Psychiatry in severe and profound learning disability is essentially behavioural psychiatry. Some clinical and research observations of disorders of behaviour in this group are summarised in this study.

After inspection of the literature, I postulated a clinical syndrome of violence and self-injury in the severely learning disabled. A check-list of behavioural symptoms was developed and used in a community survey.

Behaviour, assessed by the check-list, supported the existence of organic behaviour disorder, as did small-scale psychophysiological testing.

Self-injury is strongly associated with violence, and with severe and profound learning disability. Pathophysiology of violence and self-injury may include high levels of psychophysiological arousal demonstrated by unstable EEGs. Reduction of arousal by antipsychotic medication is associated with clinical improvement in violent and self-injurious behaviours.

The purpose of this paper is to review the association between genetic syndromes and self-injurious behaviour.

The information available from the literature on the subject of self-injurious behaviours and genetic syndromes was collated and presented with a critical appraisal.

Self-injurious behaviours are associated with some genetic syndromes. However, the causal relationship between the genetic syndromes and the self-injurious behaviour remains far from clear.

Although self-injurious behaviour has been shown to be the part of a broader phenotype in many genetic disorders, the specificity and sensitivity of these behaviours in this context remain unclear.

Many sedative and antipsychotic agents have been used in the management of severe self-injury associated with learning disabilities. Their efficacy has been questioned. Recent research has identified some biological abnormalities associated with severe self-injury and allowed a more rational selection of treatment.

Review of published literature, including trials, previous reviews and case reports.

There is evidence for the efficacy of opiate antagonists in the management of severe self-injury, and recent research has identified potential methods of predicting treatment response. Dopamine D1 antagonists and some agents affecting serotonin turnover may also be of benefit.

More rational psychopharmacological treatments for severe self-injurious behaviour may become available. Such treatments are difficult to evaluate for methodological and ethical reasons. They usually involve the clinical use of compounds for unlicensed indications, rather than trials of agents developed specifically to treat severe self-injurious behaviour. Combining psychopharmacological and psychological interventions may provide additional benefits.

Psychological approaches to working with people with learning disabilities who self-injure have developed over the past 30 years.

The major literature is reviewed and an ecological framework is described which emphasises the importance of environmental, interpersonal and intrapersonal dynamics to understanding the multi-factorial nature of self-injury. Case examples are given.

Self-injury is seen as essentially communicative and functionally adaptive; it is the person's best attempt to deal with abusive, neglecting or traumatic environments or events.

The persistence of self-injurious behaviour once established, requires an interdisciplinary approach which addresses comprehensively the variety of factors which have contributed to the development and maintenance of self-injury.

This is the third large-scale audit in the past 20 years and compares the practice of electroconvulsive therapy (ECT) in England and Wales with the standards derived from the Royal College of Psychiatrists' 2nd ECT handbook.

Facilities, equipment, practice, personnel and training were systematically evaluated during visits to all ECT clinics in the former North East Thames and East Anglia regions and Wales. All other English ECT clinics were surveyed with a postal questionnaire. Information was obtained for 184 (84%) of the 220 ECT clinics identified.

Although some aspects of ECT administration had improved since the last audit in 1991, overall only one-third of clinics were rated as meeting College standards. Only 16% of responsible consultants attended their ECT clinic weekly and only 6% had sessional time for ECT duties. Fifty-nine per cent of all clinics had machines of the type recommended by the College and 7% were still using machines considered outdated in 1989. Only about one-third of clinics had clear policies to help guide junior doctors to administer ECT effectively.

Twenty years of activity by the Royal College of Psychiatrists and three large-scale audits have been associated with only modest improvement in local practice.

Two studies tested whether subjects with obsessive-compulsive disorder could successfully use BT STEPS, a computer-aided system, to perform self-assessment for self-treatment of obsessive-compulsive disorder by exposure and ritual prevention.

Subjects were given a self-guiding manual and could use a touch-tone telephone to access computer-controlled Interactive Voice Response interviews at their convenience from home. Using the BT STEPS system, patients rated themselves and worked out a plan for individually tailored self-exposure therapy.

Outcomes were similar in the two studies. Of the 63 subjects who used BT STEPS, 84% completed the self-assessment module. Most calls were made outside usual office hours. As expected, subjects did not improve merely by completing self-assessment. However, completion of self-assessment predicted later improvement with self-exposure therapy.

Most subjects successfully completed self-assessment using BT STEPS from their homes.

A randomised controlled trial was conducted in an acute treatment setting to examine the effectiveness of compliance therapy, a brief pragmatic intervention targeting treatment adherence in psychotic disorders, based on motivational interviewing and recent cognitive approaches to psychosis.

Seventy-four patients with psychotic disorders according to DSM–III–R criteria recruited from consecutive admissions to an acute in-patient unit, received 4–6 sessions of either compliance therapy or non-specific counselling, and were followed-up over 18 months. The principal outcome measures were observer-rated compliance, attitudes to treatment, insight and social functioning.

Significant advantages were found for the compliance therapy group post-treatment on measures of insight, attitudes to treatment and observer-rated compliance which were retained over the follow-up period. Global social functioning improved relatively more over time in the compliance therapy group compared with the control group. Survival in the community prior to readmission was significantly longer in the compliance therapy group.

The results support the effectiveness of compliance therapy in improving functioning and community tenure after an acute psychotic episode.

Non-compliance rates with antipsychotic medication can be high, and the personal and societal costs are considerable. A new psychological intervention, compliance therapy seeks to improve compliance and patient outcomes and reduce treatment costs.

A randomised controlled study examined the cost-effectiveness of compliance therapy compared to nonspecific counselling over 18 months for 74 people with psychosis admitted as inpatients at the Maudsley Hospital. Bivariate and multivariate analyses were conducted to test for differences and to explore inter-patient cost variations.

Compliance therapy is more effective and is no more expensive. Consequently, compliance therapy is more cost-effective than non-specific counselling at six, 12 and 18 months.

There are compliance, outcome and cost-effectiveness arguments in favour of compliance therapy in preference to non-specific counselling.

Emotional memory is a special category of memory for events arousing strong emotions. To investigate the effects of emotional involvement on memory retention in individuals with Alzheimer's disease we studied peoples' memories of distressing experiences during a devastating earthquake.

Fifty-one subjects with probable Alzheimer's disease who experienced the Kobe earthquake at home in the greater Kobe area were studied. Memories of the earthquake were assessed 6 and 10 weeks after the disaster in semi-structured interviews, and were compared with memories of a magnetic resonance imaging (MRI) examination given after the earthquake.

Forty-four (86.3%) of the subjects remembered the earthquake and 16 (31.4%) of subjects remembered the MRI experience. Factual content of the earthquake was lost in most of the subjects.

Fear reinforces memory retention of an episode in subjects with Alzheimer's disease but does not enhance retention of its context, despite repeated exposure to the information.

Elderly people with paranoid symptoms are a taxing group for medical and social services, but studies of the prevalence of these symptoms in the general elderly population are rare. This study aimed to estimate the community prevalence and to identify some associated variables.

A community sample of 1420 elderly people, was extensively examined by nurses and physicians.

Paranoid ideation was found in 6.3% of the sample. The prevalence in people with cognitive dysfunction (n=381, 12.1%) was higher than in those without (n=1039, 2.6%). Once cognitive impairment had been controlled the associated variables were: being divorced, being female, having depressive symptoms, using psychotropic drugs, having no friends or visitors, using community care and being an immigrant.

Paranoid symptoms in this elderly population were associated most strongly with cognitive impairment. Other associated variables pointed to a higher level of social isolation than others in the community.

We evaluated the utility of the Chinese version of the Edinburgh Postnatal Depression Scale (EPDS) and measured the prevalence of major depression six weeks after confinement among Chinese women in Hong Kong.

A prospective cohort of 145 women completed the EPDS, the 12-item General Health Questionnaire (GHQ) and the Beck Depression Inventory (BDI) six weeks after giving birth. They were then assessed with the Structured Clinical Interview for DSM–III–R, non-patient version (SCID–NP) to establish psychiatric diagnosis. The criterion validity of EPDS was tested against this clinical diagnosis, and the concurrent validity against the GHQ and BDI scores was also evaluated. The internal consistency of the scales was measured by Cronbach's α coefficient.

The Chinese EPDS had satisfactory psychometric properties and a cut-off score of 9/10 is recommended for screening depressive illness in a general postnatal population. At six weeks postpartum, 5.5% of the study population suffered from major depression.

The Chinese EPDS will be useful for screening for postnatal depression.

Several studies have explored serotonin (5-HT) transmission in people with anorexia nervosa, but their results have been inconsistent.

According to a double-blind placebo-controlled design, plasma prolactin response to the specific serotonergic probe d-fenfluramine was investigated in 10 underweight and two normal-weight women with anorexia, and in 12 age-matched healthy females. Eating-related psychopathology, depressive and obsessive-compulsive symptoms, and aggressiveness were measured by appropriate rating scales.

Compared with healthy control subjects, the women with anorexia showed reduced baseline prolactin and oestrogen levels and increased basal Cortisol concentrations. The prolactin response to d-fenfluramine was blunted and did not correlate with psychopathological measures.

These results support a dysfunction of 5-HT transmission in anorexia nervosa. This dysfunction does not seem to be related to concomitant depressive or obsessive-compulsive symptoms or to the level of aggressiveness of the patients.

Post-traumatic stress disorder (PTSD) can be a persistent and disabling psychiatric disorder. There is little systematic research into the psychiatric consequences of road traffic accidents (RTAs) in children and adolescents.

A consecutive sample of 8–16-year-olds attending an accident and emergency department following RTAs were screened for PTSD. Potential cases and their parent(s) were interviewed with semi-structured research instruments about six weeks and six months after the accident.

Fifty-three (45%) of the 119 subjects fell above PTSD cut-off on the Frederick's Reaction Index. Thirty-three (75%) of the 44 cases met DSM–IV criteria for PTSD. In half of these other psychiatric disorders were present, including major depressive disorder and anxiety disorders. Being female, involvement in car accidents and pre-existing depression and anxiety were associated with developing PTSD. Seventeen per cent of the sample continued to be symptomatic six months after the accident.

PTSD is a common consequence of RTAs. Liaison with accident and emergency departments would enhance the early detection and follow-up of children at risk of developing PTSD.