Although interventions exist to reduce violent crime, optimal implementation requires accurate targeting. We report the results of an attempt to develop an actuarial model using machine learning methods to predict future violent crimes among US Army soldiers.

A consolidated administrative database for all 975 057 soldiers in the US Army in 2004–2009 was created in the Army Study to Assess Risk and Resilience in Servicemembers (Army STARRS). Of these soldiers, 5771 committed a first founded major physical violent crime (murder-manslaughter, kidnapping, aggravated arson, aggravated assault, robbery) over that time period. Temporally prior administrative records measuring socio-demographic, Army career, criminal justice, medical/pharmacy, and contextual variables were used to build an actuarial model for these crimes separately among men and women using machine learning methods (cross-validated stepwise regression, random forests, penalized regressions). The model was then validated in an independent 2011–2013 sample.

Key predictors were indicators of disadvantaged social/socioeconomic status, early career stage, prior crime, and mental disorder treatment. Area under the receiver-operating characteristic curve was 0.80–0.82 in 2004–2009 and 0.77 in the 2011–2013 validation sample. Of all administratively recorded crimes, 36.2–33.1% (male-female) were committed by the 5% of soldiers having the highest predicted risk in 2004–2009 and an even higher proportion (50.5%) in the 2011–2013 validation sample.

Although these results suggest that the models could be used to target soldiers at high risk of violent crime perpetration for preventive interventions, final implementation decisions would require further validation and weighing of predicted effectiveness against intervention costs and competing risks.

The ‘vascular depression’ hypothesis states that brain changes located in frontal-subcortical pathways increase vulnerability for specific depressive symptom profiles, but studies examining locations of small-vessel and degenerative changes with individual symptoms are scarce. We examined whether location and progression of white-matter lesions (WMLs), lacunar infarcts and atrophy were associated with motivational and mood symptoms in patients with symptomatic atherosclerotic disease.

In 578 patients [63 (s.d.=8) years] of the Second Manifestations of ARTerial disease (SMART)-Medea study, volumes of WMLs and atrophy and visually rated infarcts were obtained with 1.5 T magnetic resonance imaging at baseline and after 3.9 (s.d.=0.4) years' follow-up. Depressive symptoms were assessed with Patient Health Questionnaire-9 at follow-up and categorized into motivational and mood symptoms.

Regression analyses adjusted for age, gender, education, Mini-Mental State Examination, physical functioning, antidepressant use and vascular risk factors showed that location in mainly deep white-matter tracts and progression of WMLs were associated with symptoms of anhedonia, concentration problems, psychomotor retardation and appetite disturbance. Lacunar infarcts in deep white matter were associated with greater motivational [Incidence rate ratio (IRR) 1.7, 95% confidence interval (CI) 1.2–2.4] and mood (IRR 1.7, 95% CI 1.1–2.6) sumscores, and with symptoms of psychomotor retardation, energy loss and depressed mood; lacunar infarcts in the thalamus were associated with psychomotor retardation only. Cortical atrophy was associated with symptoms of anhedonia and appetite disturbance. Excluding patients with major depression did not materially change the results.

Our findings suggest that disruption of frontal-subcortical pathways by small-vessel lesions leads to a symptom profile that is mainly characteristic of motivational problems, also in the absence of major depression.

Transcranial direct current stimulation (tDCS) could be a side-effect-free alternative to psychostimulants in attention-deficit/hyperactivity disorder (ADHD). Although there is limited evidence for clinical and cognitive effects, most studies were small, single-session and stimulated left dorsolateral prefrontal cortex (dlPFC). No sham-controlled study has stimulated the right inferior frontal cortex (rIFC), which is the most consistently under-functioning region in ADHD, with multiple anodal-tDCS sessions combined with cognitive training (CT) to enhance effects. Thus, we investigated the clinical and cognitive effects of multi-session anodal-tDCS over rIFC combined with CT in double-blind, randomised, sham-controlled trial (RCT, ISRCTN48265228).

Fifty boys with ADHD (10–18 years) received 15 weekday sessions of anodal- or sham-tDCS over rIFC combined with CT (20 min, 1 mA). ANCOVA, adjusting for baseline measures, age and medication status, tested group differences in clinical and ADHD-relevant executive functions at posttreatment and after 6 months.

ADHD-Rating Scale, Conners ADHD Index and adverse effects were significantly lower at post-treatment after sham relative to anodal tDCS. No other effects were significant.

This rigorous and largest RCT of tDCS in adolescent boys with ADHD found no evidence of improved ADHD symptoms or cognitive performance following multi-session anodal tDCS over rIFC combined with CT. These findings extend limited meta-analytic evidence of cognitive and clinical effects in ADHD after 1–5 tDCS sessions over mainly left dlPFC. Given that tDCS is commercially and clinically available, the findings are important as they suggest that rIFC stimulation may not be indicated as a neurotherapy for cognitive or clinical remediation for ADHD.

Studies investigating neurocognitive impairment in subjects with eating disorders (EDs) have reported heterogeneous patterns of impairment and, in some instances, no dysfunction. The present study aimed to define the pattern of neurocognitive impairment in a large sample of bulimia nervosa (BN) patients and to demonstrate that neuroendocrine, personality and clinical characteristics influence neurocognitive performance in BN.

Attention/immediate memory, set shifting, perseveration, conditional and implicit learning were evaluated in 83 untreated female patients with BN and 77 healthy controls (HC). Cortisol and 17β-estradiol plasma levels were assessed. Cloninger's Temperament and Character Inventory – Revised (TCI-R), the Bulimic Investigation Test Edinburgh (BITE) and the Montgomery–Asberg Depression Rating Scale (MADRS) were administered.

No impairment of cognitive performance was found in subjects with BN compared with HC. Cortisol and ‘Self-directedness’ were associated with better performance on conditional learning whereas 17β-estradiol had a negative influence on this domain; ‘Reward dependence’ was associated with worse performance on implicit learning; and depressive symptomatology influenced performance on the Wisconsin Card Sorting Test (WCST) negatively.

No cognitive impairment was found in untreated patients with BN. Neuroendocrine, personality and clinical variables do influence neurocognitive functioning and might explain discrepancies in literature findings.

Even though recent research indicates that sexual symptoms are highly prevalent in post-traumatic stress disorder following childhood sexual abuse and cause severe distress, current treatments neither address them nor are they effective in reducing them. This might be due to a lack of understanding of sexual symptoms' specific role in the often complex and comorbid psychopathology of post-traumatic stress disorder following childhood abuse.

Post-traumatic, dissociative, depressive, and sexual symptoms were assessed in 445 inpatients with post-traumatic stress disorder following childhood sexual abuse. Comorbidity structure was analyzed using a partial correlation network with regularization.

A total of 360 patients (81%) reported difficulties engaging in sexual activities and 102 patients (23%) reported to suffer from their sexual preferences. Difficulties engaging in sexual activities were linked to depressive and hyperarousal symptoms, whereas sexual preferences causing distress were linked to anger and dissociation. Dissociative amnesia, visual intrusions, and physical reactions to trauma reminders were of central importance for the network. Dissociative amnesia, depressed mood, lack of energy, and difficulties engaging in sexual activities were identified as bridge symptoms. Local clustering analysis indicated the non-redundancy of sexual symptoms.

Sexual symptoms are highly prevalent in survivors of childhood sexual abuse with post-traumatic stress disorder. Further research is needed regarding the link of difficulties engaging in sexual activities, depression, and post-traumatic stress disorder, as well as regarding the association of dissociation and sexual preferences causing distress. Sexual symptoms require consideration in the treatment of post-traumatic stress disorder following childhood sexual abuse.

When sober, problematic drinkers display exaggerated reactivity to threats that are uncertain (U-threat). Since this aversive affective state can be alleviated via acute alcohol intoxication, it has been posited that individuals who exhibit heightened reactivity to U-threat at baseline are motivated to use alcohol as a means of avoidance-based coping, setting the stage for excessive drinking. To date, however, no study has attempted to characterize the dispositional nature of exaggerated reactivity to U-threat and test whether it is a vulnerability factor or exclusively a disease marker of problematic alcohol use.

The current investigation utilized a family study design to address these gaps by examining whether (1) reactivity to U-threat is associated with risk for problematic alcohol use, defined by family history of alcohol use disorder (AUD) and (2) reactivity to U-threat is correlated amongst adult biological siblings. A total of 157 families, and 458 individuals, participated in the study and two biological siblings completed a threat-of-shock task designed to probe reactivity to U-threat and predictable threat (P-threat). Startle potentiation was collected as an index of aversive responding.

Within biological siblings, startle potentiation to U-threat [intraclass correlation (ICC) = 0.35] and P-threat (ICC = 0.63) was significantly correlated. In addition, independent of an individuals’ own AUD status, startle potentiation to U-threat, but not P-threat, was positively associated with risk for AUD (i.e. AUD family history).

This suggests that heightened reactivity to U-threat may be a familial vulnerability factor for problematic drinking and a novel prevention target for AUD.

It has been argued that sleep disturbances are a risk factor for depression but previous longitudinal studies have had limitations and not addressed alternative explanations. The aim of this study was to examine the longitudinal association between sleep disturbances and depressive symptoms in a nationally representative sample.

Data from the 18-month follow-up of the UK National Psychiatric Morbidity survey were used (n = 2406). Sleep disturbances, depressive and other psychiatric symptoms (fatigue, concentration problems, irritability, anxiety and pain symptoms) were assessed using the Revised Clinical Interview Schedule (CIS-R). The bidirectional association between symptoms was investigated with logistic regression analyses and path analysis.

Sleep disturbances and depressive symptoms were correlated with each other cross-sectionally (r = 0.52, p < 0.001). In the longitudinal analysis, sleep disturbances at baseline did not predict depressive symptoms at follow-up [odds ratio (OR) 1.27, 95% confidence interval (CI) 0.51–3.19] and the same was observed for the reciprocal association (OR 0.87, 95% CI 0.56–1.35). In the path analysis, the reciprocal model did not have a better fit compared to the simpler first-order model without cross-lagged paths. The path from sleep disturbances at baseline to depressive symptoms at follow-up had a minimal contribution to the explained variance of the latter (<1%).

Previous studies may have overestimated the importance of sleep disturbances as an independent risk factor of depression. The strong cross-sectional association is compatible with sleep disturbances being either a prodromal or a residual symptom of depression and this may have implications for recognition and treatment of depression.

Both maternal obesity and disordered mood have adverse effects on pregnancy outcome. We hypothesized that maternal very severe obesity (SO) is associated with increased anxiety and depression (A&D) symptoms during pregnancy, with adverse effects on gestational weight gain (GWG), postpartum mood and postpartum weight retention (PPWR) and explored any mediation by circulating glucocorticoids.

We measured A&D symptoms with validated questionnaires at weeks 17 and 28 of pregnancy and 3 months postpartum in 135 lean [body mass index (BMI) ⩽25 kg/m2] and 222 SO (BMI ⩾40 kg/m2) pregnant women. Fasting serum cortisol was measured by radioimmunoassay; GWG and PPWR were recorded.

A&D symptoms were higher in the SO group during pregnancy and postpartum despite adjusting for multiple confounders including previous mental health diagnosis (p < 0.05), and were non-linearly correlated with total GWG (anxiety R2 = 0.06, p = 0.037; depression R2 = 0.09, p = 0.001). In the SO group only, increased maternal anxiety (β = 0.33, p = 0.03) and depression (β = 0.19, p = 0.04) symptoms at week 17 of pregnancy were associated with increased PPWR, independent of total GWG and breastfeeding. Anxiety symptoms at week 28 of pregnancy, but not depression, were non-linearly correlated with serum cortisol level at week 36 of pregnancy (R2 = 0.06, p = 0.02). Cortisol did not mediate the link between A&D symptoms and GWG.

Maternal SO was associated with increased A&D symptoms, and with adverse effects on GWG and PPWR independent of circulating glucocorticoids. Strategies to optimize GWG and postpartum weight management in SO women should include assessment and management of maternal mood in early pregnancy.

A wealth of clinical studies have identified objective biomarkers, which separate schizophrenia patients from healthy controls on a group level, but current diagnostic systems solely include clinical symptoms. In this study, we investigate if machine learning algorithms on multimodal data can serve as a framework for clinical translation.

Forty-six antipsychotic-naïve, first-episode schizophrenia patients and 58 controls underwent neurocognitive tests, electrophysiology, and magnetic resonance imaging (MRI). Patients underwent clinical assessments before and after 6 weeks of antipsychotic monotherapy with amisulpride. Nine configurations of different supervised machine learning algorithms were applied to first estimate the unimodal diagnostic accuracy, and next to estimate the multimodal diagnostic accuracy. Finally, we explored the predictability of symptom remission.

Cognitive data significantly classified patients from controls (accuracies = 60–69%; p values = 0.0001–0.009). Accuracies of electrophysiology, structural MRI, and diffusion tensor imaging did not exceed chance level. Multimodal analyses with cognition plus any combination of one or more of the remaining three modalities did not outperform cognition alone. None of the modalities predicted symptom remission.

In this multivariate and multimodal study in antipsychotic-naïve patients, only cognition significantly discriminated patients from controls, and no modality appeared to predict short-term symptom remission. Overall, these findings add to the increasing call for cognition to be included in the definition of schizophrenia. To bring about the full potential of machine learning algorithms in first-episode, antipsychotic-naïve schizophrenia patients, careful a priori variable selection based on independent data as well as inclusion of other modalities may be required.

Relatives of schizophrenia patients demonstrate abnormalities in prefrontal cortical activation during executive processing as measured by functional neuroimaging, albeit not consistently. A meta-analysis was conducted to determine whether reliable patterns of brain hypo- and hyperactivity, especially in the middle frontal region, were present in the relatives of patients.

Seventeen studies, containing 18 samples of relatives and controls, were included in this meta-analysis. Studies were included if relatives of schizophrenia patients were compared to controls, an executive processing task was used, and standard space coordinates were reported for the functional activations. Activation likelihood estimation (ALE) was implemented to find convergence across functional neuroimaging experiment coordinates. A separate analysis was conducted to assess the potential impact of a priori hypothesis testing used in region-of-interest (ROI) approaches on the meta-analysis results.

Relatives demonstrated hypo- and hyperactivity in statistically overlapping right middle frontal regions [Brodmann area (BA) 9/10]. Use of an ROI analysis that a priori focused on prefrontal regions resulted in more findings of reduced activity in the middle frontal region.

The cortical regions identified by this meta-analysis could potentially serve as intermediate biological markers in the search for candidate genes for schizophrenia. As neurocognitive deficits are related to functional impairments in patients, a better understanding of neural and genetic vulnerabilities would be beneficial in our efforts to remediate these important deficits.