01Purpose and principlesWhat the assessment is for and the core concepts behind it.
A single interview is a sample of a fluctuating life. Ask what the person was like before the present change and map the earliest symptom, progression, crises, admissions, treatment trials, adherence, adverse effects, substance exposure, physical illness and recovery. Functional examples are often more reliable than adjectives: missed rent, abandoned cooking, three nights without sleep, stopped collecting children or returned to work. Date uncertainty should be recorded as approximate rather than converted into false precision.
Seek permission early and offer control. The person may agree to contact a partner about medicines but not trauma, or a GP about diagnoses but not family. Explain that someone can provide information without automatically receiving information. Identify communication needs and consider separate conversations so both parties can speak freely. With children or dependent adults in the household, clarify actual caregiving tasks and current safety rather than assuming either danger or protection from a diagnostic label.
Assess the quality of each source. A family member may have excellent knowledge of baseline but limited observation of substance use; an employer may document attendance but not symptoms; previous records can contain repeated unsupported assertions. Document what was directly seen, what was reported, and the context in which it was obtained. Explore discrepancies neutrally with the patient where safe. Coercion, estrangement, cultural differences and legal disputes can shape accounts without making them automatically true or false.
Confidentiality has limits but disclosure is not all-or-nothing. With consent, share information needed for agreed care. Without consent, consider whether the person lacks capacity and sharing is necessary for their overall benefit, whether safeguarding duties apply, or whether failure to disclose exposes the patient or others to a risk of death or serious harm. Use the minimum necessary information, an appropriate recipient and secure method. Record what was considered, advice sought and why the anticipated benefit outweighed the intrusion.
Use longitudinal evidence to test the formulation. Episodic activation independent of substances differs from constant impulsivity; cognitive fluctuation with infection differs from years of decline; symptom remission after stopping a medicine may alter causation. However, collateral information remains evidence rather than verdict. Preserve the person's voice, communicate unresolved differences and ensure significant new facts change the plan rather than merely lengthen the notes.
Key points
- Ask the person's permission to contact named supporters and explain what questions will be asked, what may be shared and why the information could help.
- Collateral history is especially valuable for baseline function, onset, fluctuation, previous episodes, treatment response, adherence, substance use, risk behaviour and the effect on dependants.
- Create a dated longitudinal timeline rather than averaging episodes into a static label; include recovery periods, interventions, medicine changes and social transitions.
- Record the source's identity, relationship, opportunity to observe, exact material information and any possible conflict or limitation.
- Receiving information does not require disclosing confidential clinical details; thank the informant, clarify safety facts and explain what you can and cannot discuss.
- When a patient refuses contact, explore concerns and capacity, offer narrower choices and respect refusal unless a legal, safeguarding or public-interest justification supports proportionate action.
- Interview the patient alone for part of the assessment when coercion, abuse, exploitation or inhibited disclosure may be present.
- Triangulate records, dispensing data, previous clinicians and direct observation where material discrepancies affect diagnosis or safety.
- Tell the patient about significant information and sharing decisions unless doing so would create serious risk or frustrate a lawful safeguarding process.
02Indications, selection and cautionsWhen it is useful, when urgency changes and important limitations.
A supporter describes a clear loss of established sleep, work, self-care, judgement or social function compared with the person's usual pattern.
Distinct periods of symptoms and recovery, especially around treatment or substance changes, can discriminate diagnoses more effectively than a cross-sectional presentation.
The informant has limited opportunity to observe, a significant conflict, uncertain recall or an interpretation that exceeds the concrete facts supplied.
Another person controls access, answers every question, contradicts with intimidation or prevents private contact, prompting a safe safeguarding assessment.
New information describes a specific, credible and near-term danger that may justify proportionate action even when ordinary sharing permission is absent.
03Method and interpretationA systematic approach to the test and its findings.
Read from the initial assessment onwards. Tests may run in parallel in urgent care; first-line, preferred, confirmatory, definitive and gold-standard labels appear only when the chapter explicitly states them.
- 01
Consent and scope discussionFirst step - Why
- Agree who may be contacted, topics that may be discussed and information that may be returned.
- Interpretation and limitations
- Consent may be specific and can change. Record limits accurately; a broad contact name does not create unrestricted access to the clinical record.
- 02
Longitudinal timeline - Why
- Plot baseline, symptom onset, episodes, function, treatments, substances, physical illness and recovery.
- Interpretation and limitations
- Temporal patterns can strengthen or weaken diagnostic hypotheses, but recall errors and retrospective meaning require corroboration and expressed uncertainty.
- 03
Informant interview - Why
- Obtain observable examples of change, adherence, risk, caregiving and response to intervention.
- Interpretation and limitations
- Separate facts from conclusions and assess opportunity to observe. Document material discrepancies and potential coercion rather than choosing a preferred narrative silently.
- 04
Records and medicines reconciliation - Why
- Confirm previous diagnoses, admissions, prescriptions, monitoring, attendance and documented response.
- Interpretation and limitations
- Records improve chronology but can propagate old errors. Treat copied labels without primary evidence cautiously and compare them with the current phenotype.
- 05
Confidentiality and disclosure analysis - Why
- Determine whether consent, best interests, safeguarding, law or prevention of serious harm permits necessary sharing.
- Interpretation and limitations
- Use the least information required, appropriate recipient and secure route; record the rationale and whether informing the patient would create additional danger.
04Clinical next stepsHow the result changes management or prompts escalation.
01Routine collateralAgree contact and obtain specific evidenceFirst stepLongitudinal or functional information would materially improve assessment or care.+
- 1Explain the value of collateral, identify suitable sources and agree the scope of contact with the person.
- 2Ask for dated observable changes, baseline, treatments, risks and strengths rather than inviting a global character judgement.
- 3Compare the account with the patient's history and records, discuss discrepancies where safe and update the formulation transparently.
02Refused contactExplore, narrow and decide lawfullyThe person does not consent to contacting a potentially important informant.+
- 1Explore concerns, offer limited questions or a different source and assess capacity only if the relevant decision is genuinely in doubt.
- 2Respect a capacitous refusal unless a safeguarding, statutory or serious-harm justification makes proportionate action necessary.
- 3AlternativeRecord the decision, information unavailable, resulting uncertainty and any alternative monitoring or safety arrangements.
03Urgent third-party reportReceive, verify and protectSomeone supplies credible information suggesting immediate serious harm or abuse.+
- 1Listen without promising reciprocal disclosure, obtain specific facts, location, timing, means and people currently exposed to danger.
- 2Take immediate clinical or emergency action, seek senior or safeguarding advice and share only what the responsible recipient needs.
- 3Document the source, reliability limits, consent position, lawful rationale, information disclosed and plan for telling the patient when safe.
05Risks, monitoring and follow-upComplications, safety checks and further assessment.
- Update the timeline after admissions, medicine changes, crises, periods of remission and major social transitions so formulation reflects course rather than the latest encounter.
- Review consent and preferred contacts when relationships change, particularly after separation, safeguarding disclosure or transition between services.
- Check whether information received actually changed diagnosis, risk management or support; unacted collateral can create false reassurance about thoroughness.
- Correct demonstrable record errors and distinguish disputed information from verified fact to reduce propagation across future handovers.
- Reassess the safety of contacting an informant when coercive control, stalking or family conflict becomes apparent.
06Special situationsVariants, exceptions and circumstances that change the usual approach.
Information can flow inward
A clinician may listen to a concerned relative even when confidentiality prevents confirming attendance, diagnosis or the content of care.
Function anchors chronology
Observable changes in work, money, sleep, parenting or self-care often date an episode more reliably than retrospective symptom labels.
Records inherit bias
Repeated statements can appear corroborated when every note derives from one original account, so trace important claims to their source.
Permission can be granular
A patient may authorise questions about medication and safety while withholding other sensitive subjects, and the agreed boundary should be respected.
Discrepancy is clinical data
Different accounts may expose memory difficulty, coercion, shame, altered insight or distinct observation periods rather than simple dishonesty.
07Common pitfallsFrequent interpretation and management errors.
- 01
Refusing to hear urgent information because the patient has not consented to disclosure.
- 02
Treating next of kin as a legal power or unrestricted recipient of confidential information.
- 03
Contacting a relative secretly for convenience when no safety or legal justification exists.
- 04
Documenting the informant's diagnosis of the patient as though it were an observed fact.
- 05
Using collateral to erase the patient's account rather than explaining differences.
- 06
Ignoring the possibility that the proposed informant is abusive, coercive or conflicted.
- 07
Failing to record why non-consensual information sharing was necessary and proportionate.