To assess the feasibility, patient and clinician acceptability and test-retest reliability of the Mini-International Neuropsychiatric Interview (MINI) used by non-psychiatrists in an acute psychiatric ward.

Of 268 consecutive patients included in a cross-sectional study, 176 (66%) completed MINI, and were compared to patients not interviewed. Patients and clinicians were questioned about the interview, using Visual Analogue Scales (VAS). For 38 patients, test-retest reliability was assessed with Cohen's kappa and observed agreement.

MINI was not feasible for all patients. Among factors associated with not being interviewed were early discharge, psychosis, substance use and involuntary admissions. Although evaluations by patients and clinicians completing the postinterview questionnaire varied, MINI was generally perceived as being useful and feasible. Psychotic symptoms were associated with a less positive experience with MINI for both patients and clinicians. In the test-retest analyses, kappa values indicated excellent agreement for six diagnoses, fair to good for six and poor for seven, whereas observed agreement was 75% or above for all disorders.

Among patients admitted to an acute psychiatric ward willing and able to complete the interview, MINI was well accepted by patients and clinicians, and has moderately good test-retest reliability.

Clinical diagnosis has been shown to be unreliable compared to structured diagnostic schedules. However, clinicians rarely use structured diagnostic schedules due to concerns about the feasibility in clinical practice and about patient acceptance. Mini International Neuropsychiatric Schedule is a short diagnostic instrument validated against SCID and CIDI but its feasibility and patient acceptance has not been studied.

One hundred and eleven patients admitted to a partial program were administered Mini International Neuropsychiatric Schedule and the interview was timed. A short questionnaire was administered to assess patients’ views about the interview. For a subgroup of patients, diagnoses by both open interviews and Mini International Neuropsychiatric Interview (MINI) were available. These were compared to look for agreement in primary diagnoses and co-morbid conditions.

MINI took an average of 16.4 min to administer. Patients’ views of MINI were positive. It was considered comprehensive enough to cover all patient symptoms and at the same time not unduly lengthy. Patients were not bothered by the interview format. There was disagreement between MINI primary diagnosis and open diagnosis in 42% cases. In 33% the disagreement was of substantial clinical significance. MINI diagnosed more co-morbid conditions (average 2.05 compared to 0.5 in open interview).

MINI is a short diagnostic interview schedule that can be easily incorporated into routine clinical interviews. It has good acceptance by patients.