The development of DSM-III is traced from the Feighner Criteria through Spitzer's Research Diagnostic Criteria. The specific criteria for the diagnosis of schizophrenia are listed, and the symptoms which contribute to the diagnosis are categorized. Differences from the DSM-II concept are highlighted, and the complicated question of schizoaffective disorder is discussed. The DSM-III resolutions to the major controversies in the study of schizophrenia are outlined. Finally, the major strengths and weaknesses of the DSM-III approach are assessed.

In a recent review of the structure and function of aminergic systems Iversen (1980) paid particular attention to the dopaminergic (DA) and noradrenergic neurones (NA) and only mentioned in passing those with 5HT as a transmitter. Although the original work on the low concentration of 5HIAA, the metabolite of 5HT, in CSF of depressed patients (Ashcroft et al, 1966) which brought serotonin to the attention of biological psychiatrists may not have stood the test of time, nevertheless, recent information on the behavioural consequences in animals of drugs and lesions which influence the output of the 5HT system suggests it plays a role in the development of pathological mood and behaviour in affective illness. This is reflected in the resurgence of clinical interest in antidepressant drugs designed to affect specifically the 5HT system and emphasizes the need to continue animal and clinical exploratory experiments on serotonergic function.

Three cases of monosymptomatic delusion of alimentary stench are described. Each patient was referred for behaviour therapy as a last resort, after extensive medical investigation for halitosis or flatulence and failure of other psychiatric treatment. Two of the patients showed clinical improvement after treatment sustained at follow-up. Monosymptomatic delusion is now a treatable condition and is important to diagnose.

Patients with anorexia nervosa may minimize their symptoms with the result that doctors often fail to appreciate the distress they experience. The drastic avoidance of food is passed off euphemistically as ‘dieting’, and in later stages of the illness the patient's suffering is increased by depression, loss of concentration and insomnia. Much misery is also endured by other members of the family who observe with dread the young girl's progressive wasting and appear overwhelmed by feelings of helplessness. The illness usually runs a prolonged course even when the outcome is eventually favourable. An uncertain but sizeable proportion of patients enter a chronic phase with persistent preoccupations about food, uncontrollable overeating, self-induced vomiting, and even more severe depression (bulimia nervosa). Anorexia nervosa is therefore an illness which exacts a high toll in terms of human suffering. For this reason and in view of an increased incidence of the illness (Kendell et al, 1973; Crisp et al, 1976), it is desirable to improve methods of treatment and apply those currently available with the utmost skill.

A wide range of risk-taking behaviours may be associated with serious damage to the self although not consciously related to a wish for self-destruction: car driving and chronic alcoholism are two examples. Here, however, suicidal behaviour will be taken to refer only to that motivated by a conscious and deliberate aim to destroy the self. There is much evidence that prevention of suicide is a feasible proposition. The majority of suicides make contact with helping agencies in the few weeks prior to their death, they declare their intentions in a variety of ways and they tend to be recognizably unwell. It is just not true that most suicides decide to end their lives without declaring their intention to others. It is therefore important to acquire skills in recognizing and reaching out to those at risk in such a way that hope is rekindled and help accepted.

This paper evaluates the predictive power of two classes of variables which are thought to be related to psychotherapy outcome: patient characteristics assessed at the start of treatment and process characteristics of the psychotherapy. The treatment was Short Term Interpersonal Psychoatherapy which was performed with ambulatory patients during a major depressive episode as defined by Research Diagnostic Criteria. Univariate correlations indicate that patients with generally healthier functioning on the Prognostic Index received better ratings on measures of psychotherapy process and experienced greater improvement in depressive symptoms and social functioning. However, in a multiple regression analysis, patient characteristics accounted for 64 per cent of the variance in treatment outcome, while psychotherapy process variables were not predictive of outcome.

Group psychotherapy is perhaps a misleading term and the concept of the group psychotherapies probably better represents the broad range of psychological treatments in psychiatry in which the group process is an essential component: from psychotherapy with a married couple to the large community group meeting in a psychiatric hospital. Group psychotherapy most commonly refers to the long-term treatment of neurotic out-patients in small groups, and this will be my main focus. Marital and family therapy have their own distinctive literatures, while the therapeutic community has been well covered in a previous column (Mandelbrote, 1979).

Few people (and the present writer is not one of them) will know enough about psychiatric books to be able to declare which are the greatest. My own list is a sample biased by chance, ignorance and prejudice. But within these limitations I have worked to a plan and chosen those books which seem to me to represent or reflect important stages in the historical development of psychiatry.