THE CONCEPT of depressive disorder in childhood is one which causes controversy (Rutter, 1971;Graham, 1974;Petti, 1978). In adult psychiatry the clarification of the diagnostic category has been aided by the establishment of operational criteria for depressive illness (Feighner et al., 1972), the development of rating scales to measure mood states (Zung, 1965;Hamilton, 1967), the use of careful research methods to compare different patient populations and the effectiveness of antidepressant treatments. While confusion and dispute have not been entirely eliminated, the diagnostic confidence of clinicians now rests on a firm base (Kendell, 1976).In child psychiatry, the situation is complicated by the fact that the child is a developing organism whose mood states can fluctuate rapidly. There is considerable confusion over the use of the term "depression" and, in the current classification scheme for psychiatric disorders in childhood and adolescence, a,child with depressive features can be classified in no less than fourteen different ways. Moreover, the theoretical models used by child psychiatrists may outnumber even those of their colleagues in adult psychiatry (Sandier and Joffe, 1965; Akiskal and McKinney, 1975).Most previous studies in the field suffer from deficiencies which leave the reader in some doubt about the validity of the diagnosis of depressive disorder in childhood. Frommer (1968) and other writers are less than clear about what they mean by "depression" so that claims that one quarter of children referred to a clinic suffer from depressive disorder must be regarded with caution. Pearce (1978) studied children with depressive affect as a symptom and confirmed that this did occur in approximately one-quarter of clinic referrals. While he provides evidence that a proportion of these children probably do suffer from a depressive syndrome which resembles that found in adults, the proportion of children who actually meet the operational criteria for depressive disorder that he proposes is not clear.Studies which have sought to use a rating scale i:o measure depressive disorder in childhood have been reviewed recently by Petti (1978). He concludes that depression as a concept is difficult to operationalise and reveals that there is currently no satisfactory self-rating scale for use with children. The Beck Depression inventory seems over-sensitive, rating over 50% of normal 11-15 year olds as depressed

The many conceptual and methodological difficulties involved in evaluating depression rating scales for children are discussed. A clinical validation of the Depression Self-Rating Scale for Children (DSRSC) is described. The instrument is easy to use and has a predictive value comparable with that of a psychiatric global rating of depressed appearance and history of depression obtained at interview. There was confirmation that the DSRSC can tap an internal dimension of depression and that children are able to evaluate their feeling states. An examination of misclassified children pointed to diagnostic overlap and some unreliability of diagnosis by clinicians.

Dropout of treatment is one of the key issues in outcome in a child and adolescent mental health service. We report two studies focusing on the treatment process and the dropout rate of children with persistent conduct problems presenting to a community mental health service, using a prospective design. The first study included 32 children and used a randomised controlled treatment design comparing a CBT approach with conjoint family therapy and an eclectic approach. The overall dropout rate was 36%. Dropout occurred significantly less frequently in the CBT group. The dropout group was associated with mothers who were younger and less educated, a poorer rating by the clinicians at the last meeting, parental dissatisfaction with the treatment service and perception that the treatment was less organised and having less behavioural tasks. In the second study we used a naturalistic follow-up design. Forty-six children were included. The overall dropout rate was 48%. Again, the children who defaulted were rated by clinicians as less likely to have improved and dropout was also significantly associated with parental perception of a less organised treatment. In both studies dropout usually occurred after assessment and at the early phase of treatment.

The learning organisation model offers a more comprehensive framework for designing adaptive mental health services and supporting quality management practices than any other recent organisational form.

As half of all mental disorders appear during childhood and early adolescence, more research into the origin and emergence of these problems should focus on this early period of life. Prevention research should also focus largely on children and adolescents, with interventions targeted through sequential assessment at pivotal stages. The current problem of access to mental health care for youth in the adult mental health service system could be improved with additional funding and culture change - it is not necessary to create a new service system. Indeed, there are dangers in establishing new services for 12-25-year-olds, as the 12-17-year-old population has different needs from 18-25-year-olds. In particular, the younger group are at risk of 'adultification' by being grouped together with young adults, and this risk needs to be actively managed. Health service planning must pay attention to developmental differences. Several suggestions are proposed for addressing the mental health needs of young adults.