We read with considerable interest the recent paper by Kéri et al. (31, 915–922) in which they observed that siblings of schizophrenia patients could be distinguished from siblings of bipolar patients on the basis of dysfunctions of visual backward masking and spatial working memory.

Background. The aims of the study were: first to examine, using clinical symptoms of patients as a template, whether the correlated but independent dimensions of positive, negative and depressive symptoms that have been identified in clinical psychosis, also have a distribution as non-clinical experiences in the general population; and second, to establish to what degree population variation in experience of positive and negative features of psychosis is actually independent of experience of depression.

Method. In a representative population sample of 932 young men, we measured experiences of positive, negative and depressive features of psychosis, using a 40-item self-report instrument. Confirmatory factor analysis was used to compare the fit of hypothesized one-, two- and three-factor solutions.

Results. A three-factor model of separate depressive, positive and negative dimensions provided a better fit to the data than either a two-factor or unidimensional model. All three dimensions were correlated with each other, but also showed good discriminant validity in relation to established scales, confirming their relative independence.

Conclusion. The data suggest that the correlated dimensions of clinical psychosis also have a distribution in the general population, and that depressive symptoms may form an integral part of psychosis-like experiences in the general population.

Background. This study documents changes in household composition and effects on women's quality of life of children leaving and returning home.

Methods. A 9-year annual prospective study of a population-based sample of mid-aged Australian-born women who were premenopausal at baseline (N = 438) was conducted. Documentation was made of household composition and change, well-being, bothersome symptoms, daily hassles, feelings for partner and frequency of sexual activities.

Results. There was an increase in the number of women living alone, and a reduction in number of households in which there were children or parents. Each year >25% of women reported a change in household composition. In the first year after the last child departed (N = 155), there was an improvement in women's positive mood and total well-being and a reduction in negative mood and the number of daily hassles. This improvement in mood was confined to those women who at baseline were not worried about children leaving home. In the first year after children return home there was a trend towards reduced frequency of sexual activities but no mood changes.

Conclusions. For the majority of women, the departure of the last child from the household leads to positive changes in women's mood state and a reduced number of daily hassles. Return of offspring may have an adverse effect on sexual relating of the parents.

Background. It has been suggested that the increased incidence of psychosis in African-Caribbeans living in England may be due to illnesses in which social stress plays an important aetiological role. If this is the case, the prevalence of factors associated with psychosis that predate illness onset such as obstetric complications, pre-morbid neurological illness and poor childhood social adjustment may be expected to be lower in African-Caribbean than Whites psychotic patients.

Method. Details of obstetric complications, pre-morbid neurological illness, and pre-morbid social adjustment were obtained for 337 psychotic patients by patient interview, interviews of mothers and chart review. The proportions of patients with each ‘risk factor’ in the African-Caribbean (N = 103) and White (N = 184) groups were compared using regression analysis; age, sex, social class, diagnosis and referral status were possible explanatory variables.

Results. African-Caribbean patients were less likely to have suffered a pre-morbid neurological disorder than their White counterparts (odds ratio 0.19, 95% CI 0.06–0.61). There was no significant difference in pre-morbid social adjustment or obstetric complications between the two groups, though fewer obstetric complications were reported in the African-Caribbean group (21.5%) than the White group (30.9%).

Conclusions. African-Caribbean patients with psychosis have experienced less pre-morbid neurological illness.

Background. This study examined the effects of psychological stress on platelet α2-adrenergic receptor (α2-AR) binding sites in relation to stress-induced anxiety and changes in the inflammatory response system (IRS).

Methods. The maximum number of binding sites (Bmax) and their affinity (Kd) for [3H]-rauwolscine, a selective α2-AR antagonist, and the stimulated production of tumor necrosis factor-α (TNFα), the Th1-like cytokine, interferon-γ (IFNγ), and the Th2-like cytokines, interleukin-10 (IL-10) and IL-5, were measured in 35 university students a few weeks before (baseline) as well as on the day before a difficult, oral examination (stress condition). The State-Trait-Anxiety Inventory (STAI)- was recorded during both conditions. The Minnesota Multiphase Personality Inventory (MMPI-2TM) was used to assess psychasthenia (Scale 7).

Results. Academic examination stress induced a significant increase in α2-AR density in students whose STAI scores increased in the stress period, in female students and in students who scored higher on psychasthenia. There were significant and positive correlations between stress-induced anxiety and changes in α2-AR density. Stress-induced anxiety was accompanied by a pro-inflammatory and Th1-like response, i.e. increased IFNγ and TNFα production. The stress-induced changes in platelet α2-AR density were significantly and positively related to the production of TNFα, IL-10 and IL-5 and negatively to that of IFNγ.

Conclusions. Subchronic psychological stress in humans induces increased α2-AR density, which is related to stress-induced anxiety, an anxiety-prone constitution and female sex. Increased α2-AR density is accompanied by a Th2-like response and increased TNFα production. The results suggest that: (i) α2-AR density is sensitive to graded differences in stress-induced anxiety; and (ii) psychological stress is accompanied by intertwined responses in the catecholaminergic system, such as α2-ARs, and the IRS, such as Th1/Th2-like functions and the production of TNFα.

Background. The differential strength of correlation between symptoms, cognition and other patient characteristics with community functioning in first-episode psychosis has not been fully investigated.

Method. In a sample of 66 first-episode psychosis patients demographic variables, ratings of pre-morbid adjustment, positive and negative symptoms, duration of untreated psychosis (DUP) and assessment of cognitive functions at baseline, and symptoms, cognitive functions and adherence to medication at 1 year, were correlated with scores on social relations and activities of daily living (ADL) (outcome) at 1 year. Hierarchical regression analysis was used to confirm the independent contribution of baseline and concurrent variables to functional outcome at 1 year.

Results. Scores on functioning related to social relations and ADL were both significantly correlated with pre-morbid adjustment, all dimensions of residual positive and negative symptoms and adherence to medication at 1 year. Scores on social relations were also modestly correlated with DUP and several cognitive measures at baseline and 1 year (verbal IQ, attention, visual memory, word fluency and working memory). Hierarchical regression confirmed independent contribution of pre-morbid adjustment, total residual symptoms and adherence to medication at 1 year for both dimensions of outcome, and psychomotor poverty and working memory for social relations.

Conclusions. In addition to pre-morbid adjustment potentially malleable variables such as level of residual (but not acute) symptoms, adherence to medication and cognitive deficits are likely to influence outcome on aspects of community functioning in individuals treated for first episode of psychosis.

Background. The DSM-IV definition of hypochondriasis is contrasted with hypochondriacal dimensions as provided by the Whiteley Index (WI) and Illness Attitude Scales (IAS).

Methods. Exploratory factor analysis was conducted on self-report data from 570 patients with mental and psychophysiological disorders. Of these, 319 were additionally diagnosed according to DSM-IV by structured interviews.

Results. The three ‘classic’ factors of the WI labelled disease phobia, somatic symptoms and disease conviction were confirmed. The IAS consisted of two dimensions indicating health anxiety and illness behaviour. The overall scores of both instruments were highly correlated (0·80). Optimal cut-off points for case identification yielded sensitivity/specificity rates of 71/80% (WI) and 72/79% (IAS). The IAS was superior to the WI when patients with hypochondriacal disorder were to be discriminated from non-hypochondriacal somatizers. Largest group differences were found for scales related to affective components (health anxieties), smallest for illness behaviours. Affective components of hypochondriasis explained more variance of diagnostic group membership than somatization symptoms. The subscales of disease phobia (WI) and health anxiety (IAS) were most sensitive to treatment-related changes.

Conclusions. The self-rating scales are valid for screening, case definition and dimensional assessment of hypochondriacal disorder, including the differentiation between hypochondriasis and somatization. The existence of distinguishable affective and cognitive components was confirmed.

Background. Little is known about the relation between coping, social support, and psychological and somatic symptoms among survivors of torture living outside the West.

Method. In a population-based dataset of 315 tortured Bhutanese refugees, univariate and multivariate relationships between coping and social support and symptoms were estimated. These relationships were verified in a second sample of 57 help-seeking Nepalese torture survivors.

Results. A relationship was observed between the total number of coping strategies used and anxiety and depression. Negative coping, in contrast to positive coping, was related to all symptom outcome measures. Received social support was stronger related to symptoms than perceived social support. The findings from the first sample were replicated in the second sample.

Conclusion. We found hypothesized relationships between coping, social support, and psychological and somatic symptoms among survivors of torture living in Nepal. The findings from this study confirm the importance of understanding specific types of coping and social support to develop intervention programmes for torture survivors in Non-western settings.

Background. Three prior population-based twin studies, none of which was nationally representative, suggested that both genetic and familial–environmental factors contribute to family resemblance for lifetime cannabis use. We seek to replicate these results in a US national probability sample of twin and sibling pairs examining only last year cannabis use.

Methods. Cannabis use in the last year was assessed by self-report questionnaire. Biometrical twin analyses were performed.

Results. Twin and sibling resemblance for last-year cannabis use was substantial, and much higher in monozygotic pairs than in dizygotic and sibling pairs, where levels of resemblance were similar. Modeling suggested that sibling resemblance was due to genetic factors – with a heritability of at least 60% – and probably family environmental factors. No evidence was found that cannabis use was influenced by a special twin environment.

Conclusions. Consistent with prior studies, use of cannabis is substantially influenced by genetic factors but family–environment is also possibly of importance.

Background. Animal studies have shown that neuroactive steroids modulate the activity of the gamma-aminobutyric acid type A/benzodiazepine receptor complex and that these steroids display anxiolytic or anxiogenic activity depending on their positive (e.g. allopregnanolone) or negative allosteric modulation (e.g. dehydroepiandrosterone sulphate) of this receptor. This study compared plasma levels of allopregnanolone, dehydroepiandrosterone sulphate and pregnenolone sulphate in healthy controls and in patients with generalized social phobia, as assessed with the Mini-International Neuropsychiatric Interview.

Methods. Plasma concentrations of allopregnanolone, pregnenolone sulphate, and dehydroepiandrosterone sulphate were measured in 12 unmedicated male patients with generalized social phobia and 12 matched healthy male volunteers.

Results. Concentrations of pregnenolone sulphate were significantly lower in patients with generalized social phobia than in healthy controls. No statistically significant differences were found for the concentrations of allopregnanolone and dehydroepiandrosterone sulphate in plasma.

Conclusions. These results are particularly interesting since we also observed lower pregnenolone sulphate concentrations in male patients suffering from generalized anxiety disorder. Their relevance to the pathophysiology of social anxiety disorder remains to be determined.

Background. Neuroticism is an important personality dimension associated with depressive and anxiety disorders. Both genetic and social factors are thought to contribute to neuroticism. This study aims to examine whether early childhood adversity may be a determinant of neuroticism.

Method. Five hundred and thirty-two abstinent substance dependent patients completed both the Childhood Trauma Questionnaire (CTQ) and the Eysenck Personality Questionnaire (EPQ).

Results. There was a significant relationship between total childhood trauma scores on the CTQ and neuroticism scores on the EPQ. There were also significant relationships between neuroticism and CTQ subscores for emotional abuse, physical abuse, sexual abuse, emotional neglect and physical neglect.

Conclusions. Childhood trauma may be a determinant of neuroticism. This may be one way in which childhood trauma plays a role in the development of psychiatric disorders. General population studies are needed.

Background. Previous researches have suggested that late onset mania is a distinct subtype associated with medical and neurological disorders. Few studies, however, have focused on vascular risk factors.

Methods. Records of 366 bipolar patients were reviewed and age of first psychiatric hospitalization determined. Late-onset cases were determined empirically from a distribution histogram. Late onset cases were matched to early onset cases and histories of vascular disease/risks and current cholesterol levels compared.

Results. The distribution of age of first psychiatric hospitalization was bimodal with an intermode at age 47. Using that threshold, 6·3% of the cohort was classified as having late onset mania. Vascular risks factors were greater and current cholesterol levels higher in the late onset group.

Conclusions. Late onset mania is associated with greater vascular risk factors. The bimodal appearance of age of first psychiatric hospitalization in this study provides further support of late onset mania as a distinct manic subtype with possibly a different, vascular aetiology. Control of these vascular risks may impact on the incidence of late onset mania, as well as on its clinical management.

Objective. To determine the relationship between self-perceived health and suicidal ideation among patients in general internal medicine practice settings.

Method. A representative sample of 4007 patients was assessed for current suicidal ideation, self-perception of health, current medical disorders and current mental disorders (major depression, generalized anxiety disorder, panic attacks and alcohol use disorder) with the PRIME-MD Patient Health Questionnaire.

Results. Patients with self-perception of poor health, compared with those who reported more favourable health perception, were significantly more likely to report current suicidal ideation (35.2% v. 8.3%; χ2 = 97.4, df = 1, P<0.0001). Multivariate logistic regression analyses revealed that the perception of poor health was associated with a significantly increased risk of suicidal ideation (OR = 3.1, CI 1.9, 5.0), even after adjusting for sociodemographic characteristics, mental disorders and common physical illnesses.

Conclusion. Self-perception of poor health is associated with a significantly increased risk of suicidal ideation among general medical patients, even in the absence of common mental and physical disorders. These findings add to a growing literature on the importance of self-perceived health in the treatment and outcomes of mental and physical well-being by documenting self-perceived poor health as a risk factor for suicidal ideation in medical patients.

Objective. To determine the association between anxiety disorders, panic attack and the risk of major depression among adults in the community.

Method. Data were drawn from the Epidemiologic Catchment Area Program survey waves 1 (N = 20291) and 2 (N = 15849). Multivariate logistic regression analyses were used to determine the risk of incident major depression at 12-month follow-up (wave 2) associated with each anxiety disorder and panic attacks assessed at wave 1, adjusting for differences in sociodemographic characteristics, and then controlling simultaneously for all anxiety disorders, and other psychiatric co-morbidity.

Results. Specific phobia (OR = 1.7 (1.6, 1.8)), agoraphobia (OR = 2.3 (2.2, 2.5)), obsessive–compulsive disorder (OR = 5.4 (5.0, 5.8)) and panic attack (OR = 1.9 (1.8, 2.1)) each made an independent contribution to the risk of major depression, which persisted after adjusting simultaneously for sociodemographic differences and other psychiatric co-morbidity.

Conclusions. Each anxiety disorder and panic attacks appear to confer an independent risk for the onset of major depression within 12-months among adults in the community. Understanding the key role played by anxiety in depression onset is needed for prevention strategies.

Background. A portion of the genetic risk factors for the personality trait neuroticism (N) may also increase risk for major depression (MD). Females have both higher levels of N and higher rates of MD than males, suggesting that these traits may be more genetically correlated in females.

Methods. Structured interviews, including a lifetime assessment for MD by DSM-III-R criteria, were administered to 863 male–male MZ (monozygotic), 649 male–male DZ (dizygotic), 506 female–female MZ, 345 female–female DZ, and 1408 opposite-sex twin pairs. N was assessed using the short-form of the Eysenck Personality Questionnaire. A sex-limited Cholesky model was fitted which allowed us to decompose into additive genetic, common environmental, and individual-specific environmental components two main classes of correlations: within-sex between-variable and between-sex within-variable.

Results. Our best-fitting model contained only additive genetic and individual-specific environmental factors for both N and MD. The within-sex genetic correlations between N and MD were estimated at +0·68 in men and +0·49 in women. This model fitted only slightly better than one in which the N–MD within-sex genetic correlation was constrained to be equal across the sexes, and estimated at +0·55. There may be sex-specific genes influencing both N and MD.

Conclusion. Our best-fitting model failed to establish a significant sex difference in the genetic correlation between N and MD. These results, as well as evidence for sex-specific genetic factors for both traits, have implications for the diagnosis, classification, and treatment of the affective disorders, and molecular genetic approaches to the study of these traits.

Background. There is discrepancy in findings on spousal concordance for major depression. Here we report the risk of depression and its determinants in spouses of persons with or without depression, taking into account several known risk factors for major depression.

Methods. A random sample of non-institutionalized Finnish individual aged 15–75 years was interviewed in the 1996 National Health Care Survey. The sample included 1708 male–female spouse pairs. Major depressive episode (MDE) during the last 12 months was assessed using the Short Form of the University of Michigan version of the Composite International Diagnostic Interview (the UM-CIDI Short Form). Risk factors were assessed in the same interview.

Results. Factors associating with MDE were spouse's MDE, own alcohol intoxication at least once a week and own chronic medical conditions. In addition, there was a strong association between female's current smoking and male's MDE, independently of other risk factors and spousal MDE. The association of MDE with spouses's MDE was not affected by taking into account other assessed risk factors (own or spouse's).

Conclusions. The results indicate elevated spouse concordance for MDE independent of the risk factors assessed in the present study.

Background. Although there is evidence that genetic factors influence individual differences in environmental risk exposure, there are few findings on genetic effects on differential parenting. The present study sought to examine this issue.

Methods. The sample comprised 1117 pairs of like-sex male and female twins, aged 8–16 years, and their parents, recruited from the school population of Virginia. Differential ratings of the within-family experiences were provided by the Twin Inventory of Relationships and Experiences (TIRE).

Results. Dimensions describing the within-family environment based on differential ratings contrasting the twins with one another, were influenced, to an approximately equal extent, by both genetic and environmental factors.

Conclusions. The findings suggest that genetic differences between like-sex siblings lead them to experience their family environment differently, but also that environmental influences significantly affect interactions within the family.

Background. A number of challenge studies have reported abnormalities of serotonergic function in borderline personality disorder (BPD). There are, however, problems with the pharmacological probes used in these studies since fenfluramine and m-CPP are not only serotonergic agents but also induce release of catecholamines, particularly dopamine. Therefore, we tested whether subjects with BPD showed a blunted prolactin (PRL) response to flesinoxan, a highly potent and selective 5-HT1A agonist.

Methods. Flesinoxan challenge test was carried out in 20 BPD in-patients and 20 healthy controls matched for gender but not for age. Since 16 BPD in-patients exhibited major depressive co-morbidity, a group of 20 depressed in-patients matched for gender but not for age was also included.

Results. BPD in-patients exhibited blunted PRL responses as compared to controls, whereas depressed in-patients did not differ from controls. Moreover, PRL responses were lower among BPD in-patients than among depressed in-patients. Among the BPD in-patients, PRL responses to flesinoxan were lower in patients with past history of suicide attempts (N = 8) than in those with a negative history.

Conclusions. The results show major involvement of serotonergic function in BPD and are consistent with previous studies linking lower serotonergic activity with impulsivity. More particularly, our data suggest that BPD is characterized by lower 5-HT1A receptor sensitivity. Moreover, the data support the involvement of 5-HT1A activity in suicidal behaviour. However, this conclusion is limited because other hormonal responses such as ACTH and cortisol were not assessed, and because BPD was assessed by a self-report questionnaire and not a structured clinical interview.