Background. Some doctors who initially choose psychiatry do not pursue it as a long-term career. The study seeks to identify reasons for leaving psychiatry.

Method. Postal questionnaire survey of UK medical graduates of 1988, 1993, 1996 and 1999 identified as having left psychiatry; for comparison, doctors who left general practice or trauma and orthopaedics.

Results. Response rate was 74% (572/778); 488 respondents satisfied study criteria (59 psychiatry, 318 general practice, 111 trauma and orthopaedics). The speciality's poor public image, perceived lack of respect from medical peers, perceived threat of violence from patients, under-resourcing and low morale were problems for psychiatry leavers. Job stress, self-assessed unsuitability, and concerns about the lack of evidence-based treatments also influenced decisions to leave psychiatry.

Conclusions. Early exposure to psychiatry may help trainees assess their suitability. Negative perceptions of workforce issues (e.g. low morale) and of clinical issues (e.g. perceived lack of ability to improve prognosis) need addressing to increase retention.

Background. There is a high incidence of panic and somatoform disorders after vestibular neuritis. However, the occurrence of psychiatric disorders has been shown to be unrelated to persistent functional vestibular impairment. The aim of the present study was to examine the role played by cognitions in the development of panic and somatoform disorders.

Method. In a 2-year prospective study, cognitions were recorded at various stages using the Agoraphobic Cognitions Questionnaire and Body Sensations Questionnaire. Our analysis focused on whether body-related anxiety or panic-related thoughts can predict the development of a panic or somatoform disorder.

Results. Fear arising on the first day of an acute vestibular episode did not predict the development of panic or somatoform disorders. One week after the dysfunction, however, the fear of vertigo was a significant predictor, explaining 20% of the variance in the development of either disorder. After 6 weeks, persistent fear of vertigo or vomiting predicted approximately 30% of the variance, and after 6 months panic-related thoughts predicted 40% of the variance and, with the inclusion of body-related fears, as much as 60% of the variance in the development of panic or somatoform disorders.

Conclusion. Our data confirm prospectively a number of fundamental assumptions of cognitive theory concerning the development of anxiety disorders. Subjects who experience vertigo as particularly alarming focus more intensely than other patients on the negative symptoms they perceive as being related to the disorder. Patients with panic-related cognitions were prone to develop panic or somatoform disorders.

Objective. Eating disorders are more common in females than in males. Gender differences may be due to organizational (i.e. prenatal) and activational (i.e. post-natal) gonadal hormone effects that influence sex differences in behavior. This preliminary set of studies examined these effects by investigating relationships between eating disorder symptoms, prenatal testosterone exposure, and adult levels of estrogen in women.

Method. We examined organizational associations by investigating relationships between disordered eating and finger-length ratios, which are known to be somatic markers of prenatal testosterone exposure. Participants included 113 adult female twins drawn from the community. Disordered eating was assessed with the total score from the Minnesota Eating Behavior Survey (MEBS). Finger lengths were hand scored using a ruler and photocopies of both hands. We also investigated activational influences by examining associations between circulating levels of estradiol and disordered eating symptoms. Two independent samples of adult females (n's=24 and 25) drawn from the community were used for this study. Disordered eating was again assessed with the MEBS total score, while saliva samples were used for assessing estradiol.

Results. Positive associations were found between disordered eating and both finger-length ratios and circulating estradiol levels.

Conclusions. Findings suggest that lower levels of prenatal testosterone exposure and higher adult levels of estradiol are associated with increased eating disorder symptoms. We hypothesize that the relatively low level of testosterone before birth in females permits their brains to respond to estrogens at puberty, when the hormones activate the genes contributing to disordered eating in vulnerable girls.

Background. The global toll of suicide is estimated to be one million lives per year, which exceeded the number of deaths by homicide and war combined. A key step to suicide prevention is to prevent less serious suicidal behaviour to preclude more lethal outcomes. Although 61% of the world's suicides take place in Asia and the suicide rates among middle age groups have been increasing since the economic crisis in many Asian countries, population-based studies of suicidal behaviour among working-age adults in non-western communities are scarce.

Method. Data from a population-based survey with 2015 participants were used to estimate the prevalence of suicidal ideation and behaviour among the working-age population in Hong Kong, and to study the associated socio-economic and psychological correlates. We focused particularly on potential modulating factors between life-event-related factors and suicidal ideation.

Results. Six per cent of the Hong Kong population aged 20–59 years considered suicide in the past year, while 1·4% attempted suicide. Hopelessness, reasons for living, and reluctance to seek help from family and friends had direct association with past-year suicidal ideation. Reasons for living were found to moderate the effect of perceived stress on suicidal ideation.

Conclusions. Suicidality is a multi-faceted problem that calls for a multi-sectored, multi-layered approach to prevention. Prevention programmes can work on modulating factors such as reasons for living to reduce suicidal risk in working-age adults.

Background. Effective community-based care for older people requires the integration of assessment approaches by old age psychiatry, geriatric medicine and social services care management. This study examines the value of such collaboration in the assessment of older people at risk of entering care homes.

Method. A randomized controlled trial of integrated assessment (care management with additional clinical assessment by old age psychiatrist or geriatrician) versus care management assessment only, for older people at risk of care-home admission. Older people's characteristics, physical and cognitive functioning, depression, behaviour and quality of care measures were recorded at assessment and 6 months. Differences in clinicians' and care managers' recommendations and placement rates between experimental and control groups were analysed. Logistic regression, examining associations between different needs-related characteristics and the likelihood of care-home admission, was used to characterize placement decisions.

Results. Clinicians recommended fewer older people for placement than did care managers in their usual assessments (a reduction of 85%). This did not, however, translate into a reduction in actual numbers of placements, although placements were delayed in those receiving the integrated assessment. Placements of those receiving the integrated assessment were influenced more by clinical factors, most notably cognitive impairment, which clinicians detected more readily than care managers.

Conclusions. In this at-risk group, integrated assessment offers a means of potentially both delaying care-home admissions and taking account of a wider range of factors in placement decisions. Such an approach is viewed as an important policy goal both in the UK and internationally.

Background. Although a correlation between the central dopaminergic system and intelligence may exist, the results from imaging studies remain inconclusive. The aim of this study was to explore the relationship between striatal dopamine D2/D3 receptor availability and verbal intelligence quotient (VIQ) using single photon emission computed tomography (SPECT).

Method. Striatal D2/D3 receptor availability of 64 healthy subjects was determined with the [123I]iodobenzamide ([123I]IBZM) ligand. Intelligence quotients (IQs) of the subjects were measured by the Wechsler Adult Intelligence Scale – Revised (WAIS-R).

Results. In addition to age, left striatal D2/D3 receptor availability correlated positively with VIQ. In females, left striatal D2/D3 receptor availability was the only variable that correlated significantly with the similarities subtest of VIQ.

Conclusions. There is a relationship between left striatal D2/D3 receptor availability and verbal intelligence, which varies, predominantly in males.

Background. Comprehensive studies of mortality among patients with chronic fatigue (CF) and chronic fatigue syndrome (CFS) have not been published, but several sources suggest that CFS is associated with an elevated risk for suicide.

Method. Data on 1201 chronically fatigued patients followed in a university-affiliated tertiary-care clinic for up to 14 years were submitted to the Center for Disease Control and Prevention (CDC) National Death Index (NDI) to evaluate all-cause and suicide-caused death rates against standardized mortality rates (SMRs). We used Life Table Analysis to examine the influence of sex and diagnoses of CFS and depression.

Results. All-cause mortality in chronically fatigued patients was no higher than expected, but suicide-caused death rates were more than eight times higher than in the US general population. The significant elevation in the SMR of suicide was restricted to those who did not meet criteria for CFS [SMRCF=14·2, 95% confidence interval (CI) 5·7–29·3 versus SMRCFS=3·6, 95% CI 0·4–12·9]. Among chronically fatigued patients who did not meet CFS criteria, those with a lifetime history of major depression (MD) had higher suicide-caused death rates than among their non-depressed counterparts (SMRMD=19·1, 95% CI 7·0–41·5 versus SMRNMD=5·6, 95% CI 0·1–31·4), although the difference was not significant.

Conclusions. CFS does not appear to be associated with increased all-cause mortality or suicide rates. Clinicians, however, should carefully evaluate patients with CF for depression and suicidality.

Background. Developmental adversities may be risk factors for adult suicidal behavior, but this relationship has rarely been studied prospectively. The present study examined the association between childhood adversities and new onset suicidal ideation and attempts in an adult population-based sample.

Method. The study used a large community mental health survey (the Netherlands Mental Health Survey and Incidence Study; n=7076, age range 18–64 years). Logistic regression analyses were used to evaluate the relationship between childhood adversities and new onset of suicidal ideation and attempts over 3 years of longitudinal follow-up.

Results. During the study period 85 new cases of suicidal ideation and 39 new onset suicide attempts were observed. The incidence rate for new suicide ideation was 0·67% per year and the incidence rate for new suicide attempts was 0·28% per year. Childhood neglect, psychological abuse and physical abuse were strongly associated with new onset suicidal ideation and suicide attempts. Odds ratios (ORs) ranged from 2·80 to 4·66 for new onset suicidal ideation and from 3·60 to 5·43 for new onset suicide attempts. The total number of adversities reported had a strong graded relationship to new onset suicidal ideation and attempts. These associations remained significant after controlling for the effects of mental disorders.

Conclusions. Childhood abuse and multiple adversities are strongly associated with future suicidal behavior and the mental disorders assessed in the present study do not fully account for this effect. A comprehensive understanding of suicidal behavior must take childhood adversities into account.

Background. Spousal concordance for common mental disorders provides evidence of the relevance of social contextual factors. There are, however, limitations within the existing literature examining spousal similarity in mental health and little consensus as to the causes of spousal similarity. This study considers a large representative sample and examines an extensive range of risk factors using multilevel statistical methods to explore spousal similarity in common mental disorders.

Method. Data were from the Household, Income and Labour Dynamics in Australia (HILDA) Survey, a large nationally representative survey of Australian households. Analysis focused on 3808 mixed-sex couples in marriage and de facto relationships. Multilevel models assessed the mental health scale of the SF-36. Analyses considered risk factors at the individual, couple and area levels, examined the effects of relationship duration on concordance, and considered longitudinal data to assess the consistency with cross-sectional analysis.

Results. Significant spousal concordance on the mental health scale was demonstrated (r=0·25) and was independent of, and unexplained by mental health risk factors, including experience of multiple shared life events. Spousal similarity for mental health increased across the first 5 years of relationships.

Conclusions. Evidence of spousal concordance for common mental disorders highlights the importance of the social context of marriage in the aetiology of mental illness and identifies an important direction for further research.

Background. Children of alcoholics are significantly more likely to experience high-risk environmental exposures, including prenatal substance exposure, and are more likely to exhibit externalizing problems [e.g. attention deficit hyperactivity disorder (ADHD)]. While there is evidence that genetic influences and prenatal nicotine and/or alcohol exposure play separate roles in determining risk of ADHD, little has been done on determining the joint roles that genetic risk associated with maternal alcohol use disorder (AUD) and prenatal risk factors play in determining risk of ADHD.

Method. Using a children-of-twins design, diagnostic telephone interview data from high-risk families (female monozygotic and dizygotic twins concordant or discordant for AUD as parents) and control families targeted from a large Australian twin cohort were analyzed using logistic regression models.

Results. Offspring of twins with a history of AUD, as well as offspring of non-AUD monozygotic twins whose co-twin had AUD, were significantly more likely to exhibit ADHD than offspring of controls. This pattern is consistent with a genetic explanation for the association between maternal AUD and increased offspring risk of ADHD. Adjustment for prenatal smoking, which remained significantly predictive, did not remove the significant genetic association between maternal AUD and offspring ADHD.

Conclusions. While maternal smoking during pregnancy probably contributes to the association between maternal AUD and offspring ADHD risk, the evidence for a significant genetic correlation suggests: (i) pleiotropic genetic effects, with some genes that influence risk of AUD also influencing vulnerability to ADHD; or (ii) ADHD is a direct risk-factor for AUD.

Background. Patients with borderline personality disorder (BPD) frequently report unresolved life events but it is still poorly understood, how these experiences are represented in the brain. Using functional magnetic resonance imaging (fMRI), the present study aimed at investigating the neural correlates of the recall of unresolved life events in patients with BPD and healthy controls.

Method. Twenty female BPD patients and 21 healthy control subjects underwent fMRI. During measurement subjects recalled unresolved and resolved negative life events. Individual cue words were used to stimulate autobiographical memory. After scanning, subjects rated their emotional states during the recall of both types of memories.

Results. When contrasting unresolved and resolved life events, patients showed significant bilateral activation of frontotemporal areas including the insula, amygdala, and the anterior cingulate cortex, the left posterior cingulate cortex, right occipital cortex, the bilateral cerebellum and the midbrain. In healthy subjects, no differential brain activation was related to these conditions. The 2×2 factorial analysis (ΔBPD−Δcontrols) revealed similar results with bilateral activation of the frontal cortex including parts of the insula and of the orbitofrontal cortex, temporal activation including the amygdala, activation of the right occipital cortex, and parts of the cerebellum. Patients but not controls reported higher levels of anxiety and helplessness during the unresolved versus resolved memory condition.

Conclusions. The activation of both, the amygdala and prefrontal areas, might reflect an increased effortful but insufficient attempt to control intensive emotions during the recall of unresolved life events in patients with BPD.

Background. The urban environment may increase the risk for psychotic disorder in interaction with pre-existing risk for psychosis, but direct confirmation has been lacking. The hypothesis was examined that the outcome of subclinical expression of psychosis during adolescence, as an indicator of psychosis-proneness, would be worse for those growing up in an urban environment, in terms of having a greater probability of psychosis persistence over a 3·5-year period.

Method. A cohort of 918 adolescents from the Early Developmental Stages of Psychopathology Study (EDSP), aged 14–17 years (mean 15·1 years), growing up in contrasting urban and non-urban environments, completed a self-report measure of psychotic symptoms at baseline (Baseline Psychosis) and at first follow-up around 1 year post-baseline (T1). They were again interviewed by trained psychologists for the presence of psychotic symptoms at the second follow-up on average 3·5 years post-baseline (T2).

Results. The rate of T2 psychotic symptoms was 14·2% in those exposed to neither Baseline Psychosis nor Urbanicity, 12·1% in those exposed to Urbanicity alone, 14·9% in those exposed to Baseline Psychosis alone and 29·0% in those exposed to both Baseline Psychosis and Urbanicity. The odds ratio (OR) for the combined exposure was 2·46 [95% confidence interval (CI) 1·46–4·14], significantly greater than that expected if Urbanicity and Baseline Psychosis acted independently.

Conclusion. These findings support the suggestion that the outcome of the developmental expression of psychosis is worse in urban environments. The environment may impact on risk for psychotic disorder by causing an abnormal persistence of a developmentally common expression of psychotic experiences.

Background. Early alcohol use is associated with abuse and dependence of licit and illicit substances later in life. The role of genetic and environmental factors in this association is not conclusive.

Method. In 1992, data on substance use, abuse/dependence and psychiatric disorders were collected from 8169 male twin members of the Vietnam Era Twin Registry. The interview obtained age of onset of regular drinking (one drink/month for 6 or more months). Regression analyses of twin pairs discordant for early alcohol use tested whether the association between early drinking (before age 17) and adult substance use and abuse/dependence remained after controlling for genetic factors, family environment and covariates. Twin models tested for common genetic and/or environmental influences on early drinking and adult alcohol dependence and ever use and abuse/dependence on marijuana and other drugs.

Results. Co-twin analyses suggested the association between early regular alcohol use and adult alcohol dependence, marijuana and other drug use, and marijuana and other drug abuse/dependence could not be entirely explained by common genetic and shared family environmental factors. Genetic contributions to early regular drinking were significantly correlated with those on use of marijuana (rA=0·59), use of other drugs (rA=0·64), alcohol dependence (rA=0·54) and abuse/dependence of marijuana and other drugs (rA=0·63 and 0·66). Small but significant unique environmental correlations (rE range 0·11–0·22) indicated that familial factors could not entirely explain the association between early alcohol use and later substance use, abuse and dependence.

Conclusions. Early regular drinking is associated with later alcohol dependence and use, abuse/dependence on drugs. The association is not entirely explained by genetic or shared family environmental factors. This suggests unique environmental factors contribute to transitions from early regular alcohol drinking to use, abuse and dependence on alcohol and other substances.

Background. Psychiatric epidemiological surveys in China have repeatedly found much lower prevalence estimates than in most other parts of the world.

Method. Face-to-face household interviews of 5201 subjects (2633 in Beijing and 2568 in Shanghai respectively) were conducted from November 2001 to February 2002 using a multistage household probability sampling method. A Chinese version of the World Health Organization Composite International Diagnostic Interview (CIDI) was used for assessment.

Results. Twelve-month prevalence of any DSM-IV mental disorder in metropolitan China is estimated to be 7·0%, with major depressive disorder (2·0%), specific phobia (1·9%), and intermittent explosive disorder (1·7%) the most common disorders. Of these, 13·9% are classified as serious, 32·6% moderate, and 53·5% mild. Only 3·4% of respondents with any disorder sought treatment within the previous 12 months.

Conclusions. Although the general pattern of disorders, risk factors, and unmet need for treatment are similar to those in other countries, a low prevalence of mental disorders is found in metropolitan China. Resolving methodological problems that cause downward bias in estimates, such as stigma-related under-reporting and diagnostic incongruity with a somatopsychic mode of symptom presentation may lead to more accurate and probably higher prevalence estimates in future epidemiological studies. As a low prevalence still translates into an enormous number of people in China, measures are urgently needed to address the huge unmet need for treatment of mental disorders.

Background. For candidate endophenotypes to be useful for psychiatric genetic research, they first of all need to show significant genetic influences. To address the relative lack of previous data, we set to investigate the extent of genetic and environmental influences on performance in a set of theoretically driven cognitive-experimental tasks in a large twin sample. We further aimed to illustrate how test–retest reliability of the measures affects the estimates.

Method. Four-hundred 7- to 9-year-old twin pairs were assessed individually on tasks measuring reaction time, inhibition, working memory and ‘delay aversion’ performance. Test–retest reliability data on some of the key measures were available from a previous study.

Results. Several key measures of reaction time, inhibition and working-memory performance indicated a moderate degree of genetic influence. Combining data across theoretically related tasks increased the heritability estimates, as illustrated by the heritability estimates of 60% for mean reaction time and 50% for reaction-time variability. Psychometric properties (reliability or ceiling effects) had a substantial influence on the estimates for some measures.

Conclusions. The data support the usefulness of several of the variables for endophenotype studies that aim to link genes to cognitive and motivational processes. Importantly, the data also illustrate specific conditions under which the true extent of genetic influences may be underestimated and hence the usefulness for genetic mapping studies compromised, and suggest ways to address this.

Background. In order to characterize frontal lobe features and their behavioural and cognitive correlates across diagnostic categories, we performed a cross-sectional analysis of behavioural and neuropsychological data from a large, prospective Belgian study on behavioural and psychological signs and symptoms of dementia (BPSD).

Method. Patients with probable Alzheimer's disease (AD) (n=170), frontotemporal dementia (FTD) (n=28), mixed dementia (MXD) (n=29) and dementia with Lewy bodies (DLB) (n=21) were included and underwent neuropsychological and behavioural assessment by means of a battery of tests and scales. Frontal lobe symptoms were quantified by means of the Middelheim Frontality Score (MFS).

Results. In AD (and to a lesser extent in MXD), MFS total scores were negatively correlated with scores on MMSE (Spearman: r=−0·36, p<0·001) and a Verbal Fluency Task (r=−0·38, p<0·001) and were associated with increased severity and frequency of psychosis (r=0·24, p<0·01), activity disturbances (r=0·44, p<0·001) and aggressiveness (r=0·43, p<0·001). In DLB, MFS total scores were negatively correlated with MMSE scores (r=−0·50, p=0·020). No associations were found in FTD patients.

Conclusions. A cross-sectional analysis of frontal lobe features, behavioural characteristics and neuropsychological data demonstrated that, in AD (and to a lesser extent in MXD) patients, frontal lobe symptoms were associated with more pronounced cognitive deficits (of frontal origin), with increased severity and frequency of agitated and aggressive behaviour, and with increased severity of psychosis and depressive symptoms. Given the small sample sizes of the DLB and FTD patient groups, negative findings in these patient groups should be interpreted cautiously.

Background. Jacobs and Bovasso reported (Psychological Medicine 2000, 30, 669–678) that maternal death in childhood and chronic severe depression in adulthood were associated with subsequent breast cancer. We have examined the effects of parental loss in childhood and psychiatric disorder in adult life on breast cancer risk using a national birth cohort study.

Method. Eighty-three cases of breast cancer were diagnosed in a study of 2253 women followed from birth to age 59 years. Cox proportional hazards models were used to test whether breast cancer rates were higher in women who experienced parental death and divorce before age 16, psychiatric disorders between 15 and 32 years, symptoms of anxiety and depression at 36 years, or use of antidepressant medication at 31 or 36 years than in women who did not have these experiences.

Results. There was no overall association between parental death, parental divorce or psychiatric disorder and the incidence of breast cancer. There was some evidence that women with more severe psychiatric disorders between the ages of 15 and 32 years were more likely to develop breast cancer early. The interaction between parental divorce and severe psychiatric disorder was non-significant (p=0·1); however, the group who experienced both these events had an increased breast cancer risk compared with those who experienced neither [hazard ratio (HR) 2·64, 95% confidence interval (CI) 1·13–6·19].

Conclusions. Our study does not provide strong support for the hypothesis that early loss or adult psychiatric disorders are associated with breast cancer. A meta-analysis is needed that uses data from all available cohort studies and investigates possible interactive effects on breast cancer risk.

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. Pregnancy has often been associated with cognitive deficits, but results are equivocal and little is known about how these deficits progress with time.

Method. In the present study, the cognitive performance of 57 pregnant women was compared with that of 50 non-pregnant women matched for age and education, using a well-validated neurocognitive test battery at weeks 14, 17, 29, and 36 of pregnancy, and 32 weeks postpartum in the pregnant group and at comparable times in the non-pregnant group.

Results. Memory encoding and retrieval, as assessed with a word learning task, were significantly lower in the pregnant group than in the control group. This difference was still present at 32 weeks after delivery. The two groups did not differ in complex speed of information processing at any of the test moments; however, general speed of information processing was significantly compromised during early motherhood (week 32 postpartum).

Conclusion. Thus, memory performance is poorer during pregnancy and early motherhood, and general speed of information processing is slower during early motherhood.

Background. Recent epidemiological data suggest a decline in bulimia nervosa (BN) incidence in primary care. We sought to examine BN point prevalence from 1982 to 2002 in a college population.

Method. In 1982, 1992, and 2002, 800 women and 400 men were randomly sampled from a university for a study of health and eating patterns. Participation rates were 72% in women and 63% in men, resulting in n=2491 participants.

Results. BN point prevalence decreased significantly in women over the period of observation. Eating Disorder Inventory Bulimia scores decreased across cohorts, and these decreases remained significant when male and female and Caucasian and non-Caucasian students were analyzed separately.

Conclusion. These data support a decline in BN rates that cannot be attributed to changes in service utilization. Changing socio-cultural factors may explain a true decrease in BN incidence and prevalence.