Anorexia nervosa is a potentially deadly psychiatric illness that develops predominantly in females around puberty but is increasingly being recognized as also affecting boys and men and women across the lifespan. The aim of this environmental scan is to provide an overview of best practices in anorexia nervosa treatment across the age spectrum.

A triangulation approach was used. First, a detailed review of randomized controlled trials (RCTs) for anorexia nervosa published between 1980 and 2011 was conducted; second, clinical practice guidelines were consulted and reviewed; third, information about RCTs currently underway was sourced. This approach facilitated a comprehensive overview, which addressed the extant evidence base, recent advances in evidence and improvements in treatment, and future directions.

The evidence base for the treatment of anorexia nervosa is advancing, albeit unevenly. Evidence points to the benefit of family-based treatment for youth. For adults no specific approach has shown superiority and, presently, a combination of renourishment and psychotherapy such as specialist supportive clinical management, cognitive behavioral therapy, or interpersonal psychotherapy is recommended. RCTs have neither sufficiently addressed the more complex treatment approaches seen in routine practice settings, such as multidisciplinary treatment or level of care, nor specifically investigated treatment in ethnically diverse populations. Methodological challenges that hinder progress in controlled research for anorexia nervosa are explained.

The review highlights evidence-based and promising treatment modalities for anorexia nervosa and presents a triangulated analysis including controlled research, practice guidelines, and emerging treatments to inform and support clinical decision making.

Bilateral repetitive magnetic stimulation (rTMS) is a promising novel therapeutic intervention for major depression (MD). However, clinical trials to date have reported conflicting evidence concerning its overall efficacy, which might have resulted from low statistical power. Thus, meta-analytical approaches could be useful in examining this issue by allowing the integration of findings from multiple studies and thus producing more accurate estimates of the treatment effect.

We searched the literature for randomized, double-blind and sham-controlled trials (RCTs) on bilateral rTMS for treating MD from 1995 to July 2012 using EMBASE, PsycINFO, Cochrane Central Register of Controlled Trials, SCOPUS, and ProQuest Dissertations and Theses, and from October 2008 until May 2012 using Medline. The main outcome measures were response and remission rates. We used a random-effects model, odds ratios (ORs) and the number needed to treat.

Data were obtained from seven RCTs, totaling 279 subjects with MD. After an average of 12.9 (s.d. = 2.7) sessions, 24.7% (40/162) and 6.8% (8/117) of subjects receiving active bilateral rTMS and sham rTMS were classified as responders [OR 4.3, 95% confidence interval (CI) 1.95–9.52, p < 0.0001]. Also, 19% (23/121) and 2.6% (2/77) of subjects were remitters following active bilateral rTMS and sham rTMS, respectively (OR 6.0, 95% CI 1.65–21.8, p = 0.006). No difference between baseline mean depression scores for the bilateral and sham rTMS groups was found, and the former was comparable with the latter in terms of drop-out rates at study end. Furthermore, we did not find significant differences efficacy- and acceptability-wise between active bilateral and unilateral rTMS at study end. Finally, heterogeneity between the included RCTs was not significant, and the risk of publication bias was found to be low.

Bilateral rTMS is a promising treatment for MD as it provides clinically meaningful benefits that are comparable with those of standard antidepressants and unilateral rTMS. Furthermore, bilateral rTMS seems to be an acceptable treatment for depressed subjects.

Abnormalities in cortical thickness and subcortical structures have been studied in schizophrenia but little is known about corresponding changes in mania and brain structural differences between these two psychiatric conditions, especially early in the stage of the illness. In this study we aimed to compare cortical thickness and shape of the amygdala–hippocampal complex in first-episode schizophrenia (FES) and mania (FEM).

Structural magnetic resonance imaging (MRI) was performed on 28 FES patients, 28 FEM patients and 28 healthy control subjects who were matched for age, gender and handedness.

Overall, the shape of the amygdala was deformed in both patient groups, relative to controls. Compared to FEM patients, FES patients had significant inward shape deformation in the left hippocampal tail, right hippocampal body and a small region in the right amygdala. Cortical thinning was more widespread in FES patients, with significant differences found in the temporal brain regions when compared with FEM and controls.

Significant differences were observed between the two groups of patients with FES and FEM in terms of the hippocampal shape and cortical thickness in the temporal region, highlighting that distinguishable brain structural changes are present early in the course of schizophrenia and mania.

It is unclear how the course of maternal depressive symptoms affects child development. We modelled trajectories of maternal depressive symptoms from mid-pregnancy to 3 years after childbirth to better determine their associations with child problem behaviour.

Mother–child dyads (n = 4167) participated in a population-based prospective cohort in The Netherlands. Depressive symptoms were assessed with the Brief Symptom Inventory during pregnancy and at 2, 6 and 36 months postnatally. When children were 3 years old, problem behaviour was assessed with the Child Behaviour Checklist completed by each parent. A group-based modelling technique was used to model trajectories of maternal depressive symptoms and to examine their association with child problem behaviour. The added value of trajectory modelling was determined with successive linear regressions.

We identified four trajectories of maternal depressive symptoms; ‘no’ (34%), ‘low’ (54%), ‘moderate’ (11%) and ‘high’ (1.5%). Child problem behaviour varied as a function of maternal trajectory membership. Whether rated by mother or father, children of mothers assigned to higher trajectories had significantly more problem behaviours than children of mothers assigned to lower trajectories. The model including trajectories had additive predictive value over a model relying only on a summed repeated measure of severity and a predefined chronicity variable.

Depending on their course, maternal depressive symptoms have different effects on child problem behaviour. More information is gained by studying trajectories of symptoms, than only predefined measures of severity and chronicity. Moreover, trajectories can help identifying clinically depressed mothers who are possible candidates for early interventions.

Ageing is an important factor in the development of mental health problems and their treatment. We assessed age trajectories of common mental disorders (CMDs) and psychotherapy utilization from adolescence to old age, and examined whether these trajectories were modified by time period or birth cohort effects.

British Household Panel Survey (BHPS) with an 18-year follow-up between 1991 and 2009 (n = 30 224 participants, aged 15–100 years, with an average 7.3 person-observations per person). CMDs were assessed with the 12-item version of the General Health Questionnaire (GHQ). Psychotherapy treatment utilization during the past year was self-reported by the participants. The modifying influences of time period and cohort effects were assessed in a cohort-sequential longitudinal setting.

Following a moderate decrease after age 50, the prevalence of GHQ caseness increased steeply from age 75. This increase was more marked in the 2000s (GHQ prevalence increasing from 24% to 43%) than in the 1990s (from 22% to 34%). Psychotherapy utilization decreased after age 55, with no time period or cohort effects modifying the age trajectory. These ageing patterns were replicated in within-individual longitudinal analysis.

Old age is associated with higher risk of CMDs, and this association has become more marked during the past two decades. Ageing is also associated with an increasing discrepancy between prevalence of mental disorders and provision of treatment, as indicated by lower use of psychotherapy in older individuals.

Whether late-onset depression is a risk factor for or a prodrome of dementia remains unclear. We investigated the impact of depressive symptoms and early- v. late-onset depression on subsequent dementia in a cohort of elderly general-practitioner patients (n = 2663, mean age = 81.2 years).

Risk for subsequent dementia was estimated over three follow-ups (each 18 months apart) depending on history of depression, particularly age of depression onset, and current depressive symptoms using proportional hazard models. We also examined the additive prediction of incident dementia by depression beyond cognitive impairment.

An increase of dementia risk for higher age cut-offs of late-onset depression was found. In analyses controlling for age, sex, education, and apolipoprotein E4 genotype, we found that very late-onset depression (aged ⩾70 years) and current depressive symptoms separately predicted all-cause dementia. Combined very late-onset depression with current depressive symptoms was specifically predictive for later Alzheimer's disease (AD; adjusted hazard ratio 5.48, 95% confidence interval 2.41–12.46, p < 0.001). This association was still significant after controlling for cognitive measures, but further analyses suggested that it was mediated by subjective memory impairment with worries.

Depression might be a prodrome of AD but not of dementia of other aetiology as very late-onset depression in combination with current depressive symptoms, possibly emerging as a consequence of subjectively perceived worrisome cognitive deterioration, was most predictive. As depression parameters and subjective memory impairment predicted AD independently of objective cognition, clinicians should take this into account.

Sleep disturbances are persistent residual symptoms following remission of major depressive disorder (MDD) and are associated with an increased risk of MDD recurrence. The purpose of the current study was to examine the effect of exercise augmentation on self-reported sleep quality in participants with non-remitted MDD.

Participants were randomized to receive selective serotonin reuptake inhibitor (SSRI) augmentation with one of two doses of exercise: 16 kilocalories per kilogram of body weight per week (KKW) or 4 KKW for 12 weeks. Depressive symptoms were assessed using the clinician-rated Inventory of Depressive Symptomatology (IDS-C). The four sleep-related items on the IDS-C (Sleep Onset Insomnia, Mid-Nocturnal Insomnia, Early Morning Insomnia, and Hypersomnia) were used to assess self-reported sleep quality.

Significant decreases in total insomnia (p < 0.0001) were observed, along with decreases in sleep onset, mid-nocturnal and early-morning insomnia (p's <0.002). Hypersomnia did not change significantly (p = 0.38). Changes in total, mid-nocturnal and early-morning insomnia were independent of changes in depressive symptoms. Higher baseline hypersomnia predicted a greater decrease in depression severity following exercise treatment (p = 0.0057). No significant moderating effect of any baseline sleep on change in depression severity was observed. There were no significant differences between exercise treatment groups on total insomnia or any individual sleep item.

Exercise augmentation resulted in improvements in self-reported sleep quality in patients with non-remitted MDD. Given the prevalence of insomnia as a residual symptom following MDD treatment and the associated risk of MDD recurrence, exercise augmentation may have an important role in the treatment of MDD.

The cortisol awakening response (CAR) has been shown to predict major depressive episodes (MDEs) over a 1-year period. It is unknown whether this effect: (a) is stable over longer periods of time; (b) is independent of prospective stressful life events; and (c) differentially predicts first onsets or recurrences of MDEs.

A total of 270 older adolescents (mean age 17.06 years at cortisol measurement) from the larger prospective Northwestern-UCLA Youth Emotion Project completed baseline diagnostic and life stress interviews, questionnaires, and a 3-day cortisol sampling protocol measuring the CAR and diurnal rhythm, as well as up to four annual follow-up interviews of diagnoses and life stress.

Non-proportional person-month survival analyses revealed that higher levels of the baseline CAR significantly predict MDEs for 2.5 years following cortisol measurement. However, the strength of prediction of depressive episodes significantly decays over time, with the CAR no longer significantly predicting MDEs after 2.5 years. Elevations in the CAR did not significantly increase vulnerability to prospective major stressful life events. They did, however, predict MDE recurrences more strongly than first onsets.

These results suggest that a high CAR represents a time-limited risk factor for onsets of MDEs, which increases risk for depression independently of future major stressful life events. Possible explanations for the stronger effect of the CAR for predicting MDE recurrences than first onsets are discussed.

The psychosis-proneness–persistence–impairment model of psychotic disorder incorporates notions of both phenomenological and temporal continuity (persistence) of psychotic experiences (PE), but not structural continuity. Specific testable propositions of phenomenological continuity and persistence are identified.

Propositions are tested by systematic reviews of the epidemiology of PE, persistence of PE and disorder outcomes, and meta-analyses (including Monte Carlo permutation sampling, MCPS) of reported rates and odds ratios (ORs).

Estimates of the incidence and prevalence of PE obtained from 61 cohorts revealed a median annual incidence of 2.5% and a prevalence of 7.2%. Meta-analysis of risk factors identified age, minority or migrant status, income, education, employment, marital status, alcohol use, cannabis use, stress, urbanicity and family history of mental illness as important predictors of PE. The mode of assessment accounted for significant variance in the observed rates. Across cohorts, the probability of persistence was very strongly related to the rate of PE at baseline. Of those who report PE, ∼20% go on to experience persistent PE whereas for ∼80%, PE remit over time. Of those with baseline PE, 7.4% develop a psychotic disorder outcome.

Compelling support is found for the phenomenological and temporal continuity between PE and psychotic disorder and for the fundamental proposition that this relationship is probabilistic. However, imprecision in epidemiological research design, measurement limitations and the epiphenomenological nature of PE invite further robust scrutiny of the continuity theory.

Childhood adverse experiences are known to induce persistent changes in the hypothalamic–pituitary–adrenal (HPA) axis reactivity to stress. However, the mechanisms by which these experiences shape the neuroendocrine response to stress remain unclear.

We tested whether bullying victimization influenced serotonin transporter gene (SERT) DNA methylation using a discordant monozygotic (MZ) twin design. A subsample of 28 MZ twin pairs discordant for bullying victimization, with data on cortisol and DNA methylation, were identified in the Environmental Risk (E-Risk) Longitudinal Twin Study, a nationally representative 1994–1995 cohort of families with twins.

Bullied twins had higher SERT DNA methylation at the age of 10 years compared with their non-bullied MZ co-twins. This group difference cannot be attributed to the children's genetic makeup or their shared familial environments because of the study design. Bullied twins also showed increasing methylation levels between the age of 5 years, prior to bullying victimization, and the age of 10 years whereas no such increase was detected in non-bullied twins across time. Moreover, children with higher SERT methylation levels had blunted cortisol responses to stress.

Our study extends findings drawn from animal models, supports the hypothesis that early-life stress modifies DNA methylation at a specific cytosine–phosphate–guanine (CpG) site in the SERT promoter and HPA functioning and suggests that these two systems may be functionally associated.

While impaired memory and altered cortisol secretion are characteristic features of major depression, much less is known regarding the impact of antidepressant medication. We examined whether the cortisol awakening response (CAR) is increased in depressed patients with and without medication compared with healthy controls (HC) and whether CAR is associated with memory function in each group.

We examined 21 patients with major depression without medication, 20 depressed patients on antidepressant treatment, and 41 age-, sex- and education-matched healthy subjects. We tested verbal (Auditory Verbal Learning Task) and visuospatial (Rey figure) memory and measured CAR on two consecutive days.

Patient groups did not differ in severity of depression. We found a significant effect of group (p = 0.03) for CAR. Unmedicated patients exhibited a greater CAR compared with medicated patients (p = 0.04) with no differences between patient groups and HC. We found a significant effect of group for verbal (p = 0.03) and non-verbal memory (p = 0.04). Unmedicated patients performed worse compared with medicated patients and HC in both memory domains. Medicated patients and HC did not differ. Regression analyses revealed a negative association between CAR and memory function in depressed patients, but not in HC.

While in unmedicated depressed patients the magnitude of CAR is associated with impaired memory, medicated patients showed a smaller CAR and unimpaired cognitive function compared with HC. Our findings are compatible with the idea that antidepressants reduce CAR and partially restore memory function even if depressive psychopathology is still present.

The goal of this study was to explore the notion that anomalies of self-experience (ASE) are a core, ‘not-yet-psychotic’ clinical phenotype of emerging schizophrenia and its spectrum.

To accomplish this goal, we examined the relationship between ASE and commonly accepted risk markers in a sample of 87 help-seeking, non-psychotic adolescents (aged 14–18 years). ASE were assessed with the Examination of Anomalous Self-Experience (EASE), subclinical psychotic symptoms were assessed with the Prodromal Questionnaire and the Structured Interview for Prodromal Syndromes, deterioration in psychosocial functioning was assessed with the Social and Role Functioning Scales, and level of distress with the Mood and Anxiety Symptoms Questionnaire.

About 82 participants completed the entire EASE interview. The number of participants who reported ASE at a clinically meaningful level (n = 18, 22%) was smaller than that who met diagnostic criteria for a prodromal syndrome (n = 28, 34%). The degree of overlap between the two conditions was moderate but statistically significant (χ2(1) = 7.01, p = 0.008). An exploratory factor analysis revealed that ASE load on a different factor than prodromal symptoms and deterioration in functioning, but that there is a moderate correlation between the three factors.

These results suggest that ASE are prevalent among non-psychotic help-seeking adolescents, yet at a considerably lower rate than prodromal symptoms. In addition, they suggest that ASE and prodromal symptoms constitute distinct but moderately related dimensions of potential risk. Taken together, they provide preliminary support for the clinical usefulness of supplementing and refining the methods of early detection of risk with assessment of ASE.

Alterations in reward processing may represent an early vulnerability factor for the development of depressive disorder. Depression in adults is associated with reward hyposensitivity and diminished reward seeking may also be a feature of depression in children and adolescents. We examined the role of reward responding in predicting depressive symptoms, functional impairment and new-onset depressive disorder over time in the adolescent offspring of depressed parents. In addition, we examined group differences in reward responding between currently depressed adolescents, psychiatric and healthy controls, and also cross-sectional associations between reward responding and measures of positive social/environmental functioning.

We conducted a 1-year longitudinal study of adolescents at familial risk for depression (n = 197; age range 10–18 years). Reward responding and self-reported social/environmental functioning were assessed at baseline. Clinical interviews determined diagnostic status at baseline and at follow-up. Reports of depressive symptoms and functional impairment were also obtained.

Low reward seeking predicted depressive symptoms and new-onset depressive disorder at the 1-year follow-up in individuals free from depressive disorder at baseline, independently of baseline depressive symptoms. Reduced reward seeking also predicted functional impairment. Adolescents with current depressive disorder were less reward seeking (i.e. bet less at favourable odds) than adolescents free from psychopathology and those with externalizing disorders. Reward seeking showed positive associations with social and environmental functioning (extra-curricular activities, humour, friendships) and was negatively associated with anhedonia. There were no group differences in impulsivity, decision making or psychomotor slowing.

Reward seeking predicts depression severity and onset in adolescents at elevated risk of depression. Adaptive reward responses may be amenable to change through modification of existing preventive psychological interventions.

Long-term physical conditions (LTCs) consume the largest share of healthcare budgets. Although common mental disorders (CMDs) and LTCs often co-occur, the potential impact of improved mental health treatment on severe disability and hospital admissions for physical health problems remains unknown.

A cross-sectional study of 7403 adults aged 16–95 years living in private households in England was performed. LTCs were ascertained by prompted self-report. CMDs were ascertained by structured clinical interview. Disability was assessed using questions about problems with activities of daily living. Population impact and potential preventive gain were estimated using population-attributable fraction (PAF), and conservative estimates were obtained using ‘treated non-cases’ as the reference group.

Of the respondents, 20.7% reported at least one LTC. The prevalence of CMDs increased with the number of LTCs, but over two-thirds (71.2%) of CMD cases in people with LTCs were untreated. Statistically significant PAFs were found for CMDs and recent hospital admission [13.5%, 95% confidence intervals (CI) 6.6–20.0] and severe disability (31.3%, 95% CI 27.1–35.2) after adjusting for LTCs and other confounders. Only the latter remained significant when using the most conservative estimate of PAF (21.8%, 95% CI 14.0–28.9), and this was reduced only slightly when considering only participants with LTCs (18.5%, 95% CI 7.9–27.9).

Better treatments for CMDs in people with LTCs could achieve almost the same population health gain in terms of reducing severe disability as those targeted at the entire population. Interventions to reduce the prevalence of CMDs among people with LTCs should be part of routine medical care.

Increased sensitivity and exposure to stress are associated with psychotic symptoms in schizophrenia and its risk states, but little is known about the co-evolution of stress sensitivity and exposure with positive and other symptoms in a clinical high-risk (CHR) cohort.

A combined cross-sectional and longitudinal design was used to examine the associations over time of stress sensitivity and exposure (i.e. life events) with ‘prodromal’ symptoms in a cohort of 65 CHR patients assessed quarterly for up to 4 years, and at baseline in 24 healthy controls similar in age and gender.

Impaired stress tolerance was greater in patients, in whom it was associated over time with positive and negative symptoms, in addition to depression, anxiety and poor function. By contrast, life events were comparable in patients and controls, and bore no association with symptoms. In this treated cohort, there was a trajectory of improvement in stress tolerance, symptoms and function over time.

Impaired stress tolerance was associated with a wide range of ‘prodromal’ symptoms, consistent with it being a core feature of the psychosis risk state. Self-reported life events were not relevant as a correlate of clinical status. As in other treated CHR cohorts, most patients improved over time across symptom domains.

There are no evidence-based treatments for severe and enduring anorexia nervosa (SE-AN). This study evaluated the relative efficacy of cognitive behavioral therapy (CBT-AN) and specialist supportive clinical management (SSCM) for adults with SE-AN.

Sixty-three participants with a diagnosis of AN, who had at least a 7-year illness history, were treated in a multi-site randomized controlled trial (RCT). During 30 out-patient visits spread over 8 months, they received either CBT-AN or SSCM, both modified for SE-AN. Participants were assessed at baseline, end of treatment (EOT), and at 6- and 12-month post-treatment follow-ups. The main outcome measures were quality of life, mood disorder symptoms and social adjustment. Weight, eating disorder (ED) psychopathology, motivation for change and health-care burden were secondary outcomes.

Thirty-one participants were randomized to CBT-AN and 32 to SSCM with a retention rate of 85% achieved at the end of the study. At EOT and follow-up, both groups showed significant improvement. There were no differences between treatment groups at EOT. At the 6-month follow-up, CBT-AN participants had higher scores on the Weissman Social Adjustment Scale (WSAS; p = 0.038) and at 12 months they had lower Eating Disorder Examination (EDE) global scores (p = 0.004) and higher readiness for recovery (p = 0.013) compared to SSCM.

Patients with SE-AN can make meaningful improvements with both therapies. Both treatments were acceptable and high retention rates at follow-up were achieved. Between-group differences at follow-up were consistent with the nature of the treatments given.

Clinical trials are typically designed to test the effect of a specific treatment on a single diagnostic entity. However, because common internalizing disorders are highly correlated (‘co-morbid’), we sought to establish a practical and parsimonious method to characterize and quantify changes in a broad spectrum of internalizing psychopathology targeted for treatment in a clinical trial contrasting two transdiagnostic psychosocial interventions.

Alcohol dependence treatment patients who had any of several common internalizing disorders were randomized to a six-session cognitive-behavioral therapy (CBT) experimental treatment condition or a progressive muscle relaxation training (PMRT) comparison treatment condition. Internalizing psychopathology was characterized at baseline and 4 months following treatment in terms of the latent structure of six distinct internalizing symptom domain surveys.

Exploratory structural equation modeling (ESEM) identified a two-factor solution at both baseline and the 4-month follow-up: Distress (measures of depression, trait anxiety and worry) and Fear (measures of panic anxiety, social anxiety and agoraphobia). Although confirmatory factor analysis (CFA) demonstrated measurement invariance between the time-points, structural models showed that the latent means of Fear and Distress decreased substantially from baseline to follow-up for both groups, with a small but statistically significant advantage for the CBT group in terms of Distress (but not Fear) reduction.

The approach demonstrated in this study provides a practical solution to modeling co-morbidity in a clinical trial and is consistent with converging evidence pointing to the dimensional structure of internalizing psychopathology.

Reduced hippocampal size and increased stress sensitivity are associated with psychotic disorder and familial risk for psychosis. However, to what degree the hippocampus is implicated in daily life stress reactivity has not yet been examined. The current study investigated (i) whether familial risk (the contrast between controls, patients and siblings of patients) moderated the relationship between hippocampal volume (HV) and emotional daily stress reactivity and (ii) whether familial risk (the contrast between controls and siblings of patients) moderated the relationship between HV and cortisol daily stress reactivity.

T1-weighted magnetic resonance imaging (MRI) scans were acquired from 20 patients with schizophrenia, 37 healthy siblings with familial risk for schizophrenia and 32 controls. Freesurfer 5.0.0 was used to measure HV. The experience sampling method (ESM), a structured momentary assessment technique, was used to assess emotional stress reactivity, that is the effect of momentary stress on momentary negative affect (NA). In addition, in the control and sibling groups, cortisol stress reactivity was assessed using momentary cortisol levels extracted from saliva.

Multilevel linear regression analyses revealed a significant three-way interaction between group, HV and momentary stress in both the model of NA and the model of cortisol. Increased emotional stress reactivity was associated with smaller left HV in patients and larger total HV in controls. In line with the results in patients, siblings with small HV demonstrated increased emotional and cortisol stress reactivity compared to those with large HV.

HV may index risk and possibly disease-related mechanisms underlying daily life stress reactivity in psychotic disorder.

There is evidence that patients with schizophrenia benefit from standard cognitive behaviour therapy (CBT) only if active techniques are used (‘full therapy’). By contrast, attending sessions but not proceeding beyond engagement and assessment strategies (‘partial therapy’), or simply not attending sessions (‘no therapy’), is not associated with better outcomes. The factors leading to full therapy are unknown. We hypothesized that patients' initial ideas about the nature and extent of their problems would predict use of CBT. A match between patients' views of their problems and the principles underlying treatment would lead to better outcomes.

Ninety-two patients with a recent relapse of psychosis completed the Illness Perception Questionnaire (IPQ) before receiving CBT. We examined whether their illness perceptions predicted the take-up of therapy.

Patients who did not attend sessions believed their problems would not last as long as those who attended them. Those who attended sessions but did not proceed to full therapy had a lower sense of control over their problems and a more biological view of their causes. Patients who took up full therapy were more likely to attribute the cause of their problems to their personality and state of mind. The take-up of therapy was predicted neither by levels of psychiatric symptoms nor by insight.

People with psychosis who have psychologically orientated views of their problems, including the potential to gain control over them, may be more likely to engage fully and do well with standard CBT for psychosis, irrespective of the severity of their problems.

Cigarette smoking is strongly associated with mental illness but the causal direction of the association is uncertain. We investigated the causal relationship between smoking and symptoms of anxiety and depression in the Norwegian HUNT study using the rs1051730 single nucleotide polymorphism (SNP) variant located in the nicotine acetylcholine receptor gene cluster on chromosome 15 as an instrumental variable for smoking phenotypes. Among smokers, this SNP is robustly associated with smoking quantity and nicotine dependence.

In total, 53 601 participants were genotyped for the rs1051730 SNP and provided information on smoking habits and symptoms of anxiety and depression using the Hospital Anxiety and Depression Scale (HADS).

Self-reported smoking was positively associated with the prevalence of both anxiety and depression, and the measured polymorphism was positively associated with being a current smoker and the number of cigarettes smoked in current smokers. In the sample as a whole, risk of anxiety increased with each affected T allele [odds ratio (OR) 1.06, 95% confidence interval (CI) 1.02–1.09, p = 0.002] but there was no association with depression (p = 0.31). However, we found no clear association of the polymorphism with either anxiety (OR 1.03, 95% CI 0.97–1.09, p = 0.34) or depression (OR 1.02, 95% CI 0.95–1.09, p = 0.62) among smokers.

As there was no association of the smoking-related rs1051730 SNP with anxiety and depression among smokers, the results suggest that smoking is not a cause of anxiety and depression.