In this issue

In the first paper, Senn & Carey (pp. 355–363) report findings from a systematic review of 18 studies of the prevalence and correlates of, and interventions to promote, HIV testing in those with a severe mental illness (SMI). The authors found that, on average, less than 50% of those with a SMI had been tested for HIV in the previous year. The factors most consistently associated with testing were risky sexual and drug-related behaviours. The authors further found that interventions to promote testing are not well developed. The authors conclude by noting the need for more research in this area.


Schizophrenia
In the first of three papers on aspects of schizophrenia, Schü rhoff et al. (pp. 365-370) investigated the association between schizotypal traits and childhood trauma in a sample of 67 first-degree relatives of probands with schizophrenia and 71 first-degree relatives of probands with bipolar disorder. The authors found that schizotypal traits (particularly on the positive dimension) were associated with childhood trauma in the relatives of those with schizophrenia, but not bipolar. The authors conclude that susceptibility genes for schizophrenia may interact with childhood trauma in the emergence of schizotypal traits. , using magnetic resonance imaging, examined the cerebral correlates of neurological soft signs (NSS) in a sample of 42 cases with a first episode of schizophrenia or schizophreniform disorder and 22 controls. The authors found that, in the cases, NSS were associated with reduced grey-and white-matter densities in the pre-and post-central gyrus, pre-motor area, middle and inferior frontal gyri, cerebellum, caudate nucleus and thalamus. These associations were not evident in controls. The authors conclude that the pattern of associations supports the model of ' cognitive dysmetria ' with a disrupted cortico-cerebellar-thalamic-cortical circuit in schizophrenia.  present findings from a meta-analysis of 88 studies of associations between schizophrenic syndromes and executive impairment. In pooled correlation analyses, the authors found evidence of small to modest associations between negative and disorganized syndromes (but not positive or ' reality distortion ') and executive impairment. These associations were strongest in those with a stable, compared with a relapsing and remitting, clinical picture. In addition, negative and disorganized syndromes were associated with poorer general intellectual function.

Depression and suicide
In the first of six papers on aspects of depression and suicide,  investigated the relationship between cognitive function and clinical and functional outcomes in a sample of 48 subjects assessed on admission to hospital with major depressive disorder (MDD) and 3 months later. The authors found that more perseverative errors on the shortened Wisconsin Card Sorting Test were associated with both worse clinical and social and occupational outcomes. In addition, poor event-based prospective memory was associated with a worse social and occupational outcome. The authors conclude that a brief cognitive screen on admission, focusing on executive function, may have prognostic value in depression.  sought to devise an index of the course of MDD and test its predictive validity in a sample of 426 subjects with a history of MDD by age 30, using principal components and confirmatory factor analyses. The authors found that three indicators loaded highly on a chronic course index with predictive validity : early age of onset, number of episodes, and amount of time ill. The authors further found that a chronic course was associated with symptom severity, likelihood of treatment utilization, and psychosocial impairment.  investigated the relationship between cognitive symptoms of depression and lowgrade chronic inflammatory markers (C-reactive protein and interleukin-6) in a 12-year follow-up of the Whitehall II study. In a sample of over 3000 subjects the authors found that both markers were associated with cognitive symptoms of depression at follow-up, independent of the number of potential confounders. Baseline symptoms of depression did not predict inflammatory markers at follow-up. The authors conclude that the findings suggest inflammation precedes the onset of cognitive symptoms of depression.
Chen et al. (pp. 425-430) investigated the temporal relationship between syndromes of depression and dementia in samples of 1736 people aged o65 years in China and 5222 people of similar age in the UK who were followed for 1-5 years. The authors found significant cross-sectional associations between syndrome levels of depression and dementia in both samples. At follow-up, however, these associations were less marked. The authors conclude that the relationship between syndromes of depression and dementia may be temporal. ) examined the relative contributions over time of assaultive trauma, nonassaultive trauma, and familial effects on risk of depression in 5266 members of a volunteer Australian twin panel. The authors found that the strongest predictor of immediate and long-term depression was assaultive trauma that occurred in childhood. This outweighed familial effects on childhood-onset depression for most twins. The authors conclude that early assaultive trauma increases risk for depression into adulthood.
Hunt et al. (pp. 443-449) investigated risk factors for suicide, including aspects of health care received, in the period post-discharge from hospital in a sample of 238 psychiatric patients who died by suicide within 3 months of discharge and 238 matched controls. The authors found that risk of suicide was highest in the period immediately following discharge : 43 % of suicides occurred within the first month, and 47 % of these before the first follow-up appointment. The factors associated with suicide included a history of self-harm, self-discharge and a recent last contact with services. Initial compulsory admission to hospital and enhanced aftercare were associated with a reduced risk.

Anorexia nervosa
Two papers examine aspects of anorexia. In the first, Jacobs et al. (pp. 451-461) sought to identify patterns of behavioural traits in a sample of 433 complete anorexia nervosa (AN) trios (probands, mother, father) drawn from the Price Foundation Genetic Study of AN. The authors identified three classes involving probands and mothers only. Fathers did not differ across the classes. Class 1 included low-symptom probands and mothers. Class 2 included probands with marked elevations in drive for thinness, body dissatisfaction and neuroticism and mothers with mild anxious-perfectionistic traits. Class 3 included probands and mothers with elevations on eating-disorder symptoms and anxious-perfectionistic traits.  investigated the genetic and environmental contributions to liability to AN symptoms in a sample of female monozygotic (448 complete pairs, 4 singletons) and dizygotic (263 complete pairs, 4 singletons) twins. The authors found that the heritability of a diagnosis of AN was modest (a 2 =0.22) and that the heritability of specific symptoms varied (e.g. weight loss items : a 2 =0.31-0.34 ; weight concern items : a 2 =0.18-0.29). The authors further found that the variance of amenorrhoea was most strongly influenced by unshared environmental factors.

Other topics
In the first of the final four papers,  examined pathways to mental health care in 1044 patients commencing a new episode of care at Amanuel Hospital in Addis Ababa, Ethiopia. The authors found that 41 % of patients contacted the hospital direct. The remaining patients sought help from up to four different sources prior to contact, including priests/holy water/ church. Delays between onset and contact were associated with no formal education, joblessness and a physical health diagnosis. The authors draw conclusions concerning proposals for improving access to mental health care in Ethiopia.  investigated patterns of admissions to public and private in-patient facilities in Italy during an index period in 2004, capturing 1577 admitted patients. The authors found that non-affective psychoses (36 %) were the most common diagnoses, and that private facilities were more likely to admit patients with organic and substance use disorders. Family sought admission in 54 % of cases. The main factors associated with admission were : impairment in occupational and social function, social withdrawal, and conflict with family members.  examined the influence of genetic and environmental factors on the co-occurrence of four functional somatic illnesses (chronic widespread pain, chronic fatigue, irritable bowel syndrome, recurrent headache) and depression and anxiety in a sample of 31 318 twins aged 41-64 years drawn from the Swedish Twin Registry. The authors found that a common pathway model, with two latent traits shared by all six illnesses, best fit the women's data. In addition, all functional somatic illnesses were affected by genetic influences specific to each.
In the final paper,  investigated changes in functional magnetic resonance imaging activation patterns during the recall of unresolved adverse life events over 1 year in a sample of 13 female patients with borderline personality disorder (BPD). The authors found, between time 1 and time 2, major right more than left differences in activation of the posterior more than the anterior cingulate, superior temporal lobes, insula, and right middle and frontal lobes. No changes were observed in subjective or clinical data. The authors raise the question of whether decreases in neural activation precede clinical improvement in BPD.

C R A I G M O R G A N
Institute of Psychiatry, London, UK