Approximately 25% of people with bulimia nervosa (BN) who undertake therapy are treated in groups. National guidelines do not discriminate between group and individual therapy, yet each has potential advantages and disadvantages and it is unclear how their effects compare. We therefore evaluated how group therapy for BN compares with individual therapy, no treatment, or other therapies, in terms of remission from binges and binge frequency.

We performed a systematic review and meta-analysis of randomized controlled trials of group therapies for BN, following standard guidelines.

A total of 10 studies were included. Studies were generally small with unclear risk of bias. There was low-quality evidence of a clinically relevant advantage for group cognitive behavioural therapy (CBT) over no treatment at therapy end. Remission was more likely with group CBT versus no treatment [relative risk (RR) 0.77, 95% confidence interval (CI) 0.62–0.96]. Mean weekly binges were lower with group CBT versus no treatment (2.9 v. 6.9, standardized mean difference = −0.56, 95% CI −0.96 to −0.15). One study provided low-quality evidence that group CBT was inferior compared with individual CBT to a clinically relevant degree for remission at therapy end (RR 1.24, 95% CI 1.03–1.50); there was insufficient evidence regarding frequency of binges.

Conclusions could only be reached for CBT. Low-quality evidence suggests that group CBT is effective compared with no treatment, but there was insufficient or very limited evidence about how group and individual CBT compared. The risk of bias and imprecise estimates of effect invite further research to refine and increase confidence in these findings.

A meta-analysis of the serotonin1A (5-HT1A) receptor partial agonist of the azapirone class as an anxiolytic drug for the treatment of major depressive disorder (MDD) has not previously been reported.

We carried out a systematic review of the literature available in PubMed, the Cochrane Library database and PsycINFO up to 12 October 2013, and conducted a meta-analysis of randomized controlled trials (RCTs) comparing 5-HT1A agonists with placebo and RCTs of 5-HT1A agonist augmentation therapies for MDD treatment. We calculated the risk ratio (RR), number needed to treat (NNT)/number needed to harm (NNH) and 95% confidence intervals (CIs).

Fifteen RCTs comparing 5-HT1A agonists with placebo (total n = 2469, four studies with buspirone, seven with gepirone, three with ipsapirone and one with zalospirone) were identified. Pooled 5-HT1A agonists had significantly more responders (RR 0.74, 95% CI 0.65–083, p < 0.00001, NNT = 6, 12 trials, n = 1816) than placebo. Pooled 5-HT1A agonists were superior to placebo in discontinuation due to inefficacy (RR 0.49, p = 0.02, NNH = 16, p = 0.03, 10 trials, n = 1494) but were inferior to placebo in discontinuation due to side-effects (RR 1.88, p < 0.0001, NNH = 17, p = 0.001, 13 trials, n = 2196). However, all-cause discontinuation was similar in both groups (RR 0.99, p = 0.85, 14 trials, n = 2402). Four 5-HT1A agonist augmentation studies were identified (total n = 365, three buspirone studies and one tandospirone study). There were no statistically significant effects of 5-HT1A agonist augmentation therapies on response rate (RR 0.98, p = 0.85, four trials, n = 341). 5-HT1A agonist-related side-effects including gastrointestinal symptoms, dizziness, insomnia, palpitation, paresthesia and sweating were greater than with placebo (p < 0.00001 to p = 0.03).

Our results suggest that 5-HT1A agonist has a more beneficial effect on MDD than placebo, but has several side-effects.

Testosterone may be a biological factor that protects males against eating disorders. Elevated prenatal testosterone exposure is linked to lower levels of disordered eating symptoms, but effects emerge only after mid-puberty. Whether circulating levels of testosterone account for decreased risk for disordered eating in boys after mid-puberty is currently unknown; however, animal data support this possibility. In rodents, prenatal testosterone's masculinizing effects on sex-differentiated behaviors emerge during puberty when circulating levels of testosterone increase and ‘activate’ the expression of masculinized phenotypes. This study investigated whether higher levels of circulating testosterone predict lower levels of disordered eating symptoms in adolescent boys, and in particular whether effects are associated with advancing pubertal maturation.

Participants were 213 male twins from the Michigan State University Twin Registry. The Minnesota Eating Behavior Survey and Eating Disorder Examination Questionnaire assessed several disordered eating symptoms. The Pubertal Development Scale assessed pubertal status. Afternoon saliva samples were assayed for testosterone using enzyme immunoassays.

Consistent with animal data, higher levels of circulating testosterone predicted lower levels of disordered eating symptoms in adolescent boys and effects emerged with advancing puberty. Results were not accounted for by several important covariates, including age, adiposity, or mood/anxiety symptoms.

Findings suggest that elevated circulating testosterone may be protective and underlie decreased risk for eating pathology in males during/after puberty, whereas lower levels of testosterone may increase risk and explain why some, albeit relatively few, males develop eating disorders.

Depressive disorders are highly prevalent in adolescence and confer a heightened risk of recurrence in adulthood. Insight into the developmental neurocircuitry of depression could advance our understanding of depression and aid the development of effective treatment strategies. Whereas white-matter (WM) abnormalities are strongly implicated in adult depression, we still lack a firm understanding of WM architecture in adolescent depression. Using diffusion tensor imaging (DTI), we set out to investigate WM microstructure in a sample of clinically depressed adolescents relative to matched controls.

We employed tract-based spatial statistics (TBSS) to examine WM microstructure in 25 treatment-naive adolescents with clinical depression relative to 21 matched controls. Using TBSS, we examined fractional anisotropy (FA), axial diffusivity (AD), radial diffusivity (RD) and mean diffusivity (MD). Threshold-free cluster enhancement (TFCE) with family-wise error (FWE) correction was used to control for multiple comparisons.

Our analysis revealed abnormal WM microstructure in clinically depressed adolescents. More specifically, whole-brain analysis revealed that patients had lower FA values in the body of the corpus callosum (CC), coupled with elevated RD and MD, and preserved AD. Conversely, region-of-interest analysis revealed that patients had higher FA values in the uncinate fasciculus (UF), coupled with elevated AD, reduced RD and preserved MD.

In line with neurocircuitry models of depression, our findings suggest that WM abnormalities within pathways facilitating cognitive and emotional functioning are involved in the pathophysiology of depression. Importantly, our findings show that these WM abnormalities are already present early in the course of the disorder.

Antidepressants are frequently prescribed but results regarding their efficacy have been equivocal for different spectra of the severity continuum and their side-effects are often burdensome. Non-adherence is a likely consequence. The objective was therefore to examine patients’ trade-offs between the efficacy, side-effects and other drawbacks of antidepressants and whether these trade-offs predicted non-adherence.

Trade-offs from 225 antidepressant users, recruited through community pharmacies, were assessed with an Adaptive Conjoint Analysis (ACA) choice task that was customized to each individual patient. From the estimated utilities, relative importance scores of treatment properties were calculated. Non-adherence was measured through self-report and pharmacy refill data.

Relapse prevention and symptom relief were on average equally important. Side-effects were as important and the side-effect stomach and intestine complaints was on average even slightly more important than relapse prevention and symptom relief. Additional treatment with psychotherapy was preferred by 61% of the patients. A benefit/drawback ratio revealed that 18% of the patients did not consider the efficacy to outweigh the drawbacks. A higher benefit/drawback ratio was associated with a decreased odds of intentional non-adherence [odds ratio (OR) 0.2, 95% confidence interval (CI) 0.07–0.7, Wald = 6.7, p = 0.01).

For nearly one in five patients, the efficacy of antidepressants does not outweigh their drawbacks. Knowing patients’ trade-offs is likely to aid both physicians and patients to identify important treatment preferences, to improve adherence and to make more deliberate decisions on whether or not to continue treatment.

Postpartum depression (PPD) affects approximately 13% of women and has a negative impact on mother and infant, hence reliable biological tests for early detection of PPD are essential. We aimed to identify robust predictive biomarkers for PPD using peripheral blood gene expression profiles in a hypothesis-free genome-wide study in a high-risk, longitudinal cohort.

We performed a genome-wide association study in a longitudinal discovery cohort comprising 62 women with psychopathology. Gene expression and hormones were measured in the first and third pregnancy trimesters and early postpartum (201 samples). The replication cohort comprised 24 women with third pregnancy trimester gene expression measures. Gene expression was measured on Illumina-Human HT12 v4 microarrays. Plasma estradiol and estriol were measured. Statistical analysis was performed in R.

We identified 116 transcripts differentially expressed between the PPD and euthymic women during the third trimester that allowed prediction of PPD with an accuracy of 88% in both discovery and replication cohorts. Within these transcripts, significant enrichment of transcripts implicated that estrogen signaling was observed and such enrichment was also evident when analysing published gene expression data predicting PPD from a non-risk cohort. While plasma estrogen levels were not different across groups, women with PPD displayed an increased sensitivity to estrogen signaling, confirming the previously proposed hypothesis of increased sex-steroid sensitivity as a susceptibility factor for PPD.

These results suggest that PPD can be robustly predicted in currently euthymic women as early as the third trimester and these findings have implications for predictive testing of high-risk women and prevention and treatment for PPD.

Extremely high rates of suicide localized within subgroups of populations where suicide is rare have been reported. We investigated this intriguing observation in a population of South-East Asia, where local culture should theoretically be preventative of suicide.

A team including an anthropologist and a psychiatrist surveyed all cases of suicide that had occurred over 10 years in four isolated regions. A psychological autopsy was carried out comparing each suicide case with two matched control cases.

In a region of 1192 inhabitants, 16 suicides occurred, leading to an annual suicide rate of 134/1 000 00 which is 10 times the rate in the USA or Canada. By contrast, three ethnically similar distant communities showed low to null rates. The gender ratio was three males to one female and two-thirds of cases were aged below 35 years. Methods of suicide were poisoning and hanging and motives mainly included interpersonal discord. The pattern of developmental and clinical risk factors was somewhat different from Western countries, showing no childhood maltreatment, only one case of alcohol/substance abuse and impulsive–aggressive personality but elevated rates of social anxiety. Suicide cases had very high frequencies of second-degree biological relatives who committed suicide.

Our study confirms a persistent phenomenon of high suicide rates restricted to a subgroup of a pre-industrialized population. We hypothesized this might be explained by isolation and endogamy, which may have promoted the selection/amplification of genetic vulnerability factors, or a contagion effect. These findings shed light on suicide from both a singular and a universal perspective, suggesting that particular local conditions may significantly modulate the rate of this complex behavior.

The aim of the present study was to investigate trajectories of suicide attempt risks before and after granting of disability pension in young people.

The analytic sample consisted of all persons 16–30 years old and living in Sweden who were granted a disability pension in the years 1995–1997; 2000–2002 as well as 2005–2006 (n = 26 624). Crude risks and adjusted odds ratios for suicide attempt were computed for the 9-year window around the year of disability pension receipt by repeated-measures logistic regressions.

The risk of suicide attempt was found to increase continuously up to the year preceding the granting of disability pension in young people, after which the risk declined. These trajectories were similar for women and men and for disability pension due to mental and somatic diagnoses. Still, the multivariate odds ratios for suicide attempts for women and for disability pension due to mental disorders were 2.5- and 3.8-fold increased compared with the odds ratios for men and disability pension due to somatic disorders, respectively. Trajectories of suicide attempts differed for young individuals granted a disability pension during 2005–2006 compared with those granted during 1995–1997 and 2000–2002.

We found an increasing risk of suicide attempt up until the granting of a disability pension in young individuals, after which the risk decreased. It is of clinical importance to monitor suicide attempt risk among young people waiting for the granting of a disability pension.

Despite the inclusion of disruptive mood dysregulation disorder (DMDD) in DSM-5, little empirical data exist on the disorder. We estimated rates, co-morbidity, correlates and early childhood predictors of DMDD in a community sample of 6-year-olds.

DMDD was assessed in 6-year-old children (n = 462) using a parent-reported structured clinical interview. Age 6 years correlates and age 3 years predictors were drawn from six domains: demographics; child psychopathology, functioning, and temperament; parental psychopathology; and the psychosocial environment.

The 3-month prevalence rate for DMDD was 8.2% (n = 38). DMDD occurred with an emotional or behavioral disorder in 60.5% of these children. At age 6 years, concurrent bivariate analyses revealed associations between DMDD and depression, oppositional defiant disorder, the Child Behavior Checklist – Dysregulation Profile, functional impairment, poorer peer functioning, child temperament (higher surgency and negative emotional intensity and lower effortful control), and lower parental support and marital satisfaction. The age 3 years predictors of DMDD at age 6 years included child attention deficit hyperactivity disorder, oppositional defiant disorder, the Child Behavior Checklist – Dysregulation Profile, poorer peer functioning, child temperament (higher child surgency and negative emotional intensity and lower effortful control), parental lifetime substance use disorder and higher parental hostility.

A number of children met DSM-5 criteria for DMDD, and the diagnosis was associated with numerous concurrent and predictive indicators of emotional and behavioral dysregulation and poor functioning.

Threshold and subthreshold forms of generalized anxiety disorder (GAD) are highly prevalent and impairing conditions among adults. However, there are few general population studies that have examined these conditions during the early life course. The primary objectives of this study were to: (1) examine the prevalence, and sociodemographic and clinical characteristics of threshold and subthreshold forms of GAD in a nationally representative sample of US youth; and (2) test differences in sociodemographic and clinical characteristics between threshold and subthreshold forms of the disorder.

The National Comorbidity Survey-Adolescent Supplement is a nationally representative face-to-face survey of 10 123 adolescents 13 to 18 years of age in the continental USA.

Approximately 3% of adolescents met criteria for threshold GAD. Reducing the required duration from 6 months to 3 months resulted in a 65.7% increase in prevalence (5.0%); further relaxing the uncontrollability criterion led to an additional 20.7% increase in prevalence (6.1%). Adolescents with all forms of GAD displayed a recurrent clinical course marked by substantial impairment and co-morbidity with other psychiatric disorders. There were few significant differences in sociodemographic and clinical characteristics between threshold and subthreshold cases of GAD. Results also revealed age-related differences in the associated symptoms and clinical course of GAD.

Findings demonstrate the clinical significance of subthreshold forms of GAD among adolescent youth, highlighting the continuous nature of the GAD construct. Age-related differences in the associated symptoms and clinical course of GAD provide further support for criteria that capture variation in clinical features across development.

Despite their high prevalence, the global burden of anxiety disorders has never been calculated comprehensively. The new Global Burden of Disease (GBD) study has estimated burden due to morbidity and mortality caused by any anxiety disorder.

Prevalence was estimated using Bayesian meta-regression informed by data identified in a systematic review. Years of life lived with disability (YLDs) were calculated by multiplying prevalent cases by an average disability weight based on severity proportions (mild, moderate and severe). Disability-adjusted life years (DALYs) were then calculated and age standardized using global standard population figures. Estimates were also made for additional suicide mortality attributable to anxiety disorders. Findings are presented for YLDs, DALYs and attributable burden due to suicide for 21 world regions in 1990 and 2010.

Anxiety disorders were the sixth leading cause of disability, in terms of YLDs, in both high-income (HI) and low- and middle-income (LMI) countries. Globally, anxiety disorders accounted for 390 DALYs per 100 000 persons [95% uncertainty interval (UI) 191–371 DALYs per 100 000] in 2010, with no discernible change observed over time. Females accounted for about 65% of the DALYs caused by anxiety disorders, with the highest burden in both males and females experienced by those aged between 15 and 34 years. Although there was regional variation in prevalence, the overlap between uncertainty estimates means that substantive differences in burden between populations could not be identified.

Anxiety disorders are chronic, disabling conditions that are distributed across the globe. Future estimates of burden could be further improved by obtaining more representative data on severity state proportions.

Although prior genetic studies of interview-assessed fears and phobias have shown that genetic factors predispose individuals to fears and phobias, they have been restricted to the DSM-III to DSM-IV aggregated subtypes of phobias rather than to individual fearful and phobic stimuli.

We examined the lifetime history of fears and/or phobias in response to 21 individual phobic stimuli in 4067 personally interviewed twins from same-sex pairs from the Virginia Adult Twin Study of Psychiatric and Substance Abuse Disorders (VATSPSUD). We performed multivariate statistical analyses using Mx and Mplus.

The best-fitting model for the 21 phobic stimuli included four genetic factors (agora-social-acrophobia, animal phobia, blood-injection-illness phobia and claustrophobia) and three environmental factors (agora-social-hospital phobia, animal phobia, and situational phobia).

This study provides the first view of the architecture of genetic and environmental risk factors for phobic disorders and their subtypes. The genetic factors of the phobias support the DSM-IV and DSM-5 constructs of animal and blood-injection-injury phobias but do not support the separation of agoraphobia from social phobia. The results also do not show a coherent genetic factor for the DSM-IV and DSM-5 situational phobia. Finally, the patterns of co-morbidity across individual fears and phobias produced by genetic and environmental influences differ appreciably.

Panic disorder with agoraphobia is characterized by panic attacks and anxiety in situations where escape might be difficult. However, neuroimaging studies specifically focusing on agoraphobia are rare. Here we used functional magnetic resonance imaging (fMRI) with disorder-specific stimuli to investigate the neural substrates of agoraphobia.

We compared the neural activations of 72 patients suffering from panic disorder with agoraphobia with 72 matched healthy control subjects in a 3-T fMRI study. To isolate agoraphobia-specific alterations we tested the effects of the anticipation and perception of an agoraphobia-specific stimulus set. During fMRI, 48 agoraphobia-specific and 48 neutral pictures were randomly presented with and without anticipatory stimulus indicating the content of the subsequent pictures (Westphal paradigm).

During the anticipation of agoraphobia-specific pictures, stronger activations were found in the bilateral ventral striatum and left insula in patients compared with controls. There were no group differences during the perception phase of agoraphobia-specific pictures.

This study revealed stronger region-specific activations in patients suffering from panic disorder with agoraphobia in anticipation of agoraphobia-specific stimuli. Patients seem to process these stimuli more intensively based on individual salience. Hyperactivation of the ventral striatum and insula when anticipating agoraphobia-specific situations might be a central neurofunctional correlate of agoraphobia. Knowledge about the neural correlates of anticipatory and perceptual processes regarding agoraphobic situations will help to optimize and evaluate treatments, such as exposure therapy, in patients with panic disorder and agoraphobia.

Individuals with borderline personality disorder (BPD) frequently display co-morbid mental disorders. These disorders include ‘internalizing’ disorders (such as major depressive disorder and anxiety disorders) and ‘externalizing’ disorders (such as substance use disorders and antisocial personality disorder). It is hypothesized that these disorders may arise from latent ‘internalizing’ and ‘externalizing’ liability factors. Factor analytic studies suggest that internalizing and externalizing factors both contribute to BPD, but the extent to which such contributions are familial is unknown.

Participants were 368 probands (132 with BPD; 134 without BPD; and 102 with major depressive disorder) and 885 siblings and parents of probands. Participants were administered the Diagnostic Interview for DSM-IV Personality Disorders, the Revised Diagnostic Interview for Borderlines, and the Structured Clinical Interview for DSM-IV.

On confirmatory factor analysis of within-person associations of disorders, BPD loaded moderately on internalizing (factor loading 0.53, s.e. = 0.10, p < 0.001) and externalizing latent variables (0.48, s.e. = 0.10, p < 0.001). Within-family associations were assessed using structural equation models of familial and non-familial factors for BPD, internalizing disorders, and externalizing disorders. In a Cholesky decomposition model, 84% (s.e. = 17%, p < 0.001) of the association of BPD with internalizing and externalizing factors was accounted for by familial contributions.

Familial internalizing and externalizing liability factors are both associated with, and therefore may mutually contribute to, BPD. These familial contributions account largely for the pattern of co-morbidity between BPD and internalizing and externalizing disorders.

Oxidative stress and neurotrophic factors have been implicated in the pathophysiology of bipolar disorder. Our objective was to determine whether plasma glutathione or brain-derived neurotrophic factor (BDNF) levels were abnormal in bipolar disorder and therefore useful as possible biomarkers.

Blood samples were collected from subsyndromal, medicated bipolar I patients (n = 50), recruited from OXTEXT, University of Oxford, and from 50 matched healthy controls. Total and oxidized glutathione levels were measured using an enzymatic recycling method and used to calculate reduced, percentage oxidized, ratio of reduced:oxidized and redox state. BDNF was measured using an enzyme-linked immunoassay. Self-monitored mood scores for the bipolar group were available (Quick Inventory of Depressive Symptomatology and the Altman Self-Rating Mania Scale) over an 8-week period.

Compared with controls, bipolar patients had significantly lower levels of total glutathione and it was more oxidized. BDNF levels were not different. Age of illness onset but not current mood state correlated with total glutathione levels and its oxidation status, so that lower levels of total and reduced glutathione were associated with later onset of disease, not length of illness.

Plasma glutathione levels and redox state detect oxidative stress even in subsyndromal patients with normal BDNF. It may relate to the onset and development of bipolar disorder. Plasma glutathione appears to be a suitable biomarker for detecting underlying oxidative stress and for evaluating the efficacy of antioxidant intervention studies.

The extent to which different symptom dimensions vary according to epidemiological factors associated with categorical definitions of first-episode psychosis (FEP) is unknown. We hypothesized that positive psychotic symptoms, including paranoid delusions and depressive symptoms, would be more prominent in more urban environments.

We collected clinical and epidemiological data on 469 people with FEP (ICD-10 F10–F33) in two centres of the Aetiology and Ethnicity in Schizophrenia and Other Psychoses (AESOP) study: Southeast London and Nottinghamshire. We used multilevel regression models to examine neighbourhood-level and between-centre differences in five symptom dimensions (reality distortion, negative symptoms, manic symptoms, depressive symptoms and disorganization) underpinning Schedules for Clinical Assessment in Neuropsychiatry (SCAN) Item Group Checklist (IGC) symptoms. Delusions of persecution and reference, along with other individual IGC symptoms, were inspected for area-level variation.

Reality distortion [estimated effect size (EES) 0.15, 95% confidence interval (CI) 0.06–0.24] and depressive symptoms (EES 0.21, 95% CI 0.07–0.34) were elevated in people with FEP living in more urban Southeast London but disorganized symptomatology was lower (EES –0.06, 95% CI –0.10 to –0.02), after controlling for confounders. Delusions of persecution were not associated with increased neighbourhood population density [adjusted odds ratio (aOR) 1.01, 95% CI 0.83–1.23], although an effect was observed for delusions of reference (aOR 1.41, 95% CI 1.12–1.77). Hallucinatory symptoms showed consistent elevation in more densely populated neighbourhoods (aOR 1.32, 95% CI 1.09–1.61).

In people experiencing FEP, elevated levels of reality distortion and depressive symptoms were observed in more urban, densely populated neighbourhoods. No clear association was observed for paranoid delusions; hallucinations were consistently associated with increased population density. These results suggest that urban environments may affect the syndromal presentation of psychotic disorders.

People with Prader–Willi syndrome (PWS), a genetically defined developmental disorder, are at increased risk of developing psychotic illness. This is particularly the case for those with a genetic subtype of PWS called maternal uniparental disomy (mUPD), where rates of psychosis are more than 60% by early adult life. Little is known about the long-term course of their disorder.

Individuals who had had episodes of psychosis or were at increased risk of developing psychosis due to their genetic subtype and had taken part in a previous study were contacted. Ten people were untraceable or deceased, leaving a total of 38 potential participants. Of these, 28 agreed to take part in a follow-up interview or complete a questionnaire about their mental health and medication. This represented 20/35 (57.1%) people from the original study who had had psychosis and 8/13 (61.5%) people who were at risk due to their genetic subtype. They were thought to be representative of those groups as a whole based on IQ and number of episodes of psychosis.

Two individuals had had recurrent episodes of psychosis while all others remained well. There were no new-onset cases of psychosis in those at risk. Individuals with PWS remained on high levels of psychiatric medication throughout the follow-up period.

Recurrent episodes of psychosis may be rare in people with PWS once stability has been achieved in the management of their illness. We speculate that this may be due to the protective influence of medication.

Observing another person performing a complex action accelerates the observer's acquisition of the same action and limits the time-consuming process of learning by trial and error. Learning by observation requires specific skills such as attending, imitating and understanding contingencies. Individuals with autism spectrum disorder (ASD) exhibit deficits in these skills.

The performance of 20 ASD children was compared with that of a group of typically developing (TD) children matched for chronological age (CA), IQ and gender on tasks of learning of a visuomotor sequence by observation or by trial and error. Acquiring the correct sequence involved three phases: a detection phase (DP), in which participants discovered the correct sequence and learned how to perform the task; an exercise phase (EP), in which they reproduced the sequence until performance was error free; and an automatization phase (AP), in which by repeating the error-free sequence they became accurate and speedy.

In the DP, ASD children were impaired in detecting a sequence by trial and error only when the task was proposed as first, whereas they were as efficient as TD children in detecting a sequence by observation. In the EP, ASD children were as efficient as TD children. In the AP, ASD children were impaired in automatizing the sequence. Although the positive effect of learning by observation was evident, ASD children made a high number of imitative errors, indicating marked tendencies to hyperimitate.

These findings demonstrate the imitative abilities of ASD children although the presence of imitative errors indicates an impairment in the control of imitative behaviours.

The common mental disorders (CMDs) of anxiety and depression are the most common form of poor mental health in the general population. Evidence from the small number of previous cohort studies on the role of neighbourhood factors in mental health is inconclusive. We tested the hypothesis that high levels of neighbourhood social cohesion modify an adverse association between change in individual mental health and neighbourhood deprivation.

We carried out a longitudinal multilevel analysis using data from the Caerphilly Health and Social Needs Cohort Study with a 7-year follow-up (n = 4426; age range 18–74 years at baseline). Neighbourhood deprivation and neighbourhood social cohesion were assessed at baseline and change in mental health between follow-up and baseline was assessed using the five-item Mental Health Inventory (MHI-5).

Residence in the most deprived neighbourhoods was negatively associated with change in mental health, after adjusting for baseline individual socio-economic risk factors and transitions in life events. This negative effect was significantly reduced in high social cohesion neighbourhoods. The predicted change in mental health score was calculated for the 10th and 90th centiles of the household low-income distribution. The difference between them was −2.8 in the low social cohesion group and 1.1 in the high cohesion group. The difference between the groups was 3.9 [95% confidence interval (CI) 0.2–7.6].

The public health burden of poor mental health and mental health inequality could potentially be reduced by strengthening social cohesion in deprived neighbourhoods. This offers a mechanism to address the adverse effect of neighbourhood deprivation on population mental health.