An historical analysis is made of the word and of the concept of ‘dementia’ before the nineteenth century. With regard to the word, it is shown that it had legal and medical meanings and that, while the former developed during the seventeenth century, the latter did so only during the eighteenth century (earlier than psychiatric historians have suggested). As evidence for the latter point, rare historical material on ‘Démence’ from the first edition of the Encyclopédie Française is presented for the first time in English. It is also shown that the legal meaning was finally enshrined in the ‘Code Napoléon’. With regards to the concept of dementia, it is shown that it took final shape in the work of Willis, Hartley and Cullen in whose view it was made to include terminal states of behavioural incompetence due to severe failure of almost any mental function. During this period, dementia was not yet associated with a particular age group nor was specifically defined in terms of cognitive deficit. The origins of the ‘cognitive’ paradigm of dementia and of the clinical boundaries of the future concept of dementia are briefly outlined.

The increase in plasma prolactin which follows intravenous administration of L–tryptopha (LTP) was used to assess changes in brain 5-hydroxytryptamine (5-HT) function in normal male and female subjects, following a three week period of dieting. In women, but not men, there was a marked increase in the prolactin response to LTP, suggesting that dieting had caused alterations in brain 5-HT-mediated responses. In contrast, dieting did not alter the prolactinresponse to thyrotropin releasing hormone in either men or women, indicating that the changes in response to LTP could not be attributed to an increase in pituitary reserve of prolactin. These findings suggest that dieting alters brain 5–HT function in women but not in men. Biological factors as well as greater psychosocial pressures to diet may contribute to the high prevalence of eating disorders amongst women.

The psychobiology of anorexia nervosa is described and explained under four headings; (1) the psychopathology as related to the motivation for fasting; (2) metabolic and somatic consequences of starvation, including brain morphology; (3) endocrine abnormalities in the hypothalamic-pituitary-adrenal and gonadal axis; and (4) the hunger drive and its possible perversions in terms of aspects of neuroethology and the reward system in the brain.

Urinary excretions of arginine vasopressin (AVP), sodium, potassium, osmoles and creatinine were measured in three in-patients with bipolar manic-depressive psychosis on at least eight 24-hour periods in each affective phase. Mood and body weight were recorded twice daily. The excretion by each patient of sodium, water and osmoles was greater in mania than during depression. Comparison of electrolytes and osmoles suggested that the increase was due to increased intake of salt and water rather than of total diet. There was a fall of mean AVP excretion during mania, the magnitude of the fall being related to the increase of water throughput.

Compared with controls, AVP excretion was high and variable. It did not show the normal relationship to urine osmolality. Days with very high AVP were not associated with any characteristic feature of the other measurements; nor were they confined to any one point in the manic-depressive cycle.

AVP does not appear to play a major role in the salt and water changes characteristic of manicdepressive psychosis and we have no evidence of its having any direct relationship to mood changes. We suggest that the observed abnormalities of AVP excretion are another manifestation of the central defect of this disease.

In 17 benzodiazepine (BDZ) dependent in-patients a CT scan was performed before initiation of withdrawal therapy. The evaluation of the ventricular to brain ratio (VBR) by standardized and computerized measurements revealed significantly higher mean VBRs for both high-and low-dose BDZ-dependent patients compared to the mean VBR of an age- and sex-matched control group. In addition, the mean VBR of high-dose BDZ-dependent patients (N = 8) was significantly higher than the mean VBR of low-dose BDZ-dependent patients (N = 9). This difference could not be accounted for by the age of the patients or duration of BDZ-dependency and, therefore, suggests a dose-dependent effect of BDZs on the enlargement of internal CSF-spaces. On the other hand, higher values for the width of external CSF-spaces were found to be related to increasing age of the patients and duration of BDZ-dependency.

Platelet-rich plasma from healthy controls was pre-treated with neuroleptics of the phenothiazine, butyrophenone or benzamide variety before aggregation with one of the following agonist agents: ADP, adrenaline, 5-HT, collagen, platelet activating factor or ristocetin. All compounds effective as antipsychotics, except sulpiride, depressed aggregation. Unmedicated schizophrenics showed aggregation responses indistinguishable from healthy controls. However, within days of treatment with either trifluoroperazine or haloperidol responses became abnormal in acutely psychotic patients. Increased responses to 5-HT and depressed responses to platelet activating factor were detected. After 4 weeks of treatment responses tended to return to normal. Aggregation responses were normal in those patients on long-term depot neuroleptics.

Using the family history method, we assessed the morbid risk for psychiatric disorders the first-degree relatives of 69 probands with bulimia, 24 probands with major depression, and nonpsychiatric control probands. The morbid risk for major affective disorder among the first-degree relatives of the bulimic probands was 32%, significantly greater than that found in the nonpsychiatric control probands. The rate of familial major affective disorder was significantly greater in bulimic probands who had a history of major affective disorder themselves than in bulimic probands without such a history - but the latter group, in turn, displayed significantly higher rates than the nonpsychiatric control probands. Eating disorders were slightly, but not significantly, more prevalent in the families of bulimic probands than nonpsychiatric control probands. We present two alternative hypotheses which might explain these findings.

Bulimic and non-bulimic anorexics were compared on psychological variables during a hospital treatment study. Although before treatment bulimic anorexics displayed more overall psychopathology than non-bulimics, many of the differences disappeared with treatment. There was no difference in severity of depression or body size estimation in these groups both before and with treatment. There was no difference in treatment response as measured by rate of weight gain. More expression of discomfort by bulimic anorexics during the acute phase of illness may in part account for some described differences in these two groups.

Mood swings accompanying the motor fluctuations of patients with Parkinson's disease on chronic levodopa treatment frequently occur, but are poorly recognized. Occasionally, their functional impact may be greater than that caused by the motor disability itself.

In this study we have assessed the nature of, and relationship between, mood and motor fluctuations in nine Parkinsonian patients with ‘on—off’ motor swings. The results of an additional questionnaire survey confirm that ‘on–off’ mood swings occur in approximately two thirds of patients with Parkinson's disease experiencing motor fluctuations on dopaminergic treatment. Aetiological and therapeutic implications are discussed.

Before treatment 49 obsessive-compulsive (o-c) ritualizers were presented with two series of brief stimuli – 15, 100db tones (brief neutral) and 15 presentations of a ritual-evoking object (brief aversive). Habituation of skin conductance (SC) responses to the tones was reduced compared with that previously found in normal subjects. Neither habituation rates to tones nor aversive stimuli were related to coexisting anxiety or depression or to the severity of o-c symptoms. The increased arousal induced by the aversive stimuli was sustained, that induced by the tones was short-lived and SC level and subjective anxiety had returned to resting levels by the end of the tone series. Concordance between SC activity and subjective anxiety was much greater during and after the presentations of ritual evoking stimuli than of tones. There were few correlations between SC and clinical measures, though patients who strongly resisted and were able to control their compulsive urges were more aroused.

Reported cases of violence during sleep are reviewed. Two new cases are reported where the subjects were charged respectively with attempted murder and murder; both were acquitted on the grounds of non-insane automatism. Clinical features and aetiological factors are analysed. The legal concepts surrounding non-insane automatism are discussed and criticized.

Body image in 38 obese and normal weight subjects was measured using a video TV monitor. Tasks included a continuous task where subjects manipulated the size of their body image on a TV screen, and a discrete task, where they judged the accuracy of their TV image which was presented as either too heavy or too thin. Subjects viewed their images in both regular street clothes and with minimal clothing.

Results from the continuous task revealed that obese subjects overestimated body size. All subjects were more accurate on trials in which the initial image was adjusted in the heavier direction.

On the discrete task, a signal detection analysis revealed sensory sensitivity differences between obese and normal weight subjects. Obese subjects were better at detecting distortion caused by a too thin image, but were inferior in detecting a too heavy distortion. No significant differences were obtained in response criteria or between clothing conditions.

In an earlier paper (Goldberg et al. 1987) 36 common symptoms of minor psychiatric disorder in general practice were analysed using the technique of latent trait analysis. From this analysis two dimensions of illness emerged, corresponding to anxiety and depression. In the present paper, this symptom-based representation of minor psychiatric illness is used as a framework for comparing four diagnostic systems: General Practitioner (GP) diagnoses, the ID-CATEGO diagnostic system, the DSM-III system and the Bedford College diagnostic system. This analysis clarifies the reasons for disagreement among systems of diagnostic criteria and examines the practical effects of alternative diagnostic algorithms.

This paper reports an investigation of the incidence of schizophrenia in Ireland by means of standardized case-finding procedures. Our findings do not support the popular belief, based on first admissions figures, of a raised incidence in Ireland. We show this to have been considerably overestimated.

A considerable amount of research documents the negative effects of job loss on both physical and mental health. Yet we know comparatively little about the mechanisms through which these effects occur. Unemployment, like other events, is not the same experience for everyone exposed to it. An understanding of this variation might be facilitated by breaking down the analysis of unemployment into a consideration of the various stresses that it creates or exacerbates. This is our purpose in the present paper.

We demonstrate that, for one area of the United States, the effect of job loss on several health outcomes involves two mechanisms: (1) unemployment results in increased financial strain which, in turn, results in negative health effects, and (2) unemployment leaves the individual more vulnerable to the impact of unrelated life events. Controlling for financial strain, unemployed people in our sample who have not experienced an additional life event in the previous year are in no worse health than the stably employed. This provides useful insights into the nature of the unemployment experience in this particular setting. It also provides a basis for future detailed explorations of the various ways people cope with this event.

Four groups of families were investigated using an interview designed to assess expressed emotion (EE) in relatives of psychiatric patients and a family interaction questionnaire (FIQ) based upon the Leary and Benjamin schemes for coding interpersonal behaviour. The families included 25 containing a psychiatric patient aged between 18 and 46, and three groups of 12 families each, all containing a patient over 60 (patients with dementia, a functional psychiatric disorder, or a chronic physical disorder respectively). In all cases data were obtained from key relatives.

The FIQ clearly separated out the group of dementia relatives who reported the most dominant and protective behaviours and the highest levels of hostile-dominance and the lowest levels of affection. EE was not found to be so sensitive to interactions occurring in families with a member with dementia, and the possible reasons for this are discussed.

We report the development of a new procedure for assessing the needs for treatment and care of the long-term mentally ill. This procedure covers 21 areas of clinical and social functioning, and in each of these specifies appropriate interventions. Decision rules are described which permit problems in functioning to be primarily classified as a met need, an unmet need, or as involving no need, and which allow the identification of various secondary needs. We report preliminary data on the reliability and validity of this procedure and discuss its potential applications in the care of the long-term mentally ill.