Patients with panic disorder plus agoraphobia had 8 weeks of drug treatment (alprazolam or placebo) plus psychological treatment (exposure or relaxation). At the end of treatment at week 8, 40 patients who had become much/very much improved rated how much their gains were attributable to medication or to their own efforts. During the tapering-off to week 16, and treatment-free follow-up to week 43, patients who at week 8 had attributed their gains to medication and felt less confident in coping without tablets had more severe withdrawal symptoms and greater loss of gains than did patients who at week 8 had attributed their gains to their own efforts during treatment. Baseline illness severity, greater age, higher expectations from drug treatment, and more side-effects of drugs during treatment all predicted more external attributions (i.e. to the effect of drugs) but did not independently predict relapse. Patients on alprazolam compared with placebo had more drug attributions. Though drug attributions predicted relapse in both alprazolam and placebo groups, predictions were stronger in the alprazolam group.

Most clinical and genetic evidence suggests that puerperal psychosis is closely related to bipolar affective disorder. During a linkage study of bipolar disorder we ascertained a British family in which puerperal psychosis was associated with consanguinity in three sisters. All three subjects had lifetime RDC diagnoses of bipolar I or manic disorder. An inbred brother also had bipolar I disorder. The only female member of the sibship to escape puerperal psychosis was outbred. These findings are consistent with several genetic models for bipolar disorder in this family. The most interesting possibility is a single major susceptibility locus of recessive effect. Under this assumption, the family could be used for homozygosity mapping to help localise the putative recessive locus. If other inbred families can be found in which the same putative recessive locus is operating, the power to localise the gene by homozygosity mapping would be greatly increased.

Long-term outcome studies for people suffering from bulimia nervosa are few. Ten years after presentation, we followed up 50 patients (49 women, 1 man) who were originally involved in a double-blind, controlled trial of the antidepressant mianserin. Standardised interviews and questionnaires were used to assess eating attitudes and behaviour, psychiatric status and social functioning. A DSM–III–R diagnosis was given where appropriate. Sufficient information to make a diagnosis was obtained for 44 subjects (88%). Of patients traced, 52% had recovered fully and only 9% continued to suffer the full syndrome; 39% continued to experience some symptoms. Significant predictors of favourable outcome were younger age at onset, higher social class and a family history of alcohol abuse. Outcome for bulimia nervosa continues to improve over ten years with the majority of patients eventually making a full recovery or suffering only moderate abnormalities in eating attitudes. Although predictors of recovery were few, it would appear that intervention has a significant impact on ultimate outcome.

Nefazodone is a phenylpiperazine antidepressant with 5-HT2 antagonism and 5-HT reuptake inhibition. Two hundred and eighty-three out-patients with a diagnosis of DSM–III–R major depression of at least one-month duration (65% ill for over 6 months), and a mean score of 24 on the 17-item Hamilton Rating Scale for Depression (HRSD), were randomised to treatment with nefazodone, imipramine, or placebo. The double-blind treatment period was 8 weeks in duration. Nefazodone's antidepressant efficacy was comparable with imipramine's, with both drug treatments significantly better than placebo in a variety of outcome measures. For example, after 8 weeks of therapy, 78% of nefazodone and 83% of imipramine but only 55% of placebo patients (P < 0.01) were globally much or very much improved. Nefazodone was better tolerated than imipramine, with fewer drop-outs and a lower incidence of side-effects during treatment.

We report the efficacy of combined chlorpromazine and electroconvulsive therapy (ECT) in the treatment of mania. Two groups of 15 manic patients received eight ECT sessions either actual or simulated, in a double-blind, controlled study. All patients also received 600 mg of chlorpromazine daily until the sixth session. Results indicate that the group receiving the combination of chlorpromazine and ECT did significantly better than the other group.

We contacted and interviewed 150 cocaine users in south London community settings. Most were taking cocaine regularly, but not daily, and in substantial doses. Three main routes of administration were used: smoking (40%), intranasal (32%), and injecting (24%). Injectors reported having used cocaine more frequently, in higher doses and for longer periods of time. Despite their frequent and extensive use of cocaine, the overall level of dependence was low for the great majority of the sample. Two-thirds (66%) of the full sample (and two-thirds (65%) of the crack smokers) reported only minor signs of dependence during the year prior to interview. Route of drug administration was related to severity of dependence. Cocaine taken by injection was associated with the highest levels of dependence; intranasal use was associated with the lowest levels, and crack smoking was intermediate between the two. There had been substantial changes in the initial route by which cocaine was used, with a clear trend away from injection on the first occasion of use. Those who first used cocaine before 1986 were more likely to have used the drug either intranasally or by injection. Cocaine users who first tried the drug after 1987 were increasingly likely to take it by smoking crack. Women and black (predominantly Afro-Caribbean) subjects were more evident in this sample of cocaine users than has previously been reported in many UK studies.

Abnormalities in the hypothalamic–pituitary–adrenal axis (HPA) have been the most consistently demonstrated biological markers in depressive illness. Numerous other neuroendocrine disturbances have also been described, including blunted clonidine-induced growth hormone release and blunted fenfluramine-induced prolactin release. These disturbances are generally interpreted in terms of monoaminergic receptor dysfunction. The theory presented here suggests that chronic stress which activates the HPA will in certain susceptible people produce changes in central monoamines. The high level of glucocorticoid receptors on such central neurons is postulated as mediating the alterations. Thus monoamine abnormalities, rather than being a core aetiological feature of depression, are seen as secondary to HPA overdrive.

To test further the highly successful outcomes of a controlled study of in-home behavioural family management (BFM) for schizophrenic patients, a clinic-based version of this intervention was compared with customary care alone for 41 schizophrenic patients in a Veterans Administration (VA) mental health clinic. Monthly Brief Psychiatric Rating Scale (BPRS) ratings, conducted by clinic psychiatrists who were ‘blind’ to the patients' assignment, revealed that 3 (14%) patients who received behavioural family management as well as customary care, as compared with 11 (55%) patients who received customary care alone, had symptomatic exacerbations during the first year of treatment.

Forty-eight depressed panic-agoraphobic patients - clinically matched for severity of depression with 35 primary chronic depressives - had significantly better adjustment in most areas of social functioning, especially those related to interpersonal contact, work, leisure activities, and sexual life. Even when depressed, panic-agoraphobic patients appear to possess sufficient interpersonal skills - that is, by leaning on significant others - to engage in a range of social activities, both at home and beyond. These data, which are open to different interpretations, nonetheless suggest that anxiety and mood disorders differently affect certain key areas of social adjustment.

As schizophrenia is a heterogeneous condition with a variety of possible causes, it is sensible to combine biological and social treatment. Neuroleptic drugs give partial protection against environmental stress, but need to be supplemented. The programme developed on the basis of research on relatives' expressed emotion begins with education about schizophrenia for patients and relatives. Family sessions in the home are run in parallel with a relatives' group, and are aimed at improving communication, teaching problem-solving skills, lowering criticism and overinvolvement, reducing social contact between patient and relatives, expanding social networks, and lowering expectations. Individual work with the patient is often undertaken alongside the family sessions, and includes social-skills training, anxiety management, and a cognitive approach to persistent psychotic symptoms. Therapists take a central role in coordinating the variety of services needed by patient and relatives.

Analyses were undertaken to examine whether a wide range of psychiatric disorders, including anxiety, affective, substance abuse and psychotic disorders, represent variant manifestations of Tourette's syndrome (TS). Previous studies have suggested that chronic tics (CT) and obsessive-compulsive disorder (OCD) are variant expressions of TS since both CT and OCD are elevated among relatives of TS probands. In the current study, no other psychiatric disorder was significantly elevated among the relatives who did not have TS, CT or OCD when compared with a control sample. These findings are not consistent with the hypothesis that a wide range of psychiatric and behavioural disorders are variant expressions of TS.

The World Health Organization's International Pilot Study on Schizophrenia (IPSS) and subsequent Determinants of Outcomes of Severe Mental Disorder (DOSMD) studies have reported that the course of schizophrenia is more favourable in ‘developing’ than ‘developed’ societies. The 1992 DOSMD study attributes this difference to culture. We review studies concerning the course of schizophrenia and conclude that the evidence for a more favourable course in developing societies is not conclusive. Indeed, a favourable course has also been reported in various industrialised societies. We also raise questions about the findings reported in the recent DOSMD study, including the conclusion that the putatively more favourable course is a product of culture. Finally, we argue that longitudinal, direct observation of patients in their natural environments must be carried out before variations in the course of schizophrenia can be adequately understood.

There has been increasing public concern about the risks of high-dose antipsychotic (neuroleptic) treatment, arising in part from an, as yet unproven, association between high-dose treatment and death in a small minority of patients. The clinical issues related to the use of neuroleptics in doses exceeding the maximum recommended in the British National Formulary (BNF) were discussed at the Psychopharmacology Subcommittee. When, if ever, should the recommended doses be exceeded?

The hypothermic, growth hormone and corticotrophin (ACTH) responses to the 5-HT1A receptor agonist buspirone (30 mg orally) were measured in 20 unmedicated patients with major depression and 20 healthy controls. Compared with the controls, the hypothermic responses of the depressed patients to buspirone were significantly attenuated, particularly in patients with melancholic depression. In contrast, the responses of growth hormone and ACTH to buspirone were unchanged. The data suggest that major depression may be associated with impaired sensitivity of 5-HT1A autoreceptors but that the function of the post-synaptic 5-HT1A receptors that mediate growth hormone and ACTH release is unaltered. Within the limitations that attend the use of buspirone as a 5-HT1A probe, our data suggest that the decrement in serotonin neurotransmission at post-synaptic 5-HT1A receptors in depression is due to decreased serotonin release rather than impaired responsivity of post-synaptic 5-HT1A receptors.

The association of attentional, neuropsychological, and behavioural abnormalities with Tourette's syndrome (TS) suggests that the abnormal function of the disorder extends beyond the motor circuits of the basal ganglia. To explore this possibility we studied, with conventional 18-channel electroencephalography, monozygotic twins ranging from 8 to 26 years of age, where at least one member of the twin pair suffered from TS. In nine out of the 11 twin pairs that differed in clinical severity of the tic disorder, the twin with the more severe course of illness had a significantly more abnormal electroencephalogram (EEG) by qualitative visual analysis. Most of the differences were due to excessive frontocentral theta activity, suggesting dysfunction outside the basal ganglia. There was also a significant relationship between a lower global neuropsychological testing score and a worse overall EEG. In eight of nine twin sets with different global neuropsychological testing scores, the twin with the lower score had a worse EEG. A similar relationship was found between birth weight and overall EEG quality. In the nine sets that differed in birth weight, the twin with a lower birth weight had a worse EEG in seven of the sets. The EEG findings are unlikely to be a medication effect because the same result was seen in the six twin pairs who had been medication-free for at least six months before entry into the study. The origin of this slowing may relate to the interaction between environmental insults to the central nervous system and the genetic component of TS, an interaction producing damage to the cortex, thalamus, or both.

Chromosomal abnormalities associated with bipolar disorder may help in the localisation of susceptibility genes for bipolar illness by pinpointing ‘candidate’ regions of the genome for further study using molecular genetic methods. We review descriptions of chromosomal abnormalities in association with bipolar and related affective disorders and evaluate their relevance for localising susceptibility genes for bipolar disorder, using standardised criteria. We found 28 reports. We identified four genomic regions of potential interest: 11q21-25; 15q11-13; chromosome 21; Xq28. It is important that clinicians are able to recognise patients who may have chromosome abnormalities which could help in the localisation of susceptibility genes for psychiatric disorders. We suggest referral for specialist investigation and karyotyping, to a psychiatric genetics research group, of any patient with functional psychosis and one or more of the following: (a) a strong family history of functional psychosis; (b) mental retardation; (c) another disease known to be caused by a single gene; or (d) congenital abnormalities.

People with schizophrenia lack the ability to develop – to differentiate and integrate – their self- and object-representations, and suffer from primitive ‘object-relations’ conflicts, which occur when they try to develop (to differentiate and integrate) their self- and object-world. When a therapist interacts beneficially with a schizophrenic patient and enables him/her to identify with the ego functions involved in this interaction, the patient's frail psychic structure receives nourishment that will strengthen it: this process is similar to human development, where a child attains psychic organisation by interacting with the one who nurtures him/her. The recommended approach in the psychoanalytic psychotherapy of schizophrenia is to ‘allow’ the natural evolution of the fusion–defusion and introjection–projection processes to appear in the experiences of transference and counter-transference.

The purpose of this study was twofold: to measure the overall direct costs of depression for 1990 in the UK, and to develop a model to illustrate issues in the evaluation of the relative cost-effectiveness of the pharmacological treatment of depression. We compared a tricyclic antidepressant, imipramine, with paroxetine, a newer antidepressant. For assessing the cost of illness, we used a top-down approach. We calculated direct but not indirect costs. Cost-effectiveness was evaluated by developing a simulation model based on the theory of clinical decision analysis to compare the costs and outcome of each treatment. From this we estimated the expected cost per patient and the cost per successfully treated patient. The total cost to the nation of depressive illness was estimated to be £222 million. The expected costs per patient were found to be similar for paroxetine and imipramine (£430 v. £424). The costs per successfully treated patient were found to be lower for paroxetine (£824) than for imipramine (£1024). The results were stable when a sensitivity analysis was applied to the variables employed in the model. The most sensitive variable was the cost of treatment failure. Our model thus reveals that medication that appears expensive in terms of cost per day may not be so when patient compliance and the total costs of treatment are taken into account.

All new attenders at the Menopause Clinic in Edinburgh over six months were interviewed to detect current depressive disorder (MADRS) and past psychiatric disorders (SADS-L), to find out whether women who were depressed at the time of clinic attendance had a history of depression. Of the 95 subjects who entered the study, 78 had gone through a natural menopause and 17 had undergone hysterectomy with or without oophorectomy. Of the 78 who had experienced a natural menopause, 35 were found to be depressed at the time of clinic attendance and 43 were not. A strong association was found between current and past depressive illness, 29 of the patients depressed at the time of clinic attendance having had depression previously. However, a clear peak of illness was seen in the perimenopausal period (four years either side of the last menstrual period): 35% of all patients with past or current depressive illness experienced their first episode of illness in this period.