considerably longer than the continuous standard the order specified.
ActionThe unit built a dedicated staffing allocation specifically for continuous observation patients, separate from general ward duties, ensuring the observing staff member has no competing responsibility that could create a gap.
AfterThe Monitor reviewed subsequent observation logs for a high-risk patient and found genuinely continuous, gap-free documentation. Criterion verified.
If you are starting from zero — do this first
- Adopt a validated suicide risk screening tool if not already in use.
- Review actual observation logs for gaps against the ordered level.
- Build dedicated staffing allocation for continuous observation, separate from competing duties.
The most common mistake: A continuous observation order that’s correctly placed but, in actual practice, has real gaps because the observing staff member also carries other ward responsibilities that compete for their attention.
Self-assessment questions
1. Is screening performed using a validated tool, applied consistently to every admission, not left to clinical impression alone? — A structured tool catches risk more reliably than impression alone, particularly for a patient who may not spontaneously disclose risk.
Evidence: Screening record
Evidence: Screening record
2. Is a high-risk observation level genuinely continuous and documented at real intervals, not nominal with gaps in practice? — A documented observation level that isn’t actually maintained in practice provides no real protection.
Evidence: Observation log review
Evidence: Observation log review
3. Is observation level reassessed as risk changes during the admission, not fixed at the initial screening? — Risk can genuinely change during an admission; a level set once at intake may no longer reflect current status.
Evidence: Reassessment documentation
Evidence: Reassessment documentation
Common reasons for a PARTIAL answer
- Observation is ordered correctly but actual logs show gaps during busy periods.
- Risk level is set at admission but rarely formally reassessed.
Implementation plan
| When | What |
|---|---|
| Week 1 | Confirm a validated screening tool is in consistent use. |
| Week 2 | Audit actual observation logs for gaps against ordered levels. |
| Week 3 | Build dedicated staffing allocation for continuous observation. |
| Ongoing | Review observation log completeness periodically. |
How the Monitor verifies this
| Method | What | Detail |
|---|---|---|
| DOCUMENT | Observation log review | Reviews actual observation logs for a high-risk patient against the ordered observation level for real gaps. |
Evidence base
World Health Organization. Preventing Suicide: A Global Imperative. Geneva: WHO; 2014.
15.2
Environmental Ligature Risk Is Identified and Mitigated
Non-Negotiable
The psychiatric unit’s physical environment is systematically assessed for ligature and other self-harm risk points, with identified hazards genuinely mitigated through environmental design or removal — not a one-time assessment that’s never revisited as fixtures and furnishings change.
In plain terms: The unit’s physical space is genuinely assessed for self-harm risk points, and what’s found is actually fixed — not an assumption based on general psychiatric unit design standards, and not a one-time check that’s never repeated as things change.
| Facility category | Crisis | Transitional | Small | Standard |
|---|---|---|---|---|
| Applicability | Where applicable | Full | Full | Full |
Why this matters
Assuming a unit is safe because it follows general psychiatric unit design principles, without a specific, documented assessment of this particular environment’s actual fixtures and furnishings, overlooks the reality that risk points can be genuinely specific to a given space and its particular contents. A hazard identified in an assessment but left unaddressed provides no real protection — the mitigation step is what actually matters, and a hazard that’s merely noted, without follow-through, represents a documented but unresolved risk.
What good looks like
- A systematic, documented ligature risk assessment has actually been performed.
- Identified hazards are genuinely mitigated, not merely noted.
- The assessment is repeated when the physical environment changes.
Common failure modes
- Safety is assumed from general design standards without a specific unit assessment.
- Hazards are identified in an assessment but not actually corrected.
- The assessment was performed once and never repeated as furnishings changed.
Worked example
In practice
A unit that replaced bathroom fixtures without reassessing ligature risk.
BeforeA thorough ligature risk assessment had been performed when the unit originally opened, but new bathroom fixtures installed during a later renovation were never reassessed against the same risk criteria, introducing an unaddressed hazard.
ActionThe unit built a required ligature risk reassessment step triggered by any renovation or furnishing change, with a specific sign-off required before new fixtures go into clinical use.
AfterThe Monitor reviewed the current assessment record and confirmed it reflected the actual current fixtures, with a documented reassessment following the recent renovation. Criterion verified.
If you are starting from zero — do this first
- Perform or confirm a specific, documented ligature risk assessment for this actual unit.
- Build a mitigation tracking process ensuring identified hazards are genuinely corrected.
- Build a reassessment trigger for any future renovation or furnishing change.
The most common mistake: A thorough initial ligature risk assessment that’s never repeated following a later renovation or furnishing change, leaving newly introduced fixtures unassessed against the same risk criteria that governed the original design.
Self-assessment questions
1. Has a systematic, documented ligature risk assessment actually been performed for the unit, not assumed from general psychiatric unit design standards? — A specific, documented assessment of this unit’s actual environment, not an assumption based on general design principles.
Evidence: Ligature risk assessment report
Evidence: Ligature risk assessment report
2. Are identified hazards genuinely mitigated, not noted but left unaddressed? — A hazard identified but not corrected provides no real protection.
Evidence: Mitigation tracking record
Evidence: Mitigation tracking record
3. Is the assessment repeated when the physical environment changes, not treated as a one-time exercise? — New furnishings or fixtures can introduce new hazards not present at the original assessment.
Evidence: Reassessment trigger log
Evidence: Reassessment trigger log
Common reasons for a PARTIAL answer
- An original assessment exists but hasn’t been repeated after subsequent changes.
- Some identified hazards remain unaddressed in the mitigation tracking.
Implementation plan
| When | What |
|---|---|
| Week 1 | Perform or confirm a current, specific ligature risk assessment. |
| Week 2 | Build a mitigation tracking process for identified hazards. |
| Week 3 | Build a reassessment trigger for future environmental changes. |
| Ongoing | Review mitigation completion periodically. |
How the Monitor verifies this
| Method | What | Detail |
|---|---|---|
| OBSERVE | Unit inspection | Inspects the unit directly against the documented ligature risk assessment for current accuracy. |
Evidence base
World Health Organization. Preventing Suicide: A Global Imperative. Geneva: WHO; 2014.
15.3
Restraint and Seclusion Used as Genuine Last Resort, With Strict Documentation
Non-Negotiable
Physical restraint or seclusion is used only after genuine, documented attempts at de-escalation have failed, for the minimum time necessary, with each episode individually justified, time-limited, and reviewed — not a default response to agitation or a measure extended past clinical necessity.
In plain terms: Restraint or seclusion is genuinely the last resort, after real de-escalation attempts are tried and documented — each time individually justified and time-limited, not a default reaction or something that quietly extends past what’s actually needed.
| Facility category | Crisis | Transitional | Small | Standard |
|---|---|---|---|---|
| Applicability | Where applicable | Full | Full | Full |
Why this matters
Restraint used as a default or first response to agitation, rather than genuinely as a last resort after other approaches are tried, represents both a significant risk to the patient and a well-documented pattern this criterion exists specifically to prevent. A blanket or open-ended restraint order, applied without case-specific justification and active, timely reassessment, can persist well past the point where it’s actually still clinically necessary — the genuine safeguard lies in each episode being individually justified, time-limited, and genuinely reviewed.
What good looks like
- Documented evidence shows de-escalation was genuinely attempted first.
- Every episode is individually, specifically justified and time-limited.
- A genuine, timely review process checks whether restraint remains necessary.
Common failure modes
- Restraint is used as a first response rather than a documented last resort.
- A blanket order is applied without case-specific, time-limited justification.
- Restraint continues without active, timely reassessment of continued necessity.
Worked example
In practice
A unit reviewing a recent restraint episode.
BeforeA restraint episode record showed the restraint was applied but included no documentation of de-escalation attempts made beforehand, and the order was open-ended without a specific time limit or scheduled reassessment.
ActionThe unit built a required de-escalation attempt log preceding any restraint order, with every restraint order required to specify a maximum duration and a mandatory reassessment point before any extension.
AfterThe Monitor reviewed a subsequent restraint episode and found documented de-escalation attempts, a time-limited order, and a timely reassessment leading to early discontinuation. Criterion verified.
If you are starting from zero — do this first
- Build a required de-escalation attempt documentation step preceding any restraint order.
- Require every restraint order to specify a maximum duration.
- Build a mandatory, timely reassessment point before any extension.
The most common mistake: A restraint order that’s open-ended rather than time-limited, with no documented evidence of de-escalation attempts beforehand and no mandatory reassessment point, which allows the restraint to continue well past genuine clinical necessity.
Self-assessment questions
1. Is there documented evidence that de-escalation was genuinely attempted before restraint or seclusion, not restraint used as a first response? — Restraint as a last resort, with real evidence of what was tried first, not the default response to agitation.
Evidence: De-escalation attempt log
Evidence: De-escalation attempt log
2. Is every episode individually, specifically justified and time-limited, not a blanket order applied without case-specific review? — A blanket or open-ended restraint order is a specific, serious failure mode this criterion exists to prevent.
Evidence: Episode-specific justification record
Evidence: Episode-specific justification record
3. Is there a genuine, timely review process checking whether restraint remains necessary, not continued without active reassessment? — Ongoing restraint without active, timely reassessment can persist well past genuine clinical need.
Evidence: Reassessment schedule and record
Evidence: Reassessment schedule and record
Common reasons for a PARTIAL answer
- Justification is documented but de-escalation attempts aren’t clearly logged beforehand.
- Orders are time-limited but reassessment at the limit isn’t consistently timely.
Implementation plan
| When | What |
|---|---|
| Week 1 | Build a de-escalation attempt documentation requirement. |
| Week 2 | Require every restraint order to specify maximum duration. |
| Week 3 | Build a mandatory, timely reassessment point. |
| Ongoing | Audit restraint episodes periodically for full compliance. |
How the Monitor verifies this
| Method | What | Detail |
|---|---|---|
| DOCUMENT | Restraint episode review | Reviews a specific restraint episode record for de-escalation evidence, time-limited order, and timely reassessment. |
Evidence base
World Health Organization. QualityRights Guidance and Training Tools. Geneva: WHO; 2019.
15.4
Psychiatric Medication Management Follows a Defined Interaction and Monitoring Protocol
Core
Psychiatric medications, including those with significant interaction risk or requiring specific monitoring, are managed through a defined protocol covering interaction screening and required laboratory or clinical monitoring — not prescribed and administered without this structured safeguard.
In plain terms: Psychiatric medications go through a real, systematic interaction check and genuinely scheduled monitoring — not just relying on the prescriber’s own memory for every possible interaction or required lab test.
| Facility category | Crisis | Transitional | Small | Standard |
|---|---|---|---|---|
| Applicability | Where applicable | Full | Full | Full |
Why this matters
Relying on an individual prescriber’s memory for every known interaction, rather than a systematic screening step, is considerably less reliable than a structured check — particularly given the number of significant interactions relevant to psychiatric medications and the volume of patients a prescriber manages. Required monitoring that’s left to chance, rather than genuinely scheduled and tracked, risks missing a clinically significant abnormal result that a defined protocol exists specifically to catch.
What good looks like
- Interaction screening is actually performed for every new medication order.
- Required monitoring is genuinely scheduled and tracked, not left to chance.
- A documented process exists for responding to an abnormal monitoring result.
Common failure modes
- Interaction screening relies on prescriber memory rather than a systematic check.
- Required monitoring is assumed to happen without active tracking.
- An abnormal result doesn’t trigger a clear, documented clinical response.
Worked example
In practice
A patient started on a medication requiring regular specific lab monitoring.
BeforeThe medication was correctly prescribed, but required monitoring was left to the prescriber’s own memory to order at the appropriate intervals, resulting in a monitoring gap that went unnoticed for several weeks.
ActionThe unit built an automated monitoring schedule tied to the specific medication, flagging the care team when a required test is due, with a documented escalation step for any result outside normal range.
AfterThe Monitor reviewed a subsequent patient’s record and found monitoring genuinely tracked against the automated schedule, with no gaps. Criterion verified.
If you are starting from zero — do this first
- Build a systematic interaction screening step for new psychiatric medication orders.
- Build an automated or actively tracked monitoring schedule tied to specific medications.
- Build a documented response process for an abnormal monitoring result.
The most common mistake: Required ongoing monitoring for a specific medication left to the individual prescriber’s own memory to order at the correct intervals, rather than a systematic, tracked schedule that flags when a test is due.
Self-assessment questions
1. Is interaction screening actually performed for every new psychiatric medication order, not assumed from the prescriber’s general knowledge alone? — A systematic check, not reliance on an individual prescriber’s memory for every known interaction.
Evidence: Interaction screening record
Evidence: Interaction screening record
2. Is required monitoring, such as specific lab tests for certain medications, genuinely scheduled and tracked, not left to chance? — Specific, required monitoring actually completed on schedule, not assumed to happen.
Evidence: Monitoring schedule tracking
Evidence: Monitoring schedule tracking
3. Is there a documented process for what happens when monitoring reveals an abnormal result? — Detection without a genuine clinical response provides limited protective value.
Evidence: Abnormal-result response protocol
Evidence: Abnormal-result response protocol
Common reasons for a PARTIAL answer
- Screening happens but isn’t consistently systematic across all prescribers.
- Monitoring is scheduled but tracking has occasional gaps.
Implementation plan
| When | What |
|---|---|
| Week 1 | Build a systematic interaction screening step. |
| Week 2 | Build a tracked monitoring schedule tied to specific medications. |
| Week 3 | Build an abnormal-result response protocol. |
| Ongoing | Audit monitoring schedule adherence periodically. |
How the Monitor verifies this
| Method | What | Detail |
|---|---|---|
| DOCUMENT | Monitoring schedule review | Checks a specific patient’s required monitoring schedule against actual completed tests for gaps. |
Evidence base
World Health Organization. mhGAP Intervention Guide for Mental, Neurological and Substance Use Disorders. Geneva: WHO; 2016.
15.5
Staff Are Trained and Drilled in De-escalation, Not Only Briefed
Core
Staff working on the psychiatric unit are trained and practically drilled in de-escalation technique, not only given a written policy to read, with training refreshed on a defined schedule, not completed once at hiring.
In plain terms: Staff genuinely practice de-escalation technique, not just read about it once — refreshed on a real schedule, not a single onboarding session years in the past.
| Facility category | Crisis | Transitional | Small | Standard |
|---|---|---|---|---|
| Applicability | Where applicable | Full | Full | Full |
Why this matters
A written de-escalation policy, however well-crafted, functions very differently under real stress than a technique that’s genuinely been practiced through scenario-based drilling — the gap between reading about an approach and having actually rehearsed it is considerably more significant for a skill that needs to be applied calmly under genuine pressure. Training completed once at hiring, with no refresher, risks skill fade, particularly for staff who use de-escalation technique infrequently and may lack the ongoing reinforcement that regular use would otherwise provide.
What good looks like
- Training genuinely includes practical drilling, not only a written policy or lecture.
- Training is refreshed on a defined schedule, not completed once at hiring.
- Staff can describe their actual de-escalation approach confidently when asked.
Common failure modes
- Training is a written policy distributed with no practical, scenario-based component.
- Training happened once at hiring with no refresher since.
- Staff struggle to articulate a concrete de-escalation technique when asked.
Worked example
In practice
A unit reviewing its de-escalation training approach.
BeforeDe-escalation training consisted of a written policy document reviewed during onboarding, with no practical scenario-based component and no refresher since initial hiring, in some cases years earlier.
ActionThe unit built a scenario-based de-escalation training program with actual role-play practice, required for all staff with direct patient contact, refreshed on an annual schedule.
AfterThe Monitor asked several staff members to describe their de-escalation approach and received confident, concrete answers, and reviewed recent annual refresher completion records. Criterion verified.
If you are starting from zero — do this first
- Build a scenario-based, practical de-escalation training program.
- Set a defined annual or similar refresher schedule.
- Ask several staff to describe their actual approach as a quick confidence check.
The most common mistake: Treating a written de-escalation policy, reviewed once during onboarding, as sufficient training, without the practical, scenario-based drilling that actually prepares staff to apply the technique confidently under real pressure.
Self-assessment questions
1. Has de-escalation training genuinely included practical drilling, not only a written policy or lecture-style session? — Practical, scenario-based training functions differently under real stress than a document only read.
Evidence: Training curriculum content
Evidence: Training curriculum content
2. Is training refreshed on a defined schedule, not completed once at hiring and never revisited? — Skills can fade without periodic reinforcement, particularly for a skill used infrequently by some staff.
Evidence: Refresher completion record
Evidence: Refresher completion record
3. Can staff describe their actual de-escalation approach confidently when asked, not struggle to articulate a concrete technique? — Confident, concrete description suggests the training genuinely embedded, not just technically completed.
Evidence: Staff interview
Evidence: Staff interview
Common reasons for a PARTIAL answer
- Training includes some practical component but isn’t consistently scenario-based.
- A refresher schedule exists but completion isn’t consistently tracked.
Implementation plan
| When | What |
|---|---|
| Week 1 | Review current training content for practical, scenario-based components. |
| Week 2 | Build or source scenario-based training with role-play practice. |
| Week 3 | Set a defined refresher schedule. |
| Ongoing | Spot-check staff confidence periodically. |
How the Monitor verifies this
| Method | What | Detail |
|---|---|---|
| ASK | Staff interview | Asks staff directly to describe their de-escalation approach, checking for confident, concrete answers. |
Evidence base
World Health Organization. QualityRights Guidance and Training Tools. Geneva: WHO; 2019.
15.6
Discharge Safety Planning With Follow-Up Care Genuinely Arranged
Standard
Before discharge, every psychiatric patient has a documented safety plan and a genuinely arranged, confirmed follow-up appointment — not a verbal recommendation to “follow up with your provider” with no actual appointment secured, given the specific, well-documented elevated risk in the period immediately following psychiatric discharge.
In plain terms: Every patient leaves with a real, individualized safety plan and a genuinely confirmed follow-up appointment — not a vague suggestion to “follow up,” during the exact window when risk is known to be highest.
| Facility category | Crisis | Transitional | Small | Standard |
|---|---|---|---|---|
| Applicability | Where applicable | Full | Full | Full |
Why this matters
The period immediately following psychiatric discharge carries a specific, well-documented elevated risk, which is precisely why a verbal recommendation to seek follow-up care, without an actual confirmed appointment, leaves the patient to navigate this high-risk window largely on their own. A generic safety plan handout, given regardless of the patient’s specific situation, provides considerably less genuine protective value than an individualized plan built around that patient’s own actual risk factors, supports, and circumstances.
What good looks like
- Every discharge includes a documented, individualized safety plan.
- Follow-up care is genuinely arranged and confirmed before discharge.
- A documented process exists for following up if the first appointment is missed.
Common failure modes
- A generic safety plan handout is given regardless of individual situation.
- Follow-up care is suggested verbally with no actual appointment secured.
- A missed first follow-up appointment triggers no active re-engagement.
Worked example
In practice
A patient discharged after a psychiatric admission.
BeforeThe patient was given a generic printed safety plan and told to “follow up with your outpatient provider within a week,” with no actual appointment scheduled before discharge and no process to check whether the patient followed through.
ActionThe unit built a required step confirming an actual, scheduled outpatient appointment before discharge is finalized, with an individualized safety plan built from a structured conversation covering the patient’s specific risk factors and supports, and an active re-engagement call if the first appointment is missed.
AfterThe Monitor reviewed a recent discharge record and found a confirmed, scheduled outpatient appointment, an individualized safety plan, and a documented re-engagement call for a patient who had missed the first appointment. Criterion verified.
If you are starting from zero — do this first
- Build a required step confirming a scheduled outpatient appointment before discharge.
- Build an individualized safety plan process, not a generic handout.
- Build an active re-engagement process for a missed first follow-up appointment.
The most common mistake: Discharging a patient with a verbal recommendation to seek follow-up care, rather than an actually confirmed, scheduled appointment, leaving the patient to navigate the known highest-risk post-discharge period largely on their own.
Self-assessment questions
1. Does every discharge include a documented, individualized safety plan, not a generic handout given regardless of the patient’s specific situation? — A genuinely individualized plan, not a standard form applied identically to every patient.
Evidence: Safety plan record
Evidence: Safety plan record
2. Is follow-up care genuinely arranged and confirmed before discharge, not a verbal suggestion with no actual appointment secured? — A confirmed appointment, not a recommendation the patient must arrange themselves during a known high-risk period.
Evidence: Confirmed appointment record
Evidence: Confirmed appointment record
3. Is there a documented process for following up if the patient misses their first post-discharge appointment? — A specific, active process for re-engagement during the known highest-risk window, not simply noting the missed appointment.
Evidence: Re-engagement protocol
Evidence: Re-engagement protocol
Common reasons for a PARTIAL answer
- Appointments are usually confirmed but occasionally remain a verbal suggestion.
- No active re-engagement process exists for a missed first appointment.
Implementation plan
| When | What |
|---|---|
| Week 1 | Review current discharge follow-up confirmation practice. |
| Week 2 | Build a required confirmed-appointment step before discharge. |
| Week 3 | Build an individualized safety plan process and re-engagement protocol. |
| Ongoing | Track missed-appointment re-engagement outcomes. |
How the Monitor verifies this
| Method | What | Detail |
|---|---|---|
| DOCUMENT | Discharge record review | Checks recent discharge records for individualized safety plans and confirmed, scheduled follow-up appointments. |
Evidence base
World Health Organization. Preventing Suicide: A Global Imperative. Geneva: WHO; 2014.