Risk formulation and management linkage

8.1 Scope of the risk recommendation

NICE NG225 advises against using risk scales or global low/medium/high stratification to predict suicide or repeated self-harm, or to decide treatment access or discharge. It recommends a needs- and safety-focused assessment that includes risk formulation. This is UK self-harm guidance, not a universal ban on structured assessment, a rule about every violence instrument, or a WA statutory threshold. ^S18

Use Australian prevention and comprehensive-care principles and the applicable WA standard/local procedure alongside that guidance. The WA authority’s standards directory is undergoing updates. This revision verified the listing and update programme, not every current WA risk-documentation requirement. ^S07S09S10

8.2 A prevention-oriented structure

The following is an editorial template informed by a prevention-oriented formulation model. It is not a probability calculator. ^S43

| Element | Documentation focus | |---|---| | Current presentation | Actual thoughts, behaviour, mental state, distress and current safety needs, with source and time. | | State | What has changed relative to this person’s own usual baseline? | | Status, if used | Relative to which specifically described comparison population? Avoid an undefined label. | | Drivers and scenarios | Which change or circumstance could plausibly worsen safety, and what contributes to it? | | Available resources | Who or what is accessible now, acceptable to the person and usable in the scenario? | | Actions and contingencies | What addresses the driver or opportunity, who is responsible, and what happens if the plan cannot operate? | | Uncertainty | What remains unknown, disputed or rapidly changing, and how will it be reassessed? |

Risk domains may include self-harm or suicide, aggression or violence, vulnerability to others, exploitation, self-neglect and physical-health deterioration. Consider each relevant domain rather than assuming a single global risk label applies to all.

8.3 Safeguards for interpretation

Self-harm may serve several functions and still require explicit assessment of suicidal intent, planning, access and medical consequences. Do not infer intent from a presumed relief function. NSSI findings cannot simply be relabelled suicide-death evidence. ^S38S39

Sleep disturbance, hopelessness, impulsivity and loneliness deserve enquiry where relevant, but the cited syntheses do not establish accurate imminent prediction for an individual. Their constructs, populations and timeframes differ. Avoid numerical precision or thresholds not directly supported by the appropriate source. ^S29S30S31S32

A stated support is not necessarily available. Check practical access and consent, ability to respond, relationship safety and what could disrupt it. Do not treat the presence of children, faith, a future appointment or stated willingness to engage as a guarantee.

8.4 Management linkage without deterministic treatment matching

Use a small action table rather than inferring a treatment from a mechanism label alone:

| Supported need or hypothesis | Proposed action and observable aim | Reason to reassess | |---|---|---| | Attendance barrier appears practical | Resolve the identified access problem and observe attendance/function. | The barrier remains, a different barrier emerges or distress persists despite access. | | Repetitive abstract processing appears unhelpful | Agree a concrete problem-solving or attention/activity experiment. | The difficulty is obsessional, trauma-related or an unresolved real-world problem instead. | | A relational misunderstanding is evident | Clarify what occurred, acknowledge service contribution and agree communication. | The account suggests actual abuse, discrimination or a different clinical syndrome. | | Safety worsens in a foreseeable context | Address the modifiable driver, access/opportunity and available support, with a contingency. | Resources fail, mental state changes or new information alters the scenario. |

This is an editorial planning tool. Evidence for a whole treatment programme does not prove its benefit is mediated by the particular mechanism selected in this case. No response and treatment failure are not synonyms, and management should be revised collaboratively and proportionately. ^S04S19S49