Concerns have been expressed that staff burnout may make community mental health care difficult to sustain. This study compares stress and job satisfaction between community and hospital-based staff.

The GHQ-12, the Maslach Burnout Inventory and a job satisfaction measure were used to study 160 Inner London staff.

Community staff scored significantly higher on the GHQ-12 and the ‘emotional exhaustion’ component of the Maslach Burnout Inventory than hospital-based in-patient, day care or out-patient staff. Satisfaction did not vary significantly between settings.

These results may be explained in several ways. Community work may be inherently more stressful than hospital work, or may currently be stressful because of inadequate resources, training or supervision. The results may also reflect widespread recent changes in community services or the specific effects of working in a deprived area.

The aim was an investigation of the symptom composition of the negative syndrome of schizophrenia.

A hundred schizophrenic patients were assessed with various rating scales for positive and negative symptoms and typological criteria. A subgroup with a negative syndrome was defined, and a discriminant analysis performed using symptoms from the Scale for the Assessment of Negative Symptoms (SANS) as discriminant variables.

The discriminant function classified 92% of patients correctly. All but four SANS symptoms (inappropriate affect, both attentional items and blocking) were significantly correlated with the discriminant function.

With the exception of some items, the SANS seems to capture adequately the negative syndrome of schizophrenia.

Secondary depression is common among primary alcoholics. However, its aetiology is poorly understood.

Forty men with primary alcoholism, abstinent for two weeks or more, with a secondary major depressive episode were compared with forty matched never-depressed primary alcoholic controls for risk factors for depression and recent life events.

In the six months before the onset of depression depressed alcoholics had experienced significantly more life events, more life events with negative impact more dependent events caused by alcohol, and more independent events. Significantly more depressed alcoholics than controls had a family history of depression, a family history of suicidal behaviour and had themselves attempted suicide.

Having recent life events, particularly events with negative impact, and a family history of depression are risk factors for secondary depression in alcoholics.

This paper considers the extent to which vulnerability to the outcome of major affective disorder is conferred through personality deviance. Results are based upon a 12 year longitudinal follow-up study of 80 patients with a primary depressive disorder, originally selected from a consecutive series seeking care at a hospital in Scotland.

The study included detailed clinical course assessments in association with the formal application of diagnostic criteria. Assessments on recovery from the index episode included measures of self-confidence and of neuroticism. Experience of selected severe loss events over the study period was also determined.

Limited self-confidence was strongly related, unlike neuroticism, to the subsequent first recurrence of affective disorder. Relationships between psychosocial, clinical and demographic factors and long-term outcome revealed psychosocial factors, in particular neuroticism and a lack of self-confidence, to have the greatest prognostic significance.

These results reveal the heightened risk over the long-term of a poor outcome for depressive disorder consequent upon measures of personality deviance and of exposure to adversity. While giving only limited support to narrowly defined psychosocial models of depression, they clarify the risk gradients involved and through this may provide a firmer basis than hitherto for relapse prevention.

Alcohol and drugs use were assessed in a nationwide one year suicide study in Finland.

Alcohol and drugs use were analysed in 1348 suicides, 96.5% of all suicides in the study year. Relative suicide risks for drugs were defined by relating the number of suicides committed by use of various drugs to drug availability.

Alcohol was detected twice as often in men as in women; the opposite was the case with drugs. The drugs most commonly used for suicide were neuroleptics and antidepressants, which were, in 74.3% and 77.4% of the cases, respectively, the victim's own prescribed drugs. Antidepressants were found in 19.0% of women and only 4.8% of men. Relative suicide risk for antidepressants varied substantially between different compounds.

Undertreatment of depression is a challenge for suicide prevention. Those who commit suicide by antidepressants use their own drugs. Relative suicide risk for a drug should be considered when choosing treatment for depressive patients.

Stability of neuroleptic medication has been associated with optimal clinical effect and minimal side-effects. Depot administration is assumed to yield better stability.

Thirty patients on depot neuroleptic treatment were followed during three years with repeated measurements of plasma level and concurrent ratings of clinical symptoms and side-effects.

Of 120 blood samples 35 (29%) measurements were outside ± 2 s.d. measurement error (expected 5%). Perphenazine levels were more variable (46%) than haloperidol (25%) and flupenthixol (12.5%). No relationship was found between side-effect ratings and fluctuations of plasma levels.

Depot neuroleptic medication does not eliminate a clinically unwanted and sometimes marked variation in plasma level.

This trial evaluated the feasibility, acceptability and effectiveness of a structured approach to the management of schizophrenia in general practice.

All patients with non-affective psychosis (mainly schizophrenia) in four inner-city general practices were recruited. A check-list and a set of outcome measures were used by the general practitioner and the practice nurses. Information on attendances at the general practice and psychiatric out-patient departments was also collected.

Two practices took part in the intervention and two served as control practices. Sixty-seven patients with non-affective psychosis were identified. Thirty-three (81%) of the 41 patients in the two intervention practices attended the initial assessment by the doctor and nurse, but only 13 (32%) attended the first review assessment. The attendance for the second review, after six months, was six out of 15 (40%) in one practice, but rose to 16 out of 18 (89%) in the other practice. Significant improvements were recorded in the intervention group on the Global Assessment Scale (GAS) and the Behaviour, Speech and Other Syndromes (BSO) subscore of the Present State Examination (PSE). The absolute risk reduction and relative risk reduction as a result of the intervention as measured by the GAS scores, was 29% (95% CI 4% to 54%) and 36% (95% CI 5% to 66%), respectively, and in the case of the BSO subscores of the PSE, this was 23% (95% CI – 1.8% to 47.2%) and 28% (95% CI – 2.2% to 57%), respectively. For one patient to show an improvement in GAS and BSO scores 3.5 patients (95% CI 1.85 to 25) and 4.3 patients (95% CI – 55 to 2.1), respectively, would need to receive the intervention. There was a significant increase in consultation rates for patients in the intervention practices.

Health surveillance of patients with non-affective psychosis is possible in general practice.

For many common medical and neuropsychiatric disorders, early age at onset reflects high familial liability to illness. However, for schizophrenia, most studies do not find such a relationship.

Using Cox proportional hazard models, we investigate this question in the epidemiologically-based Roscommon family study.

No relationship was found between age at onset in schizophrenic probands and the hazard rate for schizophrenia in their relatives. Similar results were obtained when the definition of illness was expanded to include schizoaffective disorder and other non-affective psychoses.

For schizophrenia, a ‘common-sense’ model for age of onset (i.e. those with highest familial liability to illness succumb first while those with lower liability survive longer before falling ill) does not seem to apply. Our results are more consistent with a model in which variation in age at onset of schizophrenia is due to random developmental effects or to environmental experiences unique to the individual.

Rates of self-poisoning fell during the 1980s, although recent reports suggest this trend may have reversed.

Data on all hospital discharges with a diagnosis of deliberate self-poisoning were obtained from the Information and Statistics Division of the NHS in Scotland. Rates of self-poisoning were investigated by overall trend, and for trends by age and by type of drug.

Deliberate self-poisoning rates for men and women began increasing in the late 1980s, and did so steadily during the early 1990s. Because the rates are increasing faster in men, the traditional excess of self-poisoning among women is being eroded. For men rates increased in all age groups up to the age of 60, but the greatest rise occurred between ages 15–29. For women the increase was largely restricted to 15–24-year-olds, and rates fell among women over the age of 50. Self-poisoning with paracetamol increased most rapidly; in contrast to aspirin which showed little change. Opiate analgesics, antirheumatics, antidepressants and antipsychotics have also shown some increase in their use in both genders. There has been a substantial fall in the use of benzodiazepines in women, but little change among men.

Rapid increases in self-poisoning particularly among young adults present a serious public health problem. Controlling this epidemic is made difficult because the principal drug involved, paracetamol, is readily available.

Clozapine can cause reversible agranulocytosis and neutropenia. This study documents the occurrence of blood dyscrasias and identifies predisposing risk factors.

An analysis was made of the haematological, demographic, and dosage data from a central database on 6316 patients receiving clozapine over four and a half years in the UK and Ireland.

During the study period 2.9% of the patients developed neutropenia and 0.8% developed agranulocytosis. The peak incidence of both disorders was in the first 6–18 weeks of treatment. Fatal agranulocytosis occurred in 0.03% of patients. After the first year of treatment the incidence of agranulocytosis significantly decreased to the order noted with some phenothiazines.

The use of a patient monitoring service kept the haematological risks associated with using clozapine within acceptable limits, particularly in view of the benefits of this medication in treatment-resistant schizophrenia.

Parental emotional overinvolvement (EOI) may entail a worse outcome in schizophrenia. In the present study we examined demographic and clinical predictors of EOI.

The predictors were examined in a Norwegian sample of 41 recently admitted patients (schizophrenia or schizophreniform disorder) and 66 parents. Parents' expressed emotion was assessed by the Camberwell Family Interview.

Regression analyses showed that higher EOI was significantly related, on the part of the parent to being a mother, single, spending more time with the patient; and, on the part of the patient, to no substance misuse, more anxiety–depression, and less uncritical and aggressive behaviour. EOI was not linked to previous hospital admissions.

Our analyses indicate that characteristics of the parent and of the parent–patient dyad seem to be the most important determinants of EOI. EOI is probably not linked to psychotic relapse, but rather to affective disturbances in the patient.

We examined the functional and structural cerebral changes in chronic alcoholics, analysing their association with personality features and alcohol drinking habits.

Forty patients with alcohol dependency, including 15 with antisocial personality disorder (ASP) as defined in DSM–III–R and 10 age and sex matched healthy controls were studied after termination of withdrawal symptoms, using high resolution single photon emission tomography (SPECT), cranial computerised tomography (CT) and brainstem auditory evoked potentials (BAEP).

We found significant reductions in regional cerebral blood flow (rCBF) measurements of alcoholic patients. Low flow in frontal regions encountered in 67.5% of the patients was associated with the duration of alcohol consumption, while no such relation existed with the amount of daily intake. Patients with ASP exhibited more marked frontal hypoperfusion. Significant brain atrophy detected by CT was present in 40% of the patients and did not correlate with frontal hypoperfusion.

Patients with ASP are more sensitive to toxic effects of alcohol. Alternatively chronic alcoholism leads to frontal lobe dysfunction recognised as ASP in the clinical setting.

The mechanism of the antidepressant action of electroconvulsive therapy (ECT) remains unknown. Based on previous work with antidepressant drugs and their effects on the noradrenergic system, we undertook this study to further determine the effects of ECT on selected indices of peripheral adrenoceptor function in depressed patients.

Binding parameters (Bmax and Kd) of platelet α2- and leukocyte β2-adrenoceptors, plasma noradrenaline (NA), 3-methoxy-4-hydroxy-phenylglycol (MHPG) and cortisol levels were determined in 18 patients, prior to treatment and 14 days after the last of a series of ECTs, and compared with samples obtained from 18 matched control subjects.

Platelet α2-adrenoceptor sites were significantly elevated in untreated patients compared with controls (P < 0.03), but leukocyte β2-adrenoceptor numbers did not differ. Treatment with ECT led to a significant reduction in platelet α2-adrenoceptor numbers, whereas leukocyte β2-adrenoceptor densities increased. Pre-ECT plasma NA, MHPG, and cortisol levels were elevated in patients, compared with controls, and decreased following ECT, but these differences were not statistically significant. Post-ECT plasma NA and β2-adrenoceptor numbers were significantly, negatively correlated (P < 0.05).

These results suggest that platelet α2-adrenoceptors are supersensitive in depressed patients and treatment with ECT results in down-regulation of these receptors, which may be interpreted as a primary therapeutic, ‘normalising’ effect. The post-ECT changes in leukocyte β2-adrenoceptors are probably only secondary to the lower circulating plasma NA levels.

Maladaptive behaviours are frequently present in individuals with learning disabilities; however, their prevalence varies between studies and the aetiology, classification and most appropriate management of such behaviours remain unclear. The aims of the study were twofold: firstly to determine the prevalence of maladaptive behaviours, and secondly to develop a classification typology.

A population of 2202 adults with learning disabilities was surveyed to determine the prevalence of maladaptive behaviours. These were subjected on the basis of 13 behavioural characteristics to cluster analysis. The clusters were further examined in respect of secondary variables.

Over 60% of individuals demonstrated the presence of at least one maladaptive behaviour and in 40% this problem was of either severe degree or frequent occurrence. Men were more likely to demonstrate physical aggression, property destruction, tantrums and verbal abuse than women. Although some severe maladaptive behaviours diminished with age, many persisted through all age groups. Six behavioural clusters were devised which demonstrated face validity and which differed in respect of demographic and developmental variables, and the prevalence of maladaptive habits and autistic symptoms.

These findings demonstrate a high prevalence of maladaptive behaviours and ‘objectionable habits’ among people with learning disabilities. A classification of behavioural symptoms may be possible on the basis of symptom clusters.

There is evidence that people with a history of sexual abuse may have an increased risk of developing alcohol and drug problems.

A self-completion sexual abuse questionnaire was designed and administered to a sample of attenders at three London alcohol services. Drinking behaviour was assessed using the Severity of Alcohol Dependence Questionnaire and the Alcohol Problems Questionnaire, and additional data were derived from case notes.

Fifty-four per cent of women and 24% of men identified themselves as victims of sexual abuse or assault. For the majority this had started before the age of 16 and involved non-relatives. Subjects with a history of sexual abuse were younger, reached drinking milestones earlier, were more likely to have a family history of alcohol misuse and had more alcohol-related problems than non-abused subjects. Sexual abuse, age and alcohol dependence predicted level of problems in a regression analysis.

The high rates of sexual abuse and its association with indications of increased morbidity suggest it is an important issue for the management of alcohol problems. More use could be made of self-completion questionnaires for the investigation of sexual abuse.

Schizophrenia is associated with expanded CAG/CTG trinucleotide repeats. We wished to determine whether the presence of such expansions correlated with specific subsyndromes or other clinical features of schizophrenia.

Seventy patients from England and Wales and 44 patients from Portugal with a DSM–III–R diagnosis of schizophrenia were rated on the opcrit checklist Patients' maximum CAG/CTG repeat length was measured using repeat expansion detection (RED). Significant differences were sought for repeat lengths in subjects categorised according to dimensional and categorical schizophrenia subsyndromes, affective episodes, individual symptoms, and a range of demographic variables.

Maximum CAG/CTG repeat length did not differ significantly for any of the clinical or demographic variables studied.

There are no subsyndromes or other clinical features of schizophrenia associated with CAG/CTG repeat expansion. Therefore, the identification of the gene(s) that contain expanded CAG/CTG repeats and which are associated with schizophrenia is unlikely to be facilitated at present by using any subsyndromes of schizophrenia as phenotypes.

Clinical and demographic information on patients seen as a result of deliberate self-harm (DSH) was collected in an attempt to identify factors in the index episode of DSH predictive of subsequent suicide.

Specific data were prospectively collected on all DSH patients who lived in Blacktown Municipality, Sydney, Australia, and seen from October 1975 to September 1976. Follow-up at 18 years was by evaluation of coroners' records and identification of ‘probable suicide’.

Two hundred and twenty-three patients harmed themselves on one or more occasions. Follow-up at 18 years showed that 15 of the 223 (6.7%) had completed suicide. The proportion at five and eight years was 4.0% and at 10 years was 4.5%. Identified predictors of suicide were: narcotic overdose; more than one episode of DSH in the year of the study; planned episode; and mental illness. Teenage narcotic-abusing males were at greatest risk, and in females a planned episode was the most powerful predictor.

Suicides continued to occur over 18 years. One of the striking differences between this and other studies is the finding of teenage male DSH, associated with narcotic abuse, as a strong predictor of subsequent suicide. These findings are particularly relevant to the issue of young male suicide, which increased from the 1970s onwards in Australia and elsewhere.

The objective of this project was to test whether there are differences in the size of the caudate nucleus in schizophrenic in-patients with and without tardive dyskinesia.

The study was cross-sectional in design, examining group differences between institutionalised schizophrenic patients with and without tardive dyskinesia, using non-enhanced computerised tomography scans of the brain. The group comprised 15 schizophrenic patients with persistent tardive dyskinesia and 21 in-patient schizophrenic controls who were group-matched for demographic variables.

The dyskinetic subjects had a significantly larger left caudate nucleus and tended to have a larger right caudate nucleus than the controls. There were no differences between the groups on any of the measures of cerebral atrophy.

The findings can be understood within the context of models of neostriatal function. It is possible that a larger caudate nucleus could be used to identify patients at risk of developing tardive dyskinesia.

Recent studies suggest that larger brain size may offer some protection against the clinical manifestations of Alzheimer's disease. However, this association has not been investigated in population-based studies.

The relationship between head circumference, a measure of premorbid brain size, and score on the Cognitive Abilities Screening Instrument (CASI) was studied in a population of 1985 Japanese–Americans aged 65+ living in King County, Washington, USA.

After adjusting for age, sex and education, head circumference was positively associated with CASI score (b=3.8, 95% CI: 2.2, 5.4; P=0.0000), but not with diagnosis of probable AD (odds ratio=0.87, 95% CI: 0.33, 1.87). When the data were stratified by AD status, no association was seen among controls (b=1.6, 95% CI: – 1.7, 5.1; P=0.4), whereas a strong effect was present among cases (b=35.3, 95% CI: 12.2, 58.4: P=0.006).

These results suggest that persons with AD with smaller head circumference either had the disease longer or progressed more rapidly than those with larger head circumference. Improvement in environmental factors in prenatal and early life that partially determine completed brain/head size may have consequences for the late-life expression of Alzheimer's disease in vulnerable individuals.

Little is known about the long-term outcome of schizophrenia that has its onset during childhood and early adolescence (early-onset schizophrenia, or EO-SZ). Whether or not EO-SZ is an aetiologically separate form of schizophrenia (SZ) is unresolved.

The study was a 14.8-year follow-up, using methods such as systematic sampling, evaluation of possible non-respondent bias, consensus best-estimate diagnoses (DSM–III–R) made independently in childhood and adulthood, measures of positive and negative dimensions, of non-psychotic behaviour disturbances (NPBD) and of developmental problems before the appearance of SZ.

There was high stability of EO-SZ (n=40) diagnoses (mean onset at 14.0 years) until adulthood (mean age at follow-up 28.8 years) but a lower stability of positive and negative schizophrenic dimensions. There was a poor outcome of EO-SZ, a strong over-representation of males but few gender differences, and no effect of age of onset on clinical features and outcome.

EO-SZ taken as a whole shows no qualitative differences to adult-onset SZ. However, a distinction through the onset of preschizophrenic developmental problems or NPBD might be a way to investigate heterogeneity within EO-SZ.