The British Journal of Psychiatry
The Reliability and Validity of the Psychiatric Assessment of the Child: I. Interview with the Child


A half-hour diagnostic psychiatric interview was described for use with 7- to 12-year-old children. Four studies of reliability and validity were reported. Inter-rater reliability was examined by having children interviewed by one psychiatrist while a second psychiatrist observed without taking part. At the end of the interview, without any discussion, both psychiatrists independently made separate ratings of the child's psychiatric state. Re-test reliability was examined by having a further series of children interviewed twice, once by each author, with an interval of 1 to 30 days between the interviews. Validity was tested by comparing the interview findings for a randomly selected sample of the general population and a sample of children judged to have psychiatric disorder on the basis of information from teachers and parents. The interviewing psychiatrist knew only the child's name and age and was unaware of the selection group from which the child was drawn.

A short psychiatric interview with a child was found to be a reasonably sensitive diagnostic instrument which could give rise to reliable and valid judgments on whether the child exhibited any psychiatric disorder. However, on the whole, individual ratings on specific aspects of behaviour proved to be less reliable than the overall psychiatric diagnosis. There were also marked differences in the reliability and usefulness of the various specific ratings.

Depression proved to be a more difficult item than anxiety to rate reliably, but both were found to be reasonably efficient in the differentiation of the child with psychiatric disorder. The child's attention or persistence on cognitive tasks was also rated fairly reliably, and proved to be a surprisingly good indicator of psychiatric disorder. The child's emotional responsiveness at interview and the relationship he formed with the psychiatrist could be rated reliably, and these items also effectively differentiated children with psychiatric disorder from "control" children in the general population.

Some of the inferences involved in the diagnostic process were examined in relation to the assessment of anxiety. It was shown, for example, that whereas fidgetiness is commonly taken to be an indicator of anxiety this assumption is frequently misleading. Tension and tremulousness were also less useful indicators of anxiety than tearfulness, an anxious expression and preoccupation with anxiety topics at interview.

The main uses and limitations of the interview have been briefly discussed. It is concluded that the interview with the child (aged 7 to 12 years) forms a useful part of the diagnostic procedure in child psychiatry.