Thirty-eight depressed patients who were treated with ECT were randomly assigned to receive lithium therapy or identical-looking placebo tablets for one year after clinical recovery in a double-blind trial. The patients who received placebo tablets spent an average of 7.8 weeks with an episode of depression (either as in-patients or day-patients) during the year. In comparison, patients who received lithium spent on average 1.7 weeks with an episode (P <0.02). The trial confirms the high rate of relapses after ECT and suggests that lithium considerably reduces this morbidity. It is suggested that ECT without continuation therapy is not a satisfactory treatment of depressive illness.

Forty-one subjects from an earlier study, who had undergone computed tomography (CT) during their in-patient care for affective disorder, were followed up clinically for a minimum of one year. Mortality at two years was also determined, and comparisons made with 50 age-matched controls. Those with affective disorder showed a higher mortality than controls, but the difference just failed to reach significance. Within the affective group, however, mortality was significantly higher in patients who had previously shown ventricular enlargement on CT, confirming our earlier suggestion that these patients might have constituted a distinct subgroup. Failure of the affective group to improve performance on a simple cognitive test at follow-up was related to persisting depression. These and other findings are discussed.

The incidence of physical symptoms in depression was studied in 51 drug-free patients and in an age and sex-matched control group. Mean symptom intensity and number of symptoms were significantly higher in the patient than in the control group. The subjects' personality structure and relationships between their symptoms and degrees of depression and anxiety were examined using the Eysenck Personality Inventory (EPI), the Beck Depression Inventory, and the Spielberger State-Trait Anxiety Inventory, respectively. Of the variables studied, only the N score of the EPI was shown to influence the symptoms significantly. The effect of such non-specific factors as age, gender, use of alcohol, coffee, tea, and cigarettes was also evaluated and found to be minimal.

The validity of the concept of situational (i.e. reactive) depression as distinct from other major depressive subtypes is examined in terms of psychosocial stressors, personality features, current symptomatology, and clinical course and follow-up. Thirty-eight patients with a recent onset of situational major depressive disorder (i.e. the disorder developed after an event or in a situation the diagnostician deemed likely to have contributed to the episode at that time) were compared with 68 non-situational major depressive patients. These patients were participants in the clinical studies of the NIMH-Clinical Research Branch Collaborative Program on the Psychobiology of Depression. No significant differences between the two groups were found in the total number, content areas, or other categorizations of life events, experienced prior to onset. Some statistically significant differences in current symptomatology and in clinical course and follow-up measures were obtained, but there were none in personality traits. Implications of these results in relation to previously published reports are discussed and caution is recommended in the use of the term situational depression until more definitive data become available.

In an attempt to clarify the intentional aspect of parasuicide, constructs in response to a supposed crisis were elicited from a group of overdose patients, using the standard method of contrasting elements. In this case the elements were presented as a list of 11 alternative behaviours (including overdose and suicide) possible in such a situation. Using 9 of the most common constructs obtained, a repertory grid was administered to a second group of overdose patients and by computer analysis consensus group grids were obtained for patients scoring high and low on the Beck suicidal intent scale. These consensus grids showed that the low intent group perceived an overdose (in comparison to other alternative behaviours) as similar to ‘being alone and crying’ and ‘getting drunk’ and construed it almost exclusively as an escape from tension. In contrast the high intent group perceived overdose and suicide in quite similar terms. The data indicated that an overdose may have a respite function for low suicidal intent patients.

Eighty-five schizophrenic patients were found to have worse two-year follow-up outcome (history of remission, clinical course, improvement in psychotic symptoms, general treatment result) than 39 non-schizophrenic psychotic patients. The outcome of 36 schizophrenic patients with derealization at initial evaluation, especially when nine hebephrenics were excluded, was better than that of 49 schizophrenic patients without derealization but similar to that of 39 non-schizophrenic patients. The hebephrenics with derealization demonstrated similar outcomes to schizophrenic patients without derealization. In non-schizophrenic patients, derealization had no prognostic significance.

Three new cases of the syndrome of Capgras are presented. An unusual variant of the syndrome is recorded. It is contended that the definition of the syndrome should be widened to include impersonators who exhibit slight differences from the prototype, and that ambivalence and projection cannot explain all cases. The syndrome is critically reviewed.

As part of an epidemiological survey of a rural community in the North of Spain we analysed the interaction between physical illness and neurotic morbidity using a two-phase cross-sectional analysis of a representative sample of the population over 15 years old. There was a very high prevalence of neurosis in the somatically ill (males—15.6 per cent, females—33.1 per cent) and much physical pathology in the group of neurotics (males—42.4 per cent, females—39 per cent). The personal characteristics of the sample, as well as some sociological factors like rural-urban way of life, educational level and social class, appear to influence the relationship between both types of illness. The neurotic morbidity of the somatic group comprised depressive, anxiety and hypochondriac types.

The Middlesex Hospital Questionnaire (MHQ) was used as a screening test for psychiatric disorder in 169 new outpatients. The profile obtained on the six subscales of the MHQ was strikingly similar in this American sample compared to four previous British reports. The MHQ significantly differentiated between diagnostic groups, most particularly between neuroses and personality disorders. Moreover, 75 per cent of the patients could be correctly classified as either neurosis or personality disorder on the basis of their MHQ total and subscale scores. The MHQ appears to be particularly useful in identifying phobic disorders, and the phobia subscale consistently discriminated between anxiety-phobic states and other diagnostic groups.

Theories on the classification and aetiology of male homosexuality are reviewed, particularly recent hypotheses on the role of prenatal hormonal influences on brain sexual differentiation and subsequent sexual object choice in the male. Female as well as male brain sexual differentiation may be hormonally determined, and so primary homosexuality in both sexes may be due to abnormalities in foetal exposure to hormones, leading first to physical mis-differentiation and later to homosexual behaviour in genetically and phenotypically normal men and women.

The outcome in clinical and social terms of a psychiatric illness arising within 1 year of confinement or termination of pregnancy and requiring on-going care was studied after a period of 1–6 years in 47 women. The outcome was related to the earlier features of the illness and to personal and historical features. Social impairment and continuing psychopathology occurred in many cases. The only factor which significantly related to poor outcome was an initial diagnosis of schizophrenia.

This paper describes a patient who has had two episodes of mania, both of which occurred immediately after weaning. Although no recent reference to weaning as an aetiological factor in psychiatric disturbances could be found, it is suggested that weaning was an aetiological factor in this case.

In the last twenty years many workers have applied multivariate statistical techniques to problems of classification in psychiatry. To what extent have the techniques resolved the problems? This personal view will attempt to appraise the current situation.