Behavioural Science & Sociocultural Psychiatry · Paper A
Expressed emotion
Expressed emotion describes the emotional climate a patient's family creates around them, and it is measured from the relative, not the patient. Two families of a patient with schizophreniaA long-term mental illness involving episodes of psychosis such as hallucinations and delusions.Read more → can differ sharply: one relative frequently criticises and stays over-involved in every decision, the other stays calm and measured. Clinicians actually assess the relative's attitude, not the patient's own report of how the family feels.
Expressed emotion is rated from the relative using the Camberwell Family Interview (CFI), a structured assessment tool. High expressed emotion: critical comments, hostility and emotional over-involvement; robustly predicts relapse in schizophreniaA long-term mental illness involving episodes of psychosis such as hallucinations and delusions.Read more →, and also in depression and eating disorders. Warmth and positive remarks are not the components that predict relapse; only the critical, hostile and over-involved components carry that risk. Take two equally warm and loving relatives, e.g. one is additionally highly critical and intrusive while the other stays calm and measured: only the first household carries a raised relapse risk despite both being loving. Measuring warmth alone would therefore miss the actual risk factor entirely.
Most relapse risk factors in (genetics, illness severity, medication non-adherence) are hard to change directly. Expressed emotion (EE) is different: it is one of the few psychosocial predictors of relapse that a structured intervention can actually lower. Hold the phrase MODIFIABLE PREDICTOR through every section below.
EXPRESSED EMOTION (EE) (the emotional climate, measured from a relative, that surrounds a patient after they return home) is rated using the CAMBERWELL FAMILY INTERVIEW (CFI) (a semi-structured interview scoring a relative's speech about the patient). High EE comprises three components: critical comments, hostility, and emotional over-involvement (EOI).
A common mistake is assuming high EE means a family is unloving or blameworthy. High-EE relatives are very often exhausted, frightened, or grieving the illness, not uncaring, and framing EE as 'bad families' undermines the therapeutic alliance clinicians need to reduce it. The target of intervention is the specific critical, hostile and over-involved behaviours, not the relative's underlying love or commitment.
Warmth and positive remarks are recorded but do NOT predict relapse; only the critical, hostile and over-involved components do. Two relatives can both be loving: for example, one warm but also intensely critical and intrusive, the other warm and measured: only the first carries raised relapse risk. This is why EE is scored on specific negative components, not on overall relationship quality.
CRITICAL COMMENTS are statements of dislike, disapproval or resentment about the patient's behaviour, counted directly from the interview transcript. HOSTILITY is generalised rejection of the patient as a person, not just specific behaviours: for example, "I can't stand having him in the house." EMOTIONAL OVER-INVOLVEMENT is excessive self-sacrifice, overprotection, or intrusive over-identification with the patient's problems. A household scoring high on any one of these three is classed as high-EE overall.
High EE robustly predicts relapse in and, to a lesser extent, in depression and eating disorders. The effect is large enough that expressed emotion is one of the most replicated psychosocial predictors in psychiatry. Relapse risk rises further when a high-EE relative and patient spend many hours together each week (high face-to-face contact), which is why reducing contact time is sometimes advised alongside other interventions.
Expressed emotion is a FAMILY-ENVIRONMENT predictor of relapse, not a cause of the underlying psychotic illness itself: reducing EE lowers relapse risk but does not treat the psychosis directly. It should not be confused with EXPRESSED AFFECT in general psychiatric interviewing (how a patient themselves shows emotion), which is an unrelated mental state examination concept. High EE is also distinct from simple family conflict; a family can argue without meeting criteria for high EE, and a quiet-seeming family can still score high on emotional over-involvement.
EE is not something the patient is assessed for directly; it is measured from the RELATIVE. The gold-standard tool is the Camberwell Family Interview, a semi-structured, one-to-two-hour interview conducted with the relative alone, scored on five scales: critical comments, hostility, emotional over-involvement, warmth, and positive remarks. A household is classically rated high-EE if the relative makes six or more critical comments, shows any marked hostility, or scores high on emotional over-involvement: any one of the three is enough. Because the full CFI is time-intensive, the Five-Minute Speech Sample offers a shorter, validated alternative: the relative speaks freely and unprompted about the patient for five minutes, and trained raters score the same underlying components from that brief sample.
Screening for high EE is not part of the formal diagnostic criteria for itself, but clinicians assess it informally when planning relapse-prevention care for a patient returning to live with family. NICE guidance on psychosis and schizophrenia (NICE guideline NG178 / the legacy CG178) recommends offering family intervention to families of patients with schizophrenia, particularly where relapse or readmission is likely, and this recommendation is built directly on the expressed-emotion evidence base. Family intervention typically includes psychoeducation about the illness, communication-skills training, and problem-solving support, delivered over multiple sessions, with the explicit goal of reducing criticism, hostility and over-involvement. This makes EE one of the few relapse-prevention targets addressed through a structured non-drug intervention, and it sits alongside first-line antipsychotic management rather than replacing it.
Questions often describe a relative who is warm and loving but also intensely critical, expecting you to recognise this household as high-EE despite the warmth. Another trap asks what NICE recommends for families of patients with frequent relapse and expects 'family intervention' specifically, not generic family counselling or psychoeducation alone. A further trap tests whether you know EE is scored from the RELATIVE's interview, not the patient's own account of their family relationships.
High expressed emotion (critical, hostile or over-involved relatives, measured by the Camberwell Family Interview) predicts relapse, and NICE-recommended family intervention is one of psychiatry's few evidence-based non-drug fixes for it.