Parents are a major supplier of alcohol to adolescents, yet there is limited research examining the impact of this on adolescent alcohol use. This study investigates associations between parental supply of alcohol, supply from other sources, and adolescent drinking, adjusting for child, parent, family and peer variables.

A cohort of 1927 adolescents was surveyed annually from 2010 to 2014. Measures include: consumption of whole drinks; binge drinking (>4 standard drinks on any occasion); parental supply of alcohol; supply from other sources; child, parent, family and peer covariates.

After adjustment, adolescents supplied alcohol by parents had higher odds of drinking whole beverages [odds ratio (OR) 1.80, 95% confidence interval (CI) 1.33–2.45] than those not supplied by parents. However, parental supply was not associated with bingeing, and those supplied alcohol by parents typically consumed fewer drinks per occasion (incidence rate ratio 0.86, 95% CI 0.77–0.96) than adolescents supplied only from other sources. Adolescents obtaining alcohol from non-parental sources had increased odds of drinking whole beverages (OR 2.53, 95% CI 1.86–3.45) and bingeing (OR 3.51, 95% CI 2.53–4.87).

Parental supply of alcohol to adolescents was associated with increased risk of drinking, but not bingeing. These parentally-supplied children also consumed fewer drinks on a typical drinking occasion. Adolescents supplied alcohol from non-parental sources had greater odds of drinking and bingeing. Further follow-up is necessary to determine whether these patterns continue, and to examine alcohol-related harm trajectories. Parents should be advised that supply of alcohol may increase children's drinking.

Excessive alcohol use is associated with brain damage but less is known about brain effects from moderate alcohol use. Previous findings indicate that patients with severe mental illness, particularly schizophrenia, are vulnerable to alcohol-related brain damage. We investigated the association between levels of alcohol consumption and cortical and subcortical brain structures in schizophrenia and bipolar disorder patients and healthy controls, and investigated for group differences for this association.

1.5 T structural magnetic resonance images were acquired of 609 alcohol-using participants (165 schizophrenia patients, 172 bipolar disorder patients, 272 healthy controls), mean (s.d.) age 34.2 (9.9) years, 52% men. Past year alcohol use was assessed with the Alcohol Use Disorder Identification Test – Consumption part (AUDIT-C). General linear models were used to investigate associations between AUDIT-C score and cortical thickness, surface area, and total brain and subcortical volumes.

Increasing AUDIT-C score was linearly associated with thinner cortex in medial and dorsolateral frontal and parieto-occipital regions, and with larger left lateral ventricle volume. There was no significant interaction between AUDIT-C score and diagnostic group. The findings remained significant after controlling for substance use disorders, antipsychotic medication and illness severity.

The results show a dose-dependent relationship between alcohol use and thinner cortex and ventricular expansion. The findings are present also at lower levels of alcohol consumption and do not differ between schizophrenia or bipolar disorder patients compared to healthy controls. Our results do not support previous findings of increased vulnerability for alcohol-related brain damage in severe mental illness.

A number of studies reports reduced hippocampal volume in individuals who engage in problematic alcohol use. However, the magnitude of the difference in hippocampal volume between individuals with v. without problematic alcohol use has varied widely, and there have been null findings. Moreover, the studies comprise diverse alcohol use constructs and samples, including clinically significant alcohol use disorders and subclinical but problematic alcohol use (e.g. binge drinking), adults and adolescents, and males and females.

We conducted the first quantitative synthesis of the published empirical research on associations between problematic alcohol use and hippocampal volume. In total, 23 studies were identified and selected for inclusion in the meta-analysis; effects sizes were aggregated using a random-effects model.

Problematic alcohol use was associated with significantly smaller hippocampal volume (d = −0.53). Moderator analyses indicated that effects were stronger for clinically significant v. subclinical alcohol use and among adults relative to adolescents; effects did not differ among males and females.

Problematic alcohol use is associated with reduced hippocampal volume. The moderate overall effect size suggests the need for larger samples than are typically included in studies of alcohol use and hippocampal volume. Because the existing literature is almost entirely cross-sectional, future research using causally informative study designs is needed to determine whether this association reflects premorbid risk for the development of problematic alcohol use and/or whether alcohol has a neurotoxic effect on the hippocampus.

Internet-based cognitive–behavioural treatment (ICBT) for anxiety disorders has shown some promise, but no study has yet examined unguided ICBT in primary care. This randomized controlled trial (RCT) investigated whether a transdiagnostic, unguided ICBT programme for anxiety disorders is effective in primary care settings, after a face-to-face consultation with a physician (MD). We hypothesized that care as usual (CAU) plus unguided ICBT would be superior to CAU in reducing anxiety and related symptoms among patients with social anxiety disorder (SAD), panic disorder with or without agoraphobia (PDA) and/or generalized anxiety disorder (GAD).

Adults (n = 139) with at least one of these anxiety disorders, as reported by their MD and confirmed by a structured diagnostic interview, were randomized. Unguided ICBT was provided by a novel transdiagnostic ICBT programme (‘velibra’). Primary outcomes were generic measures, such as anxiety and depression symptom severity, and diagnostic status at post-treatment (9 weeks). Secondary outcomes included anxiety disorder-specific measures, quality of life, treatment adherence, satisfaction, and general psychiatric symptomatology at follow-up (6 months after randomization).

CAU plus unguided ICBT was more effective than CAU at post-treatment, with small to medium between-group effect sizes on primary (Cohen's d = 0.41–0.47) and secondary (Cohen's d = 0.16–0.61) outcomes. Treatment gains were maintained at follow-up. In the treatment group, 28.2% of those with a SAD diagnosis, 38.3% with a PDA diagnosis, and 44.8% with a GAD diagnosis at pretreatment no longer fulfilled diagnostic criteria at post-treatment.

The unguided ICBT intervention examined is effective for anxiety disorders when delivered in primary care.

Attention deficit hyperactivity disorder (ADHD) patients have been reported to display deficits in action control processes. While it is known that subliminally and consciously induced conflicts interact and conjointly modulate action control in healthy subjects, this has never been investigated for ADHD.

We investigated the (potential) interaction of subliminally and consciously triggered response conflicts in children with ADHD and matched healthy controls using neuropsychological methods (event-related potentials; ERPs) to identify the involved cognitive sub-processes.

Unlike healthy controls, ADHD patients showed no interaction of subliminally and consciously triggered response conflicts. Instead, they only showed additive effects as their behavioural performance (accuracy) was equally impaired by each conflict and they showed no signs of task-goal shielding even in cases of low conflict load. Of note, this difference between ADHD and controls was not rooted in early bottom-up attentional stimulus processing as reflected by the P1 and N1 ERPs. Instead, ADHD showed either no or reversed modulations of conflict-related processes and response selection as reflected by the N2 and P3 ERPs.

There are fundamental differences in the architecture of cognitive control which might be of use for future diagnostic procedures. Unlike healthy controls, ADHD patients do not seem to be endowed with a threshold which allows them to maintain high behavioural performance in the face of low conflict load. ADHD patients seem to lack sufficient top-down attentional resources to maintain correct response selection in the face of conflicts by shielding the response selection process from response tendencies evoked by any kind of distractor.

Obsessive–compulsive disorder (OCD) has been linked to functional abnormalities in fronto-striatal networks as well as impairments in decision making and learning. Little is known about the neurocognitive mechanisms causing these decision-making and learning deficits in OCD, and how they relate to dysfunction in fronto-striatal networks.

We investigated neural mechanisms of decision making in OCD patients, including early and late onset of disorder, in terms of reward prediction errors (RPEs) using functional magnetic resonance imaging. RPEs index a mismatch between expected and received outcomes, encoded by the dopaminergic system, and are known to drive learning and decision making in humans and animals. We used reinforcement learning models and RPE signals to infer the learning mechanisms and to compare behavioural parameters and neural RPE responses of the OCD patients with those of healthy matched controls.

Patients with OCD showed significantly increased RPE responses in the anterior cingulate cortex (ACC) and the putamen compared with controls. OCD patients also had a significantly lower perseveration parameter than controls.

Enhanced RPE signals in the ACC and putamen extend previous findings of fronto-striatal deficits in OCD. These abnormally strong RPEs suggest a hyper-responsive learning network in patients with OCD, which might explain their indecisiveness and intolerance of uncertainty.

Attention-deficit/hyperactivity disorder (ADHD) is characterized by problems in regulating attention and in suppressing disruptive motor activity, i.e. hyperactivity and impulsivity. We recently found evidence that aberrant distribution of posterior α band oscillations (8–12 Hz) is associated with attentional problems in ADHD. The sensorimotor cortex also produces strong 8–12 Hz band oscillations, namely the μ rhythm, and is thought to have a similar inhibitory function. Here, we now investigate whether problems in distributing α band oscillations in ADHD generalize to the μ rhythm in the sensorimotor domain.

In a group of adult ADHD (n = 17) and healthy control subjects (n = 18; aged 21–40 years) oscillatory brain activity was recorded using magnetoencephalography during a visuo-spatial attention task. Subjects had to anticipate a target with unpredictable timing and respond by pressing a button.

Preparing a motor response, the ADHD group failed to increase hemispheric μ lateralization with relatively higher μ power in sensorimotor regions not engaged in the task, as the controls did (F1,33 = 8.70, p = 0.006). Moreover, the ADHD group pre-response μ lateralization not only correlated positively with accuracy (rs = 0.64, p = 0.0052) and negatively with intra-individual reaction time variability (rs = −0.52, p = 0.033), but it also correlated negatively with the score on an ADHD rating scale (rs = −0.53, p = 0.028).

We suggest that ADHD is associated with an inability to sufficiently inhibit task-irrelevant sensorimotor areas by means of modulating μ oscillatory activity. This could explain disruptive motor activity in ADHD. These results provide further evidence that impaired modulation of α band oscillations is involved in the pathogenesis of ADHD.

Many studies have concluded that cannabis use disorder (CUD) negatively influences outcomes in first-episode psychosis (FEP). However, few have taken into account the impact of concurrent misuse of other substances.

This 2-year, prospective, longitudinal study of FEP patients, aged between 18 and 30 years, admitted to early intervention programs in Montreal, Quebec, Canada, examined the specific influence of different substance use disorders (SUD) (alcohol, cannabis, cocaine, amphetamines) on service utilization, symptomatic and functional outcomes in FEP.

Drugs and alcohol were associated with lower functioning, but drugs had a greater negative impact on most measures at 2-year follow-up. Half of CUD patients and more than 65% of cocaine or amphetamine abusers presented polysubstance use disorder (poly-SUD). The only group that deteriorated from years 1 to 2 (symptoms and functioning) were patients with persistent CUD alone. Outcome was worse in CUD than in the no-SUD group at 2 years. Cocaine, amphetamines and poly-SUD were associated with worse symptomatic and functional outcomes from the 1st year of treatment, persisting over time with higher service utilization (hospitalization).

The negative impact attributed to CUD in previous studies could be partly attributed to methodological flaws, like including polysubstance abusers among cannabis misusers. However, our investigation confirmed the negative effect of CUD on outcome. Attention should be paid to persistent cannabis misusers, since their condition seems to worsen over time, and to cocaine and amphetamine misusers, in view of their poorer outcome early during follow-up and high service utilization.

This study investigated the potential environmental effects of peer victimization and the quality of relationships with parents and friends on diurnal cortisol secretion in mid-adolescence.

This study used the monozygotic (MZ) twin-difference design to control for genetic effects and thus estimate the unique environmental influences on diurnal cortisol. Participants were 136 MZ twin pairs (74 female pairs) for whom cortisol was assessed four times per day over four collection days grouped in a 2-week period in grade 8 (mean age = 14.07 years). Participants also provided self-reports of peer victimization from grade 4 to grade 8 and of the relationship quality with the mother, father and best friend in grade 8.

The expected pattern of diurnal cortisol secretion was observed, with high levels at awakening followed by an increase 30 min later and a progressive decrease subsequently. Controlling for a host of confounders, only within-twin pair differences in peer victimization and a problematic relationship with the mother were significantly linked to twin differences in diurnal cortisol secretion. Specifically, whereas a more problematic mother–child relationship was associated with morning cortisol secretion, peer victimization was linked to cortisol secretion later in the day (diurnal slope).

Controlling for genetic influences and other confounders, stressful relationships with peers and the mother exert unique and time-specific environmental influences on the pattern of diurnal cortisol secretion in mid-adolescence.

Cognitive deficits are predictors of functional outcome in patients with psychosis. While conventional antipsychotics are relatively effective on positive symptoms, their impact on negative and cognitive symptoms is limited. Recent studies have established a link between oxidative stress and neurocognitive deficits in psychosis. N-acetylcysteine (NAC), a glutathione precursor with glutamatergic properties, has shown efficacy on negative symptoms and functioning in patients with schizophrenia and bipolar disorder, respectively. However, there are few evidence-based approaches for managing cognitive impairment in psychosis. The present study aims to examine the cognitive effects of adjunctive NAC treatment in a pooled subgroup of participants with psychosis who completed neuropsychological assessment in two trials of both schizophrenia and bipolar disorder.

A sample of 58 participants were randomized in a double fashion to receive 2 g/day of NAC (n = 27) or placebo (n = 31) for 24 weeks. Attention, working memory and executive function domains were assessed. Differences between cognitive performance at baseline and end point were examined using Wilcoxon's test. The Mann–Whitney test was used to examine the differences between the NAC and placebo groups at the end point.

Participants treated with NAC had significantly higher working memory performance at week 24 compared with placebo (U = 98.5, p = 0.027).

NAC may have an impact on cognitive performance in psychosis, as a significant improvement in working memory was observed in the NAC-treated group compared with placebo; however, these preliminary data require replication. Glutamatergic compounds such as NAC may constitute a step towards the development of useful therapies for cognitive impairment in psychosis.

Studies from high-income countries report moderate-to-strong positive associations between alcohol use disorder (AUD) and other mental disorders, but there is little evidence about the comorbidity of AUD from low-and-middle-income countries.

A sample of 74 752 adults from five provinces that account for >12% of China's adult population was screened using the General Health Questionnaire, and the Structured Clinical Interview for DSM-IV was administered by psychiatrists to a subsample of 9619 males. The associations between AUD and other mental disorders at each site and the characteristics of men with AUD with and without comorbid mental disorders were estimated using logistic regression and summarized across sites using meta-analysis. Generalized estimation equations estimated the associations between the clinical features of alcohol dependence and comorbidity.

Robust inverse associations were found between current AUD and any mood disorder (adjusted OR = 0.6, 95% CI = 0.4–0.8) and any anxiety disorder (OR = 0.5, 95% CI = 0.3–1.0). Compared with men without AUD, men with AUD without comorbid disorders were more likely to be middle-aged, to be currently married, and to have higher family incomes. Men with comorbid AUD and other disorders were more likely to have the clinical features of alcohol dependence than men with AUD without comorbid disorders.

Inverse associations between AUD and other mental disorders and the higher social status of men with AUD than men without AUD found in this large, representative sample of community-dwelling Chinese males highlight the importance of considering the local substance-use culture when designing clinical or preventive interventions for addictive conditions.

Previous studies hypothesized that neurodevelopmental risk factors may play a role in the pathogenesis of obsessive-compulsive disorder (OCD). Cortical folding has been shown to be a reliable indicator for normal and altered neurodevelopment, but in OCD it has barely been investigated up to now. The present study investigates whether alterations in gyrification are detectable in OCD and, if so, how these are associated with clinical characteristics.

We compared the local Gyrification Index (lGI) between 75 OCD patients and 75 matched healthy subjects across the whole brain. In addition, for those regions exhibiting an altered lGI in patients we explored a potential relationship to symptom severity, age of onset, and influence of medication.

OCD patients had a significantly decreased lGI in right parietal, precentral but also insula, temporal, pars triangularis and rostral middle frontal regions compared to healthy subjects. A positive association with age of onset was found but no association with symptom severity. There was no effect of co-morbidity or medication.

The reduced gyrification found in OCD confirms previous findings in other psychiatric disorders and suggests that alterations may already occur during early stages of brain development. Our findings support the idea that altered cortical folding might represent a trait characteristic of the disorder although longitudinal studies are needed to clarify the trajectory of this morphological measure in OCD.

Childhood trauma increases risk of a range of mental disorders including psychosis. Whereas the mechanisms are unclear, previous evidence has implicated atypical processing of emotions among the core cognitive models, in particular suggesting altered attentional allocation towards negative stimuli and increased negativity bias. Here, we tested the association between childhood trauma and brain activation during emotional face processing in patients diagnosed with psychosis continuum disorders. In particular, we tested if childhood trauma was associated with the differentiation in brain responses between negative and positive face stimuli. We also tested if trauma was associated with emotional ratings of negative and positive faces.

We included 101 patients with a Diagnostic and Statistical Manual of Mental Disorders (DSM) schizophrenia spectrum or bipolar spectrum diagnosis. History of childhood trauma was obtained using the Childhood Trauma Questionnaire. Brain activation was measured with functional magnetic resonance imaging during presentation of faces with negative or positive emotional expressions. After the scanner session, patients performed emotional ratings of the same faces.

Higher levels of total childhood trauma were associated with stronger differentiation in brain responses to negative compared with positive faces in clusters comprising the right angular gyrus, supramarginal gyrus, middle temporal gyrus and the lateral occipital cortex (Cohen's d = 0.72–0.77). In patients with schizophrenia, childhood trauma was associated with reporting negative faces as more negative, and positive faces as less positive (Cohen's d > 0.8).

Along with the observed negativity bias in the assessment of emotional valence of faces, our data suggest stronger differentiation in brain responses between negative and positive faces with higher levels of trauma.

A previous small study suggested that Brain Network Activation (BNA), a novel ERP-based brain network analysis, may have diagnostic utility in attention deficit hyperactivity disorder (ADHD). In this study we examined the diagnostic capability of a new advanced version of the BNA methodology on a larger population of adults with and without ADHD.

Subjects were unmedicated right-handed 18- to 55-year-old adults of both sexes with and without a DSM-IV diagnosis of ADHD. We collected EEG while the subjects were performing a response inhibition task (Go/NoGo) and then applied a spatio-temporal Brain Network Activation (BNA) analysis of the EEG data. This analysis produced a display of qualitative measures of brain states (BNA scores) providing information on cortical connectivity. This complex set of scores was then fed into a machine learning algorithm.

The BNA analysis of the EEG data recorded during the Go/NoGo task demonstrated a high discriminative capacity between ADHD patients and controls (AUC = 0.92, specificity = 0.95, sensitivity = 0.86 for the Go condition; AUC = 0.84, specificity = 0.91, sensitivity = 0.76 for the NoGo condition).

BNA methodology can help differentiate between ADHD and healthy controls based on functional brain connectivity. The data support the utility of the tool to augment clinical examinations by objective evaluation of electrophysiological changes associated with ADHD. Results also support a network-based approach to the study of ADHD.

Obsessive–compulsive disorder (OCD) patients typically overmonitor their own behavior, as shown by symptoms of excessive doubt and checking. Although this is well established for the patients’ relationship with external stimuli in the environment, no study has explored their monitoring of internal body signals, a process known to be affected in anxiety-related syndromes. Here, we explored this issue through a cardiac interoception task that measures sensing of heartbeats. Our aim was to explore key behavioral and electrophysiological aspects of internal-cue monitoring in OCD, while examining their potential distinctiveness in this condition.

We administered a heartbeat detection (HBD) task (with related interoceptive confidence and awareness measures) to three matched groups (OCD patients, panic disorder patients, healthy controls) and recorded ongoing modulations of two task-relevant electrophysiological markers: the heart evoked potential (HEP) and the motor potential (MP).

Behaviorally, OCD patients outperformed controls and panic patients in the HBD task. Moreover, they exhibited greater amplitude modulation of both the HEP and the MP during cardiac interoception. However, they evinced poorer confidence and awareness of their interoceptive skills.

Convergent behavioral and electrophysiological data showed that overactive monitoring in OCD extends to the sensing of internal bodily signals. Moreover, this pattern discriminated OCD from panic patients, suggesting a condition-distinctive alteration. Our results highlight the potential of exploring interoceptive processes in the OCD spectrum to better characterize the population's cognitive profile. Finally, these findings may lay new bridges between somatic theories of emotion and cognitive models of OCD.

Autism Spectrum Disorders (ASD) and psychosis share deficits in social cognition. The insular region has been associated with awareness of self and reality, which may be basic for proper social interactions.

Total and regional insular volume and thickness measurements were obtained from a sample of 30 children and adolescents with ASD, 29 with early onset first-episode psychosis (FEP), and 26 healthy controls (HC). Total, regional, and voxel-level volume and thickness measurements were compared between groups (with correction for multiple comparisons), and the relationship between these measurements and symptom severity was explored.

Compared with HC, a shared volume deficit was observed for the right (but not the left) anterior insula (ASD: p = 0.007, FEP: p = 0.032), and for the bilateral posterior insula: (left, ASD: p = 0.011, FEP: p = 0.033; right, ASD: p = 0.004, FEP: p = 0.028). A voxel-based morphometry (VBM) conjunction analysis showed that ASD and FEP patients shared a gray matter volume and thickness deficit in the left posterior insula. Within patients, right anterior (r = −0.28, p = 0.041) and left posterior (r = −0.29, p = 0.030) insular volumes negatively correlated with the severity of insight deficits, and left posterior insular volume negatively correlated with the severity of ‘autistic-like’ symptoms (r = −0.30, p = 0.028).

The shared reduced volume and thickness in the anterior and posterior regions of the insula in ASD and FEP provides the first tentative evidence that these conditions share structural pathology that may be linked to shared symptomatology.

Few studies have evaluated the development in the use of antipsychotic medication and psychotic symptoms in patients with first-episode psychosis on a long-term basis. Our objective was to investigate how psychotic symptoms and the use of antipsychotic medication changed over a 10-year period in a cohort of patients with first-episode psychosis.

The study is a longitudinal prospective cohort study over 10 years with follow-ups at years 1, 2, 5 and 10. A total of 496 patients with first-episode psychosis were included in a multi-centre study initiated between 1998 and 2000 in Copenhagen and Aarhus, Denmark.

At all follow-ups, a large proportion (20–30%) of patients had remission of psychotic symptoms without use of antipsychotic medication at the time of the follow-up. Patients who were in this group at the 5-year follow-up had an 87% [95% confidence interval (CI) 77–96%] chance of being in the same group at the 10-year follow-up. This stability was also the case for patients who had psychotic symptoms and were treated with antipsychotic medication at year 5, where there was a 67% (95% CI 56–78%) probability of being in this group at the consecutive follow-up.

A large group of patients with psychotic illness were in remission without the use of antipsychotic medication, peaking at year 10. Overall there was a large degree of stability in disease courses over the 10-year period. These results suggest that the long-term outcome of psychotic illness is heterogeneous and further investigation on a more individualized approach to long-term treatment is needed.

Altered hypothalamus-pituitary-adrenal (HPA) axis function and reduced hippocampal volume (HV) are established correlates of stress vulnerability. We have previously shown an attenuated cortisol awakening response (CAR) and associations with HV specifically in male first-episode psychosis patients. Findings in individuals at ultra-high risk (UHR) for psychosis regarding these neurobiological markers are inconsistent, and assessment of their interplay, accounting for sex differences, could explain incongruent results.

Study participants were 42 antipsychotic-naive UHR subjects (24 men) and 46 healthy community controls (23 men). Saliva samples for the assessment of CAR were collected at 0, 30 and 60 min after awakening. HV was determined from high-resolution structural magnetic resonance imaging scans using a semi-automatic segmentation protocol.

Cortisol measures and HV were not significantly different between UHR subjects and controls in total, but repeated-measures multivariate regression analyses revealed reduced cortisol levels 60 min after awakening and smaller left HV in male UHR individuals. In UHR participants only, smaller left and right HV was significantly correlated with a smaller total CAR (ρ = 0.42, p = 0.036 and ρ = 0.44, p = 0.029, respectively), corresponding to 18% and 19% of shared variance (medium effect size).

Our findings suggest that HV reduction in individuals at UHR for psychosis is specific to men and linked to reduced post-awakening cortisol concentrations. Abnormalities in the neuroendocrine circuitry modulating stress vulnerability specifically in male UHR subjects might explain increased psychosis risk and disadvantageous illness outcomes in men compared to women.

The aim of the study was to evaluate the comparative risk of self-harm associated with the use of different antidepressants.

A cohort study was conducted using data from Taiwan's National Health Insurance Research Database from 2001 to 2012. A total of 751 606 new antidepressant users with depressive disorders were included. The study outcome was hospitalization due to self-harm (International Classification of Diseases, Ninth Revision, Clinical Modification codes: E950–E958 and E980–E988). Cox proportional hazards models with stratification of the propensity score deciles were used to estimate the hazard ratios of self-harm hospitalization during the first year following the initiation of antidepressant treatment.

There were 1038 hospitalization episodes due to self-harm that occurred during the follow-up of 149 796 person-years, with an overall incidence rate of 6.9 [95% confidence interval (CI) 6.5–7.4] per 1000. Compared with fluoxetine, the risk of self-harm hospitalization was higher for maprotiline [adjusted hazard ratio (aHR) = 3.00, 95% CI 1.40–6.45], milnacipran (aHR = 2.34, 95% CI 1.24–4.43) and mirtazapine (aHR = 1.40, 95% CI 1.06–1.86), lower for bupropion (aHR = 0.51, 95% CI 0.30–0.86), and similar level of risk was found for other selective serotonin reuptake inhibitors (citalopram, escitalopram, fluvoxamine, paroxetine and sertraline).

The risk of self-harm may vary across different antidepressant drugs. It would be of importance to conduct further research to investigate the influence of antidepressant use on self-harm behaviors.

There is a paucity of valid, brief instruments for the assessment of lifetime major depressive disorder (MDD) that can be used in, for example, large-scale genomics, imaging or biomarker studies on depression. We developed the LIfetime Depression Assessment Self-report (LIDAS), which assesses lifetime MDD diagnosis according to DSM criteria, and is largely based on the widely used Composite International Diagnostic Interview (CIDI). Here, we tested the feasibility and determined the sensitivity and specificity for measuring lifetime MDD with this new questionnaire, with a regular CIDI as reference.

Sensitivity and specificity analyses of the online lifetime MDD questionnaire were performed in adults with (n = 177) and without (n = 87) lifetime MDD according to regular index CIDIs, selected from the Netherlands Study of Depression and Anxiety (NESDA) and Netherlands Twin Register (NTR). Feasibility was tested in an additional non-selective, population-based sample of NTR participants (n = 245).

Of the 753 invited persons, 509 (68%) completed the LIDAS, of which 419 (82%) did this online. User-friendliness of the instrument was rated high. Median completion time was 6.2 min. Sensitivity and specificity for lifetime MDD were 85% [95% confidence interval (CI) 80–91%] and 80% (95% CI 72–89%), respectively. This LIDAS instrument gave a lifetime MDD prevalence of 20.8% in the population-based sample.

Measuring lifetime MDD with an online instrument was feasible. Sensitivity and specificity were adequate. The instrument gave a prevalence of lifetime MDD in line with reported population prevalences. LIDAS is a promising tool for rapid determination of lifetime MDD status in large samples, such as needed for genomics studies.