The clinical characteristics of 70 patients included in the Northwick Park ECT trial of real against simulated ECT were analysed to identify predictors of response to the two treatments. The initial agitated/deluded/retarded substratification, the initial assessment of delusions by PSE, the individual items and factors derived from the Hamilton depression scale were all evaluated, together with six scales previously held to predict response to ECT and the individual items of these scales. The limited size of the sample does not allow firm conclusions, but the most significant and only consistent predictor of response to real ECT appeared to be the presence of delusions. The features of ‘endogenous depression’ did not in themselves appear to predict response to real ECT. The findings are discussed in relation to the viewpoint that delusional depression may be a specific entity which is relatively resistant to tricyclic antidepressants but responsive to electroconvulsive shock.

The clinical features of 35 patients with bulimia nervosa are described. All the patients came from the Oxford area and each fulfilled conservative operational diagnostic criteria. Standardised assessment procedures were used including a structured interview designed to assess the psychopathology characteristic of patients with eating disorders. It was confirmed that these patients have grossly disturbed eating habits accompanied by morbid beliefs and values concerning their shape and weight. Although the majority had a weight within the normal range, a history of weight disturbance was common. There was a high degree of psychiatric morbidity with depressive symptoms being particularly prominent. A quarter of the sample had previously fulfilled diagnostic criteria for anorexia nervosa, and this group closely resembled those patients with no such history.

This study considered the relationship between the personal networks of parents of schizophrenic patients and ratings of their level of expressed emotion towards the patient. Our initial hypothesis was that parents with high expressed emotion would be more isolated from an effective social support system. We found that the size of parental networks did correlate with the length of the patient's illness, with smaller networks associated with longer illnesses. There was no connection between overall ratings of parental expressed emotion and the size or quality of their social networks. However, the raw score of one component of expressed emotion, emotional over-involvement, did correlate with several network variables, indicating that parents who are over-involved with the patient are also more involved with their networks rather than being isolated.

We are conducting a two year longitudinal study of mood disorders in a group of 103 stroke patients. During the first six month period following the stroke the prevalence of symptoms of major depression increased from 23 per cent to 34 per cent while the frequency of symptoms of dysthymic depression increased from 20 per cent to 26 per cent. In addition, 10 of 13 patients who had major depresssive symptoms during the initial evaluation continued to have these symptoms at six months follow-up and 4 of 9 patients who had minor depression in hospital developed major depression, while 5 of 9 continued to have minor depressive symptoms. Thus, the duration of depression following stroke is more than six months and the prevalence of major depressive symptoms increases steadily for the first half year after a cerebral hemorrhage or ischemic lesion.

Twenty-one schizophrenic patients from a depot injection clinic with frequent hospital admissions (relapsers) and 28 who had remained well for 18 months (non-relapsers) were compared using the Parental Bonding Instrument (PBI) and the Influential Relationships Questionnaire (IRQ), a modified version of the PBI. The scales measured patients' past and present representations of significant others, respectively. Results indicated that only present representations of significant others differentiated relapsers from non-relapsers.

Subjects were followed up for a 9 month period. The readmitted (n = 8) and the non-readmitted (n = 41) groups were found to be different on the maternal care scales of the PBI and four of the six scales of the IRQ. The perceived attitudes of mothers of non-relapsers and non-readmitted patients tended to improve over time. The importance of the state-trait distinction of perceived attitudes in relation to re-admission to hospital is discussed.

As part of a survey of 1011 men aged 20–60 years, a factor analysis and validation study of the General Health Questionnaire (GHQ) was carried out. Three clinically relevant factors could be isolated but factor analysis did not generally enhance the value of the GHQ.

When compared with results for a similar group of women, sex differences in answering the GHQ were found; the GHQ was perhaps a less satisfactory screening instrument in the men.

Forty-five agoraphobic out-patients were randomly assigned to treatment with imipramine or placebo, and also to brief therapist-aided exposure or relaxation. All patients did systematic self-exposure homework and recorded this in a diary. Forty of these patients were followed-up two years later with self-ratings and ratings by interviewers blind to their treatment conditions. About two-thirds of the patients remained improved or much improved in their phobias, with no significant difference between any of the four treatment conditions. Spontaneous panics also remained improved. The absence of an imipramine effect may reflect the lack of initial dysphoria (anxiety–depression) in this sample compared with other studies where drug-effects have been found. The post-treatment superiority (evident at week 28) of patients who had therapist-aided exposure was no longer present at the two-year follow-up; the others had caught up, presumably because of their self-exposure homework.

Involving the spouse as a co-therapist made no significant difference to the outcome of treatment in a controlled study of 19 married patients suffering from agoraphobia who were engaged in short-term home-based behavioural treatment. Improvement continued to 6-month follow-up. The marriages of those with some marital difficulties at the start tended to improve, whether or not the spouse was involved. Poor initial marital adjustment was not related to the degree of improvement in agoraphobic symptoms.

There is a remarkable lack of controlled investigations about the usefulness of psychotropic drugs in anorexia nervosa despite the popular use of neuroleptics in these patients. Inspired by a previous study using pimozide, a double-blind placebo-controlled cross-over study with sulpiride was carried out in 18 female anorexia nervosa inpatients. No direct effect of sulpiride has been established with regard to behavioural and attitudinal characteristics as measured by behaviour observation scales as well as by two self-report questionnaires (Eating Attitudes Test, Body Attitudes Test). Regarding daily weight gain, sulpiride was on the whole superior to placebo, especially in the first treatment period of three weeks. In the cross-over analysis, however, this effect did not reach statistical significance.

A survey of attitudes to psychosurgery among consultant psychiatrists in two regions of Britain was undertaken by a postal questionnaire. Most psychiatrists considered that there is still a role for psychosurgery in psychiatric practice and over 75 per cent of psychiatrists working in general adult psychiatry requested facilities for the referral of patients; the majority of those requesting preferred local facilities and thought that the psychosurgical team should carry out both the assessment and postoperative rehabilitation. Respondents to the questionnaire were conservative in their indications for considering psychosurgery. The effect that the revision of the Mental Health Act might have on psychosurgical practice was then considered and finally the organization of the authors' own practice is described.

We surveyed the psychotropic drug treatment of our hospital's 220 patients, and compared our findings with those of the only previous hospital survey in this country which showed much allegedly inappropriate prescribing. Using simple classifications of the patients' diagnoses and of the drugs prescribed, 30 per cent of our prescriptions appeared inappropriate in the absence of knowledge of the patients' symptoms. With such knowledge we show that only 2 of our 460 prescriptions were inappropriate.

In a consecutive series of 127 chronic schizophrenics admitted to hospital, 70 (55.1 per cent) had attempted suicide. When compared with the chronic schizophrenics who had not attempted suicide, significantly more of the attempters had had a major depressive episode (45.7 per cent) and had received antidepressant medication (65.7 per cent). They also had histories of significantly more psychiatric admissions (a mean of 5.4 compared with 3.6). These findings are discussed.

Eighty opiate takers attending a Scottish drug dependence clinic were studied on a number of characteristics. The more recent attenders since 1981 were younger but otherwise not different from longer-term attenders. Recent young heroin abusers may not be seeking treatment from the drug clinic.

90 patients with a primary depressive illness, admitted consecutively to one psychiatric ward were studied. All patients had the dexamethasone suppression test (DST) done before initial treatment. First treatment response was correlated with the DST results. The non-suppressors, as a group, improved with treatment better than the suppressors. However, unlike in previous studies, no relationship was found between non-suppression and poor response to antidepressant medication.

Forty-one demented patients admitted to a psychogeriatric assessment unit in 1977–78 were matched by diagnosis, sex and living group prior to admission with forty-one patients admitted to the same Unit in 1981–82. While the two cohorts showed no significant difference in their overall levels of disability at the time of admission, the latter group were more disabled at the time of discharge and were more likely to be placed in long-stay hospital care. The 1981–82 cohort stayed significantly longer in the assessment unit. These differences suggest that health and social service resources are failing to keep pace with the rising numbers of demented elderly patients.

Although chronic depressive illness has not attracted a great deal of attention, it is a source of considerable psychiatric morbidity in the community and occurs in a proportion of the new long-stay patients. Prior to the introduction of specific antidepressant treatments, figures for continued affective illness were high. Lundquist (1945) observed an overall recovery rate of 80 per cent for a first attack. Perhaps the best researched study is that of Winokur who followed up for between two and 20 years 225 depressive patients. He showed that 25–30 per cent of female depressives developed a chronic state, compared with 10 per cent of male patients (Winokur and Morrison, 1973).