The paper illustrates drawbacks of the popular cross-over design in clinical trials in psychiatry where treatment effects become distorted by, and confounded with, the effects of their order of administration. It is argued that research workers should analyse data from cross-over trials for asymmetries due to order of treatment and seriously doubt the reliability of data if such effects are found. The problem is avoided by the use of separate-groups designs.

A year's follow-up study of day-hospital long-attenders underlined doubts as to the adequacy of service provision in urban areas of England and Wales for the increasing numbers of chronic psychiatric patients living in the community. Very few of the 82 men and 53 women were married or capable of employment, and both psychiatric and physical morbidity were high; 2 young men committed suicide and 3 elderly men died within a year of discharge. Over the age of 45 men tended to be socially isolated and over the age of 60 they were also physically ill; a number probably required better residential care. Few women under 45 attended, possibly because the services offered were unacceptable.

Relatives of 39 schizophrenic patients attending a day hospital in South Glamorgan assessed the patients‘ behaviour at home covering a period of one week. The results pointed to the existence of a sub-group of patients with severe behavioural problems. These patients had had more psychiatric admissions, and had spent a longer total time in hospital. The proportion of forms of behaviour regarded as problems by the relatives themselves was lower than that suggested by their assessments, except in the families where the patient exhibited severe behavioural difficulties. The findings are discussed in terms of the practical difficulties regarding home-management by relatives of the schizophrenic.

Fifteen mentally retarded Swedish women, judged to be a representative sample, gave birth to 41 children. These children were followed up at the age of 1–21 years with regard to registrations in the Social Register, the Board for Provisions and Services to the Mentally Retarded and the files of the Child and Youth Psychiatric Departments. One child had died, and of the remaining 40 85% had or had exhibited severe psychosocial problems according to register data. The implications of the findings are discussed.

A study is reported in which smokers are shown to have higher psychoticism scores than non-smokers. This result is discussed in the context of the comments made by Eysenck (1983) on an earlier paper of ours (McManus & Weeks, 1982).

This paper analyses patterns of episodes of anxiety and depression as recorded by 22 general practices which took part in the Second National Morbidity Survey (RCGP, 1980). The data comprise 6 years′ records. Simple Markov chains are fitted to each age-group of men and women, and the inter-practice variation in the resulting parameter estimates are compared, using Andrews′ plots and principal components plots. The implications for the presentation and interpretation of data from large-scale morbidity surveys are discussed.

The development and promotion of a ‘common language’ in the mental health field is a lasting objective of the World Health Organization. In the 1960s, experts from 35 countries participated in the WHO programme on the standardization of psychiatric diagnosis, classification and statistics which reviewed the state of the field and resulted in the glossary of mental disorders, first produced as a companion to ICD-8 and later incorporated, with modifications, in Chapter V of ICD-9.