Psychiatric interview: objectives, structure, content, and process. Standardized assessment instruments in psychiatry
The psychiatric interview is psychiatry's main diagnostic tool and the start of treatment. This page covers what it sets out to do, its phases, the history and the mental state examination, risk, cognitive testing, the physical examination and investigations, the formulation, the skills that make it work, and the standardised instruments that support it.
- 14 min read
- 5 questions
- Checked 26 Oct 2026
Contents5 sections
At a glance
- The interview gathers the history and the mental state examination, assesses risk, builds the therapeutic relationship and ends with a formulation and a plan: in psychiatry it is the diagnostic test and the first treatment at once.
- It moves from open to closed questions: the patient's story in their own words first, then clarification, the background history and specific screening questions.
- The history covers the presenting complaint, past psychiatric and medical history, medication, substances, family history, personal history, forensic history, social circumstances and premorbid personality, with a collateral history from an informant where it can be had.
- The mental state examination describes, at the time of the interview, appearance and behaviour, speech, mood and affect, thought form and content, perception, cognition and insight.
- Every assessment asks directly about thoughts of suicide, self-harm and harm to others; asking does not put the idea in a person's head.
- A physical examination and basic investigations look for physical causes and the effects of illness and drugs, and the formulation explains why this person is ill now, by predisposing, precipitating, perpetuating and protective factors.
- Structured interviews (SCID, SCAN, MINI) and rating scales (PHQ-9, HAM-D, PANSS, AUDIT, MoCA and others) make assessment reliable and measure change, but a screening score is not a diagnosis.
Objectives
The psychiatric interview is where most psychiatric diagnoses are made: physical examination and tests play a smaller part than in other specialties, and the diagnosis rests on what the patient says, what the doctor observes, and the later course. It has several aims at once:
- To make a diagnosis, or a differential diagnosis, with the evidence for and against each possibility.
- To assess risk to the patient, to others and from others.
- To understand the person, their history, circumstances, strengths and own explanation, for the formulation.
- To begin the therapeutic alliance: a patient who feels heard is more likely to return, to tell the truth and to accept treatment.
- To agree a plan, explain it, and give information and hope.
Structure
An assessment of an adult usually takes 45 to 90 minutes, and an emergency assessment is shorter and focused on the acute problem and risk, with the rest filled in later.
- 1
Prepare
Read the referral and the notes, especially past violence and risk. Tell a colleague where and whom you are seeing, and know the alarm. Arrange an interpreter if needed. - 2
Open
Introduce yourself and your role, explain the purpose, the length and what you already know, and the limits of confidentiality. Ask whether the patient wants any companion present. - 3
Patient-led history
Open questions about what has been happening. Listen, let the patient talk, and note the mental state as you go. - 4
Doctor-led history
Clarify the symptoms (are these true hallucinations? does the mood vary through the day?) and screen the areas not yet mentioned: mood, anxiety, psychotic symptoms, substances, cognition, risk. - 5
Background
Complete the past, family, personal, forensic and social history and the premorbid personality. - 6
Examine
Complete the mental state examination, a cognitive test where indicated, and the physical examination. - 7
Close
Summarise back to the patient, ask what has been missed, give your understanding of the problem, agree the next steps and the next contact. Leave about a quarter of the time for this.
The assessment goes on afterwards, with a collateral history from relatives, the general practitioner and old records (with consent, unless risk overrides it), nurses' observations and review as the illness evolves: every psychiatric diagnosis is to some degree provisional.
Content
The psychiatric history
| Heading | What to ask |
|---|---|
| Identifying details and referral | name, age, occupation, relationships, who referred and why now; legal status |
| Presenting complaint | the main problems, in the patient's own words |
| History of the presenting complaint | onset, duration, course, severity, associated symptoms, precipitants, effect on work, relationships and self-care, treatment so far, the patient's own explanation |
| Past psychiatric history | previous diagnoses, episodes, admissions (voluntary or compulsory), treatments and their effect, self-harm and suicide attempts |
| Past medical history | physical illnesses, head injury, epilepsy, operations |
| Medication | current and past drugs, doses, adherence, side effects, allergies, over-the-counter and herbal remedies |
| Substance use | alcohol (units a week, pattern, signs of dependence), tobacco, illicit drugs, route and injecting, gambling |
| Family history | a family tree: mental illness, suicide, substance misuse, relationships within the family |
| Personal history | pregnancy and birth, development, childhood (separations, abuse), school, work history, relationships and sexual history, children, menstrual and obstetric history (episodes after childbirth) |
| Forensic history | arrests, charges, convictions, violence, outstanding charges |
| Social circumstances | housing, money and debt, supports, a typical day, dependants, driving |
| Premorbid personality | how the patient, and others, would describe them before the illness: mood, relationships, interests, coping, standards, beliefs |
A collateral history matters most where insight or memory is impaired, in psychosis, mania, delirium, dementia and intoxication, and where the patient's account and the referral differ.
The mental state examination
The mental state examination is the psychiatric equivalent of the physical examination: an ordered description of the patient's experience and behaviour at the time of the interview, recorded with examples in the patient's own words. Much of it is observed during the history.
| Heading | What is described | Examples of abnormality |
|---|---|---|
| Appearance and behaviour | apparent age, dress, grooming, hygiene, weight, scars, eye contact, rapport, attitude, motor activity, abnormal movements, posture | self-neglect, flamboyant dress in mania, agitation, retardation, catatonic signs, tremor and dyskinesia from drugs, responding to unseen voices |
| Speech | rate, volume, tone, quantity, spontaneity, fluency, articulation | pressured speech, slowed speech with long pauses, mutism, dysarthria, dysphasia |
| Mood and affect | mood as the patient describes it (subjective) and as observed (objective); the range, reactivity and congruence of affect | depressed, elevated, irritable, anxious; flat, labile or incongruous affect |
| Thought form | the speed, flow and connection of thought, judged from speech | flight of ideas, loosening of associations, tangentiality, circumstantiality, thought blocking, neologisms |
| Thought content | preoccupations, worries, obsessions, overvalued ideas, delusions (type, conviction, how they are acted on), thoughts of suicide, self-harm and harm to others | persecutory or grandiose delusions, delusions of guilt, intrusive obsessional thoughts |
| Perception | hallucinations (modality, who speaks, what they say, commands), pseudohallucinations, illusions, depersonalisation and derealisation | second- or third-person voices, visual hallucinations in delirium |
| Cognition | level of consciousness, orientation, attention and concentration, memory, an estimate of intelligence; a screening test where indicated | disorientation, inattention, poor recall |
| Insight | whether the patient sees the experiences as illness, explains them as such, and accepts treatment; graded full, partial or absent | absent insight in psychosis and mania |
The psychopathology behind each heading is described in the psychopathology topics, from consciousness and psychomotor activity to thought and perception and memory. Disagreeing with the doctor is not in itself lack of insight: a patient can understand an illness fully and still decline a treatment.
Risk, in brief
Risk is assessed in every patient and recorded as a heading of its own: risk to self (suicide, self-harm, self-neglect), to others (violence, and the safety of children and other dependants) and from others (exploitation, abuse), with risks such as driving where they apply. Thoughts of suicide are asked about in steps, from hopelessness and thoughts that life is not worth living to thoughts of ending it, plans, preparations and past attempts, together with what keeps the person going.
Risk scales predict poorly. UK guidance on self-harm advises not using risk assessment tools or scales, or a rating of low, medium or high risk, to predict suicide or to decide who is treated or discharged; the assessment instead looks at each person's needs and risks and leads to a safety plan. The full assessment and management are in suicidal and aggressive behaviour.
Cognitive testing
A cognitive examination is needed whenever an organic disorder is possible: in older patients, with confusion, memory complaints, a first psychosis or a change in personality. It covers the level of consciousness, orientation, attention (months of the year backwards, serial sevens, digit span), memory (registering and recalling a name and address, recent and remote events), language (naming, repetition, comprehension, writing), visuospatial function (copying a figure, drawing a clock) and executive function (verbal fluency, abstraction, alternating hand sequences). A standard instrument gives it a score (see below). Poor attention points to delirium, a gradual decline in memory with a clear consciousness to the dementias.
Point at a name or a part to light it.
Copying a figure and drawing a clock test visuospatial and executive function, which a conversation can miss. In the MMSE the patient copies two intersecting pentagons, and scores a point only if all 10 angles are there and the two shapes intersect. In the MoCA the patient draws a clock with all the numbers and the time set to ten past eleven, one point each for the contour, the numbers and the hands.
Simplified: example drawings, not any patient's; one impaired clock shows two kinds of error at once; real clocks vary, and a single test never makes a diagnosis; only the MMSE's and the MoCA's own scoring is shown; other clock-drawing scales score in more detail.
8 sources
- Folstein MF, Folstein SE, McHugh PR. "Mini-mental state": a practical method for grading the cognitive state of patients for the clinician. Journal of Psychiatric Research, 1975; 12: 189–198.
- Nasreddine ZS. Montreal Cognitive Assessment (MoCA): administration and scoring instructions, version 7.1 and version 8.1. MoCA Cognition.
- Nasreddine ZS, Phillips NA, Bédirian V et al. The Montreal Cognitive Assessment, MoCA: a brief screening tool for mild cognitive impairment. Journal of the American Geriatrics Society, 2005; 53: 695–699.
- Rouleau I, Salmon DP, Butters N, Kennedy C, McGuire K. Quantitative and qualitative analyses of clock drawings in Alzheimer's and Huntington's disease. Brain and Cognition, 1992; 18: 70–87.
- Soffer M, Butters MA, Herrmann N et al. About time: neurocognitive correlates of stimulus-bound and other time setting errors in the Clock Drawing Test. Journal of the International Neuropsychological Society, 2024; 30: 471–478.
- Shulman KI. Clock-drawing: is it the ideal cognitive screening test? International Journal of Geriatric Psychiatry, 2000; 15: 548–561.
- Kumar and Clark's Clinical Medicine, 10th edition. Elsevier, 2021. Chapter 26, Neurology: dementia, bedside cognitive assessment.
- Semple D, Smyth R. Oxford Handbook of Psychiatry, 4th edition. Oxford University Press, 2019. Chapter 2, Psychiatric assessment (cognitive assessment instruments), and chapter 4, Neuropsychiatry (Alzheimer’s disease 1; Dementia with Lewy bodies).
Physical examination and investigations
The physical examination is part of every full psychiatric assessment, because physical illness can cause or worsen mental symptoms, mental illness and its drugs cause physical harm, and a baseline is needed before treatment. It records weight, body mass index, pulse and blood pressure, and looks especially at the neurological examination (extrapyramidal signs, tremor, gait, focal signs), signs of self-harm, injecting, liver disease, thyroid disease and malnutrition.
| Investigation | When | Why |
|---|---|---|
| Full blood count, urea and electrolytes, liver function, thyroid function, glucose or HbA1c | most new patients | physical causes, such as hypothyroidism in low mood, and baseline before drugs |
| Calcium, vitamin B12 and folate, lipids, prolactin; syphilis and HIV serology | cognitive impairment, alcohol misuse, antipsychotic treatment, risk factors | reversible causes, metabolic monitoring |
| Urine drug screen, alcohol level | psychosis, agitation, unexplained change | substance-induced disorders |
| Pregnancy test | women of child-bearing age before many drugs | teratogenic drugs such as valproate |
| ECG | before and during antipsychotics, with cardiac risk, on lithium | QTc prolongation, arrhythmia |
| Drug levels | lithium, valproate, clozapine | toxicity, adherence |
| CT or MRI of the brain | focal signs, cognitive decline, atypical or late-onset presentations, head injury | structural causes |
| EEG | suspected epilepsy, delirium, non-convulsive status, Creutzfeldt-Jakob disease | not as a screening test |
| Lumbar puncture | suspected encephalitis, including autoimmune encephalitis | infective and antibody causes |
Formulation
The case is summarised in a formulation: a short synopsis, the differential diagnosis with the working diagnosis, the causes in this patient, a management plan and the likely prognosis. The causes are set out in the biopsychosocial model, crossed with the four Ps.
| Biological | Psychological | Social | |
|---|---|---|---|
| Predisposing: why this person | family history of schizophrenia | childhood adversity | migration, urban upbringing |
| Precipitating: why now | heavy cannabis use | exam stress | leaving home |
| Perpetuating: why it continues | stopping the antipsychotic, ongoing cannabis | lack of insight | isolation, unemployment |
| Protective: what helps | good response to treatment | good premorbid functioning | supportive family, a place at college |
The example is of a young man with a first psychosis. The plan follows the same grid: immediate, short-term and long-term measures, each biological, psychological and social, with the perpetuating factors as the main targets.
Process
Setting and safety
- A quiet, private room, chairs of the same height at an angle rather than across a desk, an easy way out for both doctor and patient, the alarm at hand and nothing that could be used as a weapon.
- Violence is rare, and most risk is to the patients themselves, but it is thought about before every new interview; its best predictor is past violence.
- Interpreters: a professional interpreter, ideally sharing the patient's culture, even when the patient speaks some of the language. Relatives interpret only in an emergency, and children are never used as interpreters.
- Notes are made during or soon after the interview, dated and signed, with key symptoms in the patient's own words.
Point at a name or a part to light it.
In psychiatry the patient does not face the doctor across a desk. Two comfortable chairs of the same height, turned towards each other at an angle, are less confrontational and still allow eye contact, and a clipboard lets the doctor write as they go. Ideally each has a door of their own; with one door, the doctor sits so that the patient is not between them and it. The alarm is found before it is needed, and anything that could be used as a weapon is taken out of the room.
Simplified: a plan seen from above: the chairs' equal height cannot show; the ideal room, with two doors; many rooms have one, and then the doctor sits nearer to it; only what the interview needs is drawn; a real room also keeps out anything that could be used as a weapon.
2 sources
- Semple D, Smyth R. Oxford Handbook of Psychiatry, 4th edition. Oxford University Press, 2019. Chapter 2, Psychiatric assessment (The clinical interview, Box 2.1; Setting the scene; Interviewing psychiatric patients).
- Bourke J. Liaison psychiatry. In Kumar and Clark's Clinical Medicine, 10th edition. Elsevier, 2021. Chapter 25, Box 25.2, Essentials of a safe psychiatric interview.
Interviewing skills
| Skill | What it is | Example |
|---|---|---|
| Open questions | do not suggest an answer, used first | "How have you been feeling?" |
| Closed questions | clarify a detail | "How many hours do you sleep?" |
| Avoiding leading questions | the patient may agree to please | "Is your mood better or worse at any time of day?", not "Is it worse in the mornings?" |
| Clarification | asking for an example | "What do you mean by 'they'?" |
| Reflection and empathy | naming the feeling heard | "It sounds as if you've been very frightened." |
| Normalising | making a sensitive question ordinary | "Some people under this much strain have thoughts of ending their life. Has that happened to you?" |
| Signposting | announcing a change of topic | "Now I'd like to ask about some unusual experiences I ask everyone about." |
| Facilitation, silence and summarising | nods, a pause that lets the patient go on, and a summary to check understanding and close a section | "So, since March..." |
A delusion is neither argued with nor agreed with: the doctor acknowledges how real and distressing it is for the patient and explores it. Some patients need their own approach: the agitated (safety, a calm voice, a shorter interview), the mute (observe, offer writing, think of catatonia and physical causes), the intoxicated (reassess when sober unless there is an emergency), the manic and over-talkative (polite, firm interruption and closed questions) and the cognitively impaired (simple questions, hearing aids and glasses, an informant).
Standardised assessment instruments
Instruments make assessment more reliable, help screening, measure severity and show change with treatment. They come in three kinds: structured diagnostic interviews, clinician-rated scales and self-report questionnaires. A fully structured interview asks fixed questions in a fixed order and can be used by trained lay interviewers; a semi-structured one leaves the clinician to probe and judge.
Diagnostic interviews
| Instrument | What it is |
|---|---|
| SCID-5 (Structured Clinical Interview for DSM-5) | semi-structured, clinician-administered, for DSM diagnoses; clinician, research and personality disorder versions |
| SCAN (Schedules for Clinical Assessment in Neuropsychiatry) | the WHO's semi-structured interview, built round the Present State Examination, for ICD-10 and DSM diagnoses |
| MINI (Mini-International Neuropsychiatric Interview) | short structured interview, about 15 to 30 minutes, for the common ICD and DSM disorders |
| CIDI (Composite International Diagnostic Interview) | fully structured, for lay interviewers in population surveys |
Rating scales
| Area | Instrument | Type | What it measures |
|---|---|---|---|
| Depression | PHQ-9 | self-report, 9 items, 0 to 27 | severity over 2 weeks: 5, 10, 15 and 20 mark mild, moderate, moderately severe and severe; 10 or more suggests clinically significant depression; item 9 asks about thoughts of death or self-harm |
| HAM-D (Hamilton) | clinician-rated, usually 17 items | severity and change, the standard in drug trials; not diagnostic | |
| MADRS (Montgomery–Åsberg) | clinician-rated, 10 items, 0 to 60 | severity, sensitive to change | |
| HADS | self-report, 14 items, anxiety and depression subscales of 0 to 21 | screening in medical patients; avoids physical symptoms | |
| EPDS (Edinburgh Postnatal) | self-report, 10 items | depression in pregnancy and after childbirth | |
| Anxiety and OCD | GAD-7 | self-report, 7 items, 0 to 21 | severity of generalised anxiety; 10 or more suggests a disorder |
| Y-BOCS (Yale–Brown) | clinician-rated, 10 items, 0 to 40 | severity of obsessions and compulsions in known OCD | |
| Mania | YMRS (Young) | clinician-rated, 11 items | severity of mania |
| Psychosis | PANSS | clinician-rated, 30 items (7 positive, 7 negative, 16 general), 30 to 210 | severity and change in schizophrenia |
| BPRS | clinician-rated, 18 items | broad psychotic and non-psychotic symptoms | |
| Substances | AUDIT | 10 items, 0 to 40 | alcohol use: 8 or more suggests hazardous or harmful drinking, 20 or more possible dependence; AUDIT-C is its 3-item short form |
| CAGE | 4 yes or no questions | 2 or more suggests an alcohol problem | |
| CIWA-Ar | clinician-rated, 10 items, up to 67 | severity of alcohol withdrawal, to guide symptom-triggered benzodiazepines; local protocols set the threshold for treatment | |
| Cognition and delirium | MMSE | 30 points | global cognition; 24 or more usually normal; weak for mild and executive impairment, and depends on education |
| MoCA | 30 points, about 10 minutes | more sensitive to mild cognitive impairment; below 26 is abnormal, with a point added for 12 years of education or fewer | |
| ACE-III (Addenbrooke's) | 100 points, five domains | attention, memory, fluency, language, visuospatial; for dementia, 88 is the usual sensitive cut-off and 82 the specific one | |
| 4AT | 4 items, under 2 minutes | delirium: 4 or more suggests possible delirium, 1 to 3 possible cognitive impairment | |
| CAM (Confusion Assessment Method) | clinician algorithm | delirium: acute onset and fluctuating course with inattention, and either disorganised thinking or an altered level of consciousness | |
| GCS (Glasgow Coma Scale) | 3 to 15 | level of consciousness | |
| Other | Bush–Francis Catatonia Rating Scale | clinician-rated, 23 items | catatonia; 2 or more of the first 14 items is a positive screen |
| AIMS, Barnes Akathisia Scale, Simpson–Angus | clinician-rated | tardive dyskinesia, akathisia, parkinsonism from antipsychotics | |
| HoNOS, WHODAS 2.0, CGI | clinician-rated or self-report | outcome, disability and global severity; others screen for eating disorders (SCOFF) and bipolar disorder (MDQ) |
Check yourselfAnswer, then say how sure you were: a lucky guess still counts as a weak point.
1A 30-year-old woman is referred by her general practitioner with low mood. After you have introduced yourself and explained the purpose and length of the interview, how do you best begin?
2Which of these findings belongs in the mental state examination rather than in the history?
3A 45-year-old man in primary care completes the PHQ-9 and scores 14 out of 27. What does this score mean?
4The family of a 78-year-old retired engineer describe two years of gradually worsening forgetfulness and difficulty managing his finances. He scores 28 out of 30 on the MMSE. What is the best next step?
5A 22-year-old student is admitted with a first psychotic episode. His uncle has schizophrenia, he began smoking cannabis heavily a month before the onset, his family is supportive, and since discharge he has carried on smoking daily and stopped his antipsychotic. In a formulation, which of these is a perpetuating factor?