Thirty-two patients who complained of episodes of ravenous overeating which they felt unable to control (bulimia) were asked to describe their behaviour and symptoms. There was considerable variation both between and within individuals, but a number of factors were defined which appeared to be common to all with the complaint. It is difficult to set up strict criteria for the recognition of bulimia, and those that have recently been proposed are criticized in the light of our present findings.

Background. This study aimed to quantify the association between low IQ and the later development of psychosis in a population-based cohort study of 18-year-old conscripts.

Methods. Fifty thousand males conscripted into the Swedish army in 1969–1970 were followed by means of the Swedish National Register of Psychiatric Care up to 1983. Tests of verbal and visuospatial abilities, general and mechanical knowledge and several psychosocial variables were recorded at conscription.

Results. One hundred and ninety-five subjects were admitted to hospital with schizophrenia and 192 with a non-schizophrenic psychosis on ICD-8 criteria. The distribution of scores in those later diagnosed as suffering from schizophrenia was shifted in a downward direction, with a linear relationship between low IQ and risk. This remained after adjustment for potential confounders. The risk for non-schizophrenic disorders was also higher in those with lower IQ but the effect was less marked and non-linear. Only poorer performance on the verbal tasks and mechanical knowledge test conferred a significantly increased risk for schizophrenia after taking into account general intellectual ability. Low IQ at conscription was not related to age of onset.

Conclusions. The results confirm the importance of low intellectual ability as a risk factor for schizophrenia and other psychoses. This is unlikely to be due to prodromal decline or known confounders. The association could be directly causal with cognitive impairment leading to false beliefs and perceptions, or could be indirect with any factors causing lower IQ, such as abnormal brain development increasing the risk for schizophrenia.

Background. Traditional instruments that measure self-esteem may not relate directly to the schema construct as outlined in recent cognitive models. The Brief Core Schema Scales (BCSS) aim to provide a theoretically coherent self-report assessment of schemata concerning self and others in psychosis. The scales assess four dimensions of self and other evaluation: negative-self, positive-self, negative-other, positive-other.

Method. We analysed the psychometric properties of the BCSS using a sample of 754 students recruited by email and 252 people with psychosis recruited as part of a trial of cognitive therapy. We report the internal consistency, stability and the factor structure of the scale, and the association of the BCSS with measures of self-esteem and with symptoms of paranoia and grandiosity.

Results. The BCSS have good psychometric properties and have more independence from mood than the Rosenberg Self-Esteem Schedule. People with chronic psychosis reported extreme negative evaluations of both self and others on these scales, but their levels of self-esteem and positive evaluations of self and others were similar to the student sample.

Conclusions. Extreme negative evaluations of self and others appear to be characteristic of the appraisals of people with chronic psychosis, and are associated with symptoms of grandiosity and paranoia in the non-clinical population. The BCSS may provide a more useful measure of schemata about self and others than traditional measures of self-esteem.

The paper reports the profile of impairment across a variety of cognitive functions with special emphasis on tests sensitive to frontal lobe dysfunction, in 24 elderly depressed patients during and on recovery from mood disorder, compared with 15 age- and sex-matched controls. Traditional neuropsychological tests and a recently developed battery of computerized tests (CANTAB) were used. Impairments were found in the depressed group compared to controls and to themselves on recovery across all domains examined. The depressed group showed deficits on visuospatial recognition memory, attentional shifting at the extra-dimensional shift stage and in measures of both processing and motor speed without impaired accuracy in a visual search task. Impairments were also found on a planning task with disproportionately increased numbers of moves needed for more difficult problems and evidence of both slowed motor response and increased processing time once the task was commenced.

Performance on recovery improved across all tasks. Comparisons were made with the performance of patients suffering from dementia of the Alzheimer type (DAT) and Parkinson's disease on similar tests. Response latencies in test performance were found to correlate with the number of episodes of depression suffered and with ventricular size on CT scan, as measured by computerized planimetry. On recovery, residual depression scores correlated with latency of test performance and with ventricular brain ratio. The results, thus, show that depression in the elderly is associated with a significant degree of deficit on tests sensitive to frontostriatal dysfunction. Some of the deficits appear specific to depression and some do not remit following clinical recovery. However, these impairments have to be interpreted in the context of a broad profile of cognitive deficit.

Background. The development of drug dependence requires prior initiation. What is the relationship between the risk factors for initiation and dependence?

Methods. Using smoking as a model addiction, we assessed smoking initiation (SI) and nicotine dependence (ND) by personal interview in 1898 female twins from the population-based Virginia Twin Registry. We developed a twin structural equation model that estimates the correlation between the liability to SI and the liability to ND, given SI.

Results. The liabilities to SI and ND were substantially correlated but not identical. Heritable factors played an important aetiological role in SI and in ND. While the majority of genetic risk factors for ND were shared with SI, a distinct set of familial factors, which were probably partly genetic, solely influenced the risk for ND. SI was associated with low levels of education and religiosity, high levels of neuroticism and extroversion and a history of a wide range of psychiatric disorders. ND was associated with low levels of education, extroversion, mastery, and self-esteem, high levels of neuroticism and dependency and a history of mood and alcohol use disorders.

Conclusions. The aetiological factors that influence SI and ND, while overlapping, are not perfectly correlated. One set of genetic factors plays a significant aetiological role in both SI and ND, while another set of familial factors, probably in part genetic, solely influences ND. Some risk factors for SI and ND impact similarly on both stages, some act at only one stage and others impact differently and even in opposite directions at the two stages. The pathway to substance dependence is complex and involves multiple genetic and environmental risk factors.

Depression and anxiety disorders are the leading cause of sickness absence and long-term work incapacity in most developed countries. The present study aimed to carry out a systematic meta-review examining the effectiveness of workplace mental health interventions, defined as any intervention that a workplace may either initiate or facilitate that aims to prevent, treat or rehabilitate a worker with a diagnosis of depression, anxiety or both. Relevant reviews were identified via a detailed systematic search of academic and grey literature databases. All articles were subjected to a rigorous quality appraisal using the AMSTAR assessment. Of the 5179 articles identified, 140 studies met the inclusion criteria, of which 20 were deemed to be of moderate or high quality. Together, these reviews analysed 481 primary research studies. Moderate evidence was identified for two primary prevention interventions; enhancing employee control and promoting physical activity. Stronger evidence was found for CBT-based stress management although less evidence was found for other secondary prevention interventions, such as counselling. Strong evidence was also found against the routine use of debriefing following trauma. Tertiary interventions with a specific focus on work, such as exposure therapy and CBT-based and problem-focused return-to-work programmes, had a strong evidence base for improving symptomology and a moderate evidence base for improving occupational outcomes. Overall, these findings demonstrate there are empirically supported interventions that workplaces can utilize to aid in the prevention of common mental illness as well as facilitating the recovery of employees diagnosed with depression and/or anxiety.

Background. Despite an extensive literature, there have been widely divergent findings regarding the direction of the association between area socio-economic characteristics and area suicide rates, with high-quality studies finding either a direct relation (higher rates of suicide in higher socio-economic areas), an inverse relation (lower rates of suicide in higher socio-economic areas) or no association.

Method. We performed a systematic review of the literature dating from 1897 to 2004 and identified 86 publications with 221 separate analyses that met our inclusion criteria. We examined the percent of direct, inverse and null findings stratified by key study characteristics including size of aggregated area, socio-economic measure used, region of study, control variables and study design.

Results. Analyses at the community level are significantly more likely to demonstrate lower rates of suicide among higher socio-economic areas than studies using larger areas of aggregation. Measures of area poverty and deprivation are most likely to be inversely associated with suicide rates and median income is least likely to be inversely associated with suicide rates. Analyses using measures of unemployment and education and occupation were equally likely to demonstrate inverse associations. Study results did not vary significantly by gender or by study design.

Conclusions. The heterogeneity of associations is mostly accounted for by study design features that have largely been neglected in this literature. Enhanced attention to size of region and measurement strategies provide a clearer picture of how suicide rates vary by region. Resources for suicide prevention should be targeted to high poverty/deprivation and high unemployment areas.

Background. This study addressed the prevalences, correlates, co-morbidity and disability of DSM-IV generalized anxiety disorder (GAD) and other psychiatric disorders in a large national survey of the general population, the National Institute on Alcohol Abuse and Alcoholism's (NIAAA) National Epidemiologic Survey on Alcohol and Related Conditions (NESARC). The study presents nationally representative data, for the first time, on prevalence, correlates, co-morbidity, and comparative disability of DSM-IV GAD.

Method. Data are taken from a large (n=43093) representative sample of the adult USA population.

Results. Prevalences of 12-month and lifetime GAD were 2·1% and 4·1%. Being female, middle-aged, widowed/separated/divorced, and low income increased risk, while being Asian, Hispanic, or Black decreased risk. GAD was highly co-morbid with substance use, and other anxiety, mood, and personality disorders. Co-morbidity in GAD was not substantially greater than for most other Axis I and II disorders. Disability and impairment in pure GAD were equivalent to pure mood disorders, but significantly greater than in pure substance use, and other anxiety and personality disorders. Individuals co-morbid for GAD and each mood disorder were more disabled than those with pure forms of GAD or each mood disorder. When co-morbid with GAD, nicotine dependence and other anxiety and personality disorders were not associated with increased disability over that associated with pure GAD, but GAD did show increased disability over that due to each of these disorders in pure form.

Conclusions. Associations between GAD and Axis I and II disorders were strong and significant, with variation among specific disorders. Results strongly support GAD as an independent disorder with significant impairment and disability.

Background. Prior research indicates that conduct disorder (CD) is associated with a range of co-morbid mental disorders. However, the actual prevalence, subtypes and patterns of co-morbidity of DSM-IV-defined CD in the general US population remains unknown.

Method. Retrospective assessment of CD and other DSM-IV disorders was conducted using fully structured diagnostic interviews among a nationally representative sample of respondents (n=3199) in the National Comorbidity Survey Replication (NCS-R).

Results. The estimated lifetime prevalence of CD in the US is 9·5% (12·0% among males and 7·1% among females), with a median age-of-onset of 11·6 (0·2) years. Latent class analysis (LCA) identified five CD subtypes characterized by rule violations, deceit/theft, aggression, severe covert behaviors, and pervasive CD symptoms. A dose–response relationship was revealed between CD subtype severity and risk of subsequent disorders. Results also indicated that CD typically precedes mood and substance use disorders, but most often occurs after impulse control and anxiety disorders. Although both active and remitted CD is associated with increased risk of the subsequent first onset of other mental disorders, remitted CD is associated with significantly lower risk of subsequent disorders.

Conclusions. CD is prevalent and heterogeneous in the US population, and more severe subtypes and the presence of active CD are associated with higher risk of co-morbid disorders. Future prospective studies using general population samples will further inform the nature and course of this disorder.

Background. The purpose of this study was to present 12-month and lifetime estimates of the prevalence, sociodemographic and clinical correlates, and psychiatric co-morbidity of DSM-IV cannabis abuse and dependence.

Method. Data were derived from a large nationally representative survey (n=43093) of US adults.

Results. The prevalence of 12-month and lifetime DSM-IV cannabis abuse (1·1% and 7·2%) exceeded the corresponding rates of cannabis dependence (0·3% and 1·3%). Being male, Native American, widowed/separated/divorced, and residing in the West increased the odds whereas being Black, Asian or Hispanic decreased the odds of cannabis abuse and dependence. Cannabis dependence was significantly associated with low income. Ages of onset for both cannabis use disorders occurred in adolescence and the majority of individuals with these disorders remained untreated. Co-morbidity was high between cannabis use disorders and other Axis I and II disorders.

Conclusions. Cannabis use disorders continue to present a widespread and serious personal and public health problem. Native Americans were found to have high rates of cannabis use disorders, warranting closer attention to the mental health needs of this subgroup. Associations between cannabis abuse and dependence and Axis I and II disorders were strong, signaling the need for more comprehensive assessment of individuals with cannabis use disorders. Further controlled treatment studies are needed, especially among co-morbid individuals, in view of growing evidence of the adverse personal, medical and societal impacts of cannabis use disorders in the USA.

The Structured Clinical Interview for DSM-III-R (SCID and SCID II) was administered to 105 eating disorder in-patients in order to examine rates of comorbid psychiatric disorders and the chronological sequence in which these disorders developed. Eighty-six patients, 81·9% of the sample, had Axis I diagnoses in addition to their eating disorder. Depression, anxiety and substance dependence were the most common comorbid diagnoses. Anorexic restrictors were significantly more likely than bulimics (all subtypes) to develop their eating disorder before other Axis I comorbid conditions.

Personality disorders were common among the subjects; 69% met criteria for at least one personality disorder diagnosis. Of the 72 patients with personality disorders, 93% also had Axis I comorbidity. Patients with at least one personality disorder were significantly more likely to have an affective disorder or substance dependence than those with no personality disorder.

Background. Body dysmorphic disorder (BDD) is a highly distressing and impairing disorder characterized by a preoccupation with imagined or slight physical defects in appearance. Well designed studies on its prevalence and on base rates for diagnostic criteria are rare. Therefore this study aimed to reveal prevalence rates of BDD in the general population and to examine clinical features associated with BDD.

Method. Of 4152 selected participants 2552, aged 14–99 years, participated in this German nationwide survey. Participants were carefully selected to ensure that the sample was representative; they were visited by a study assistant who provided instructions and help if needed. Participation rate was 62·3%. DSM-IV criteria for BDD, as well as subthreshold features (e.g. individuals who consider some part(s) of their body as ugly or disfigured, but do not fulfill all BDD criteria) were examined. We also assessed suicidal ideation associated with the belief of having an ugly body part, as well as the desire for cosmetic surgery. Furthermore, somatization symptoms were assessed.

Results. The prevalence of current BDD was 1·7% (CI 1·2–2·1%). Individuals with BDD reported higher rates of suicidal ideation (19% v. 3%) and suicide attempts due to appearance concerns (7% v. 1%) than individuals who did not meet criteria for BDD. Somatization scores were also increased in individuals with BDD, relative to those without. BDD was associated with lower financial income, lower rates of living with a partner, and higher rates of unemployment.

Conclusions. Our study shows that self-reported BDD is relatively common and associated with significant morbidity.

Patients suffering from anorexia nervosa usually exhibit extreme concern over their physical size and weight. This tendency, noted by many observers, has been variously interpreted. However, few experimental studies of this phenomenon have been reported. In the present paper an objective technique for determining bodily perception is described, together with a series of studies involving investigation of bodily perception in anorexia nervosa. In the first study a group of anorexia nervosa patients was compared with a group of normal female controls, results showing that, unlike the controls, the patients exhibited marked overestimation of the width of their own body. Moreover, this tendency was found not to extend to perception of physical objects. The results of a second study showed that the overestimation of body width was less marked when the patients were required to gauge the size of a female model. In addition, they were found to be remarkably accurate in assessing physical height, both of themselves and of the model. A third study explored the effects of refeeding on the patients' tendency to see themselves as abnormally wide (and fat). The findings showed that this tendency decreased as patients put on weight. Moreover, the degree of overestimation of body width exhibited by patients was found to bear a relationship to their progress after discharge from hospital: when this persisted after weight gain, a subsequent relapse was more likely.

Background. In animals, early trauma can produce long-lasting changes in sensitivity to the pathogenic effects of stress. To explore whether similar processes occur in humans, we examine whether childhood sexual abuse (CSA) in women alters sensitivity in adulthood to the depressogenic effects of stressful life events (SLEs).

Method. A history of CSA was obtained from a population-based sample of 1404 female adult twins. Cox Proportional hazard models were used to predict onsets of episodes of DSM-III-R major depression (MD) in the past year from previously assessed levels of neuroticism (N), CSA and past-year SLEs scored on long-term contextual threat.

Results. In the best-fit model, onset of MD was predicted by CSA, SLEs and N. Individuals with CSA (and especially with severe CSA) had both an overall increased risk for MD and a substantially increased sensitivity to the depressogenic effects of SLEs. A ‘dose–response’ relationship between severity of CSA and sensitivity to SLEs was clearer in those with low to average levels of N than in those with high levels of N.

Conclusion. As documented with physiological responses to a standardized laboratory stressor, CSA increases stress sensitivity in women in a more naturalistic setting. Both genetic and early environmental risk factors can produce long-term increase in the sensitivity of individuals to depressogenic life experiences.

Background. Cognitive therapy for psychotic symptoms often embraces self-evaluative beliefs (e.g. self-worth) but whether and how such beliefs are related to delusions remains uncertain. In previous research we demonstrated that distress arising from voices was linked to beliefs about voices and not voice content alone. In this study we examine whether the relationship with the voice is a paradigm of social relationships in general, using a new framework of social cognition, ‘ranking’ theory.

Method. In a sample of 59 voice hearers, measures of power and social rank difference between voice and voice hearer are taken in addition to parallel measures of power and rank in wider social relationships.

Results. As predicted, subordination to voices was closely linked to subordination and marginalization in other social relationships. This was not the result of a mood-linked appraisal. Distress arising from voices was linked not to voice characteristics but social and interpersonal cognition.

Conclusion. This study suggests that the power imbalance between the individual and his persecutor(s) may have origins in an appraisal by the individual of his social rank and sense of group identification and belonging. The results also raise the possibility that the appraisal of voice frequency and volume are the result of the appraisal of voices' rank and power. Theoretical and novel treatment implications are discussed.