Less than a third of patients with depression achieve successful remission with standard first-step antidepressant monotherapy. The process for determining appropriate second-step care is often based on clinical intuition and involves a protracted course of trial and error, resulting in substantial patient burden and unnecessary delay in the provision of optimal treatment. To address this problem, we adopt an ensemble machine learning approach to improve prediction accuracy of remission in response to second-step treatments.

Data were derived from the Level 2 stage of the STAR*D dataset, which included 1439 patients who were randomized into one of seven different second-step treatment strategies after failing to achieve remission during first-step antidepressant treatment. Ensemble machine learning models, comprising several individual algorithms, were evaluated using nested cross-validation on 155 predictor variables including clinical and demographic measures.

The ensemble machine learning algorithms exhibited differential classification performance in predicting remission status across the seven second-step treatments. For the full set of predictors, AUC values ranged from 0.51 to 0.82 depending on the second-step treatment type. Predicting remission was most successful for cognitive therapy (AUC = 0.82) and least successful for other medication and combined treatment options (AUCs = 0.51–0.66).

Ensemble machine learning has potential to predict second-step treatment. In this study, predictive performance varied by type of treatment, with greater accuracy in predicting remission in response to behavioral treatments than to pharmacotherapy interventions. Future directions include considering more informative predictor modalities to enhance prediction of second-step treatment response.

Researchers have identified genetic and neural risk factors for externalizing behaviors. However, it has not yet been determined if genetic liability is conferred in part through associations with more proximal neurophysiological risk markers.

Participants from the Collaborative Study on the Genetics of Alcoholism, a large, family-based study of alcohol use disorders were genotyped and polygenic scores for externalizing (EXT PGS) were calculated. Associations with target P3 amplitude from a visual oddball task (P3) and broad endorsement of externalizing behaviors (indexed via self-report of alcohol and cannabis use, and antisocial behavior) were assessed in participants of European (EA; N = 2851) and African ancestry (AA; N = 1402). Analyses were also stratified by age (adolescents, age 12–17 and young adults, age 18–32).

The EXT PGS was significantly associated with higher levels of externalizing behaviors among EA adolescents and young adults as well as AA young adults. P3 was inversely associated with externalizing behaviors among EA young adults. EXT PGS was not significantly associated with P3 amplitude and therefore, there was no evidence that P3 amplitude indirectly accounted for the association between EXT PGS and externalizing behaviors.

Both the EXT PGS and P3 amplitude were significantly associated with externalizing behaviors among EA young adults. However, these associations with externalizing behaviors appear to be independent of each other, suggesting that they may index different facets of externalizing.

Cross-sectional studies report high levels of depressive symptoms during the COVID-19 pandemic, especially in youth and females. However, longitudinal research comparing depressive symptoms before and during the pandemic is lacking. Little is known about how the pandemic affected individuals with familial history of mental illness. The present study examines the impact of the pandemic on youth depressive symptoms, including offspring of parents with major mood and psychotic disorders.

Between March 2018 and February 2020, we measured depressive symptoms in 412 youth aged 5–25 years. We measured depressive symptoms again in 371 (90%) of these youth between April 2020 and May 2022. Two thirds (249) participants had a biological parent with a major mood or psychotic disorder. We tested the effect of the pandemic by comparing depression symptoms before and after March 2020. We examined age, sex, and family history as potential moderators.

We found an overall small increase in youth depressive symptoms (b = 0.07, 95% CI −0.01 to 0.15, p = 0.062). This was driven by an increase in female youth without familial history of mental illness (b = 0.35, 95% CI 0.14 to 0.56, p = 0.001). There was no change in depressive symptoms among offspring of parents with mental illness or males.

Our results provide reassurance about the wellbeing of children of parents with mental illness during a period of restricted access to resources outside the family. Rather than increasing symptoms in established risk groups, the pandemic led to a redistribution of depression burden towards segments of the youth population that were previously considered to be low-risk.

Research suggests that most mental health conditions have their onset in the critically social period of adolescence. Yet, we lack understanding of the potential social processes underlying early psychopathological development. We propose a conceptual model where daily-life social interactions and social skills form an intermediate link between known risk and protective factors (adverse childhood experiences, bullying, social support, maladaptive parenting) and psychopathology in adolescents – that is explored using cross-sectional data.

N = 1913 Flemish adolescent participants (Mean age = 13.8; 63% girls) were assessed as part of the SIGMA study, a large-scale, accelerated longitudinal study of adolescent mental health and development. Self-report questionnaires (on risk/protective factors, social skills, and psychopathology) were completed during class time; daily-life social interactions were measured during a subsequent six-day experience-sampling period.

Registered uncorrected multilevel linear regression results revealed significant associations between all risk/protective factors and psychopathology, between all risk/protective factors and social processes, and between all social processes and psychopathology. Social processes (social skills, quantity/quality of daily social interactions) were uniquely predicted by each risk/protective factor and were uniquely associated with both general and specific types of psychopathology. For older participants, some relationships between social processes and psychopathology were stronger.

Unique associations between risk/protective factors and psychopathology signify the distinct relevance of these factors for youth mental health, whereas the broad associations with social processes support these processes as broad correlates. Results align with the idea of a social pathway toward early psychopathology, although follow-up longitudinal research is required to verify any mediation effect.

Binge-eating disorder (BED) co-occurs with neurobehavioral alterations in the processing of disorder-relevant content such as visual food stimuli. Whether neurofeedback (NF) directly targeting them is suited for treatment remains unclear. This study sought to determine feasibility and estimate effects of individualized, functional near-infrared spectroscopy-based real-time NF (rtfNIRS-NF) and high-beta electroencephalography-based NF (EEG-NF), assuming superiority over waitlist (WL).

Single-center, assessor-blinded feasibility study with randomization to rtfNIRS-NF, EEG-NF, or WL and assessments at baseline (t0), postassessment (t1), and 6-month follow-up (t2). NF comprised 12 60-min food-specific rtfNIRS-NF or EEG-NF sessions over 8 weeks. Primary outcome was the binge-eating frequency at t1 assessed interview-based. Secondary outcomes included feasibility, eating disorder symptoms, mental and physical health, weight management-related behavior, executive functions, and brain activity at t1 and t2.

In 72 patients (intent-to-treat), the results showed feasibility of NF regarding recruitment, attrition, adherence, compliance, acceptance, and assessment completion. Binge eating improved at t1 by −8.0 episodes, without superiority of NF v. WL (−0.8 episodes, 95% CI −2.4 to 4.0), but with improved estimates in NF at t2 relative to t1. NF was better than WL for food craving, anxiety symptoms, and body mass index, but overall effects were mostly small. Brain activity changes were near zero.

The results show feasibility of food-specific rtfNIRS-NF and EEG-NF in BED, and no posttreatment differences v. WL, but possible continued improvement of binge eating. Confirmatory and mechanistic evidence is warranted in a double-blind randomized design with long-term follow-up, considering dose–response relationships and modes of delivery.

Eating disorders (ED) are serious psychiatric disorders, taking a life every 52 minutes, with high relapse. There are currently no support or effective intervention therapeutics for individuals with an ED in their everyday life. The aim of this study is to build idiographic machine learning (ML) models to evaluate the performance of physiological recordings to detect individual ED behaviors in naturalistic settings.

From an ongoing study (Final N = 120), we piloted the ability for ML to detect an individual's ED behavioral episodes (e.g. purging) from physiological data in six individuals diagnosed with an ED, all of whom endorsed purging. Participants wore an ambulatory monitor for 30 days and tapped a button to denote ED behavioral episodes. We built idiographic (N = 1) logistic regression classifiers (LRC) ML trained models to identify onset of episodes (~600 windows) v. baseline (~571 windows) physiology (Heart Rate, Electrodermal Activity, and Temperature).

Using physiological data, ML LRC accurately classified on average 91% of cases, with 92% specificity and 90% sensitivity.

This evidence suggests the ability to build idiographic ML models that detect ED behaviors from physiological indices within everyday life with a high level of accuracy. The novel use of ML with wearable sensors to detect physiological patterns of ED behavior pre-onset can lead to just-in-time clinical interventions to disrupt problematic behaviors and promote ED recovery.

People with borderline intellectual functioning (BIF) encounter greater social adversities than the general population and have an increased prevalence of mental illness. However, little is known about the socio-demographic characteristics and mental health of parents with BIF.

A secondary data analysis of the Adult Psychiatric Morbidity Survey 2014 was conducted. Logistic regression models were fitted to compare differences in socio-demographic, mental health and service-use characteristics between parents and non-parents with and without BIF, and to investigate if the relationship between parent status and mental health outcomes was modified by BIF status, sex, and employment.

Data from 6872 participants was analyzed; 69.1% were parents. BIF parents had higher odds of common mental disorder, severe mental illness, post-traumatic stress disorder, self-harm/suicide and were more likely to see their General Practitioner (GP) and to receive mental health treatment than non-BIF parents. BIF parents did not have a higher prevalence of mental health problems than BIF non-parents. Being a parent, after adjusting for BIF status and other confounders, was associated with increased odds of having a common mental disorder, visits to see a GP and treatment for mental health. Female parents had higher odds of treatment for mental health problems.

Being a parent is associated with elevated rates of common mental disorders. There is a higher burden of mental health problems and service use in people with BIF. A greater provision of specialist support services including ascertainment is indicated for this group.

Dysmyelination could be part of the pathophysiology of schizophrenia spectrum (SCZ) and bipolar disorders (BPD), yet few studies have examined myelination of the cerebral cortex. The ratio of T1- and T2-weighted magnetic resonance images (MRI) correlates with intracortical myelin. We investigated the T1w/T2w-ratio and its age trajectories in patients and healthy controls (CTR) and explored associations with antipsychotic medication use and psychotic symptoms.

Patients with SCZ (n = 64; mean age = 30.4 years, s.d. = 9.8), BPD (n = 91; mean age 31.0 years, s.d. = 10.2), and CTR (n = 155; mean age = 31.9 years, s.d. = 9.1) who participated in the TOP study (NORMENT, University of Oslo, Norway) were clinically assessed and scanned using a General Electric 3 T MRI system. T1w/T2w-ratio images were computed using an optimized pipeline with intensity normalization and field inhomogeneity correction. Vertex-wise regression models were used to compare groups and examine group × age interactions. In regions showing significant differences, we explored associations with antipsychotic medication use and psychotic symptoms.

No main effect of diagnosis was found. However, age slopes of the T1w/T2w-ratio differed significantly between SCZ and CTR, predominantly in frontal and temporal lobe regions: Lower T1w/T2w-ratio values with higher age were found in CTR, but not in SCZ. Follow-up analyses revealed a more positive age slope in patients who were using antipsychotics and patients using higher chlorpromazine-equivalent doses.

While we found no evidence of reduced intracortical myelin in SCZ or BPD relative to CTR, different regional age trajectories in SCZ may suggest a promyelinating effect of antipsychotic medication.

There are many barriers to engaging in current psychological treatments, including time, cost, and availability. Ultra-brief treatments overcome some of these barriers by delivering therapeutic information and skills using significantly less time than standard-length treatments. We developed a therapist-guided online ultra-brief treatment for depression and anxiety and compared it to an existing 8-week, 5-lesson therapist-guided standard-length treatment and a waitlist control.

In a randomized controlled trial, adults with self-reported depression or anxiety were randomized (1:1:1) to the ultra-brief treatment, standard-length treatment, or waitlist control. The primary outcomes were depression symptoms and anxiety symptoms assessed at baseline, 5-weeks later, 9-weeks later (primary timepoint), and 3-months later. The trial was prospectively registered.

Between 7 February 2022, and 16 August 2022, 242 participants were enrolled in the ultra-brief treatment (n = 85), standard-length treatment (n = 80), and waitlist control (n = 77). Participants were mostly women with an average age of 48.56 years. At 9-weeks post-baseline, participants in the ultra-brief treatment group reported significantly lower depression (between groups d = 0.41) and anxiety (d = 0.53) than the waitlist control. The ultra-brief treatment was non-inferior for anxiety at both 9-weeks and 3-months follow-up. Non-inferiority for depression was observed at 9-weeks.

The online ultra-brief treatment resulted in significant reductions in depression and anxiety that were non-inferior to a longer treatment course after 9-weeks. Remotely delivered ultra-brief treatments have the potential to provide accessible and effective care for those who cannot, or would prefer not to, access longer psychological interventions.

Identifying adolescents at risk of internalizing problems is a key priority. However, studies have tended to consider such problems in simple ways using diagnoses, or item summaries. Network theory and methods instead allow for more complex interaction between symptoms. Two key hypotheses predict differences in global network properties for those at risk: altered structure and increased connectivity.

The current study evaluated these hypotheses for nine risk factors (e.g. income deprivation and low parent/carer support) individually and cumulatively in a large sample of 12–15 year-olds (N = 34 564). Recursive partitioning and bootstrapped networks were used to evaluate structural and connectivity differences.

The pattern of network interactions was shown to be significantly different via recursive partitioning for all comparisons across risk-present/absent groups and levels of cumulative risk, except for income deprivation. However, the magnitude of differences appeared small. Most individual risk factors also showed relatively small effects for connectivity. Exceptions were noted for gender and sexual minority risk groups, as well as low parent/carer support, where larger effects were evident. A strong linear trend was observed between increasing cumulative risk exposure and connectivity.

A robust approach to considering the effect of risk exposure on global network properties was demonstrated. Results are consistent with the ideas that pathological states are associated with higher connectivity, and that the number of risks, regardless of their nature, is important. Gender/sexual minority status and low parent/carer support had the biggest individual impacts on connectivity, suggesting these are particularly important for identification and prevention.

A large and accumulating body of evidence shows that loneliness is detrimental for various health and well-being outcomes. However, less is known about potentially modifiable factors that lead to decreased loneliness.

We used data from the Health and Retirement Study to prospectively evaluate a wide array of candidate predictors of subsequent loneliness. Importantly, we examined if changes in 69 physical-, behavioral-, and psychosocial-health factors (from t0;2006/2008 to t1;2010/2012) were associated with subsequent loneliness 4 years later (t2;2014/2016).

Adjusting for a large range of covariates, changes in certain health behaviors (e.g. increased physical activity), physical health factors (e.g. fewer functioning limitations), psychological factors (e.g. increased purpose in life, decreased depression), and social factors (e.g. greater number of close friends) were associated with less subsequent loneliness.

Our findings suggest that subjective ratings of physical and psychological health and perceived social environment (e.g. chronic pain, self-rated health, purpose in life, anxiety, neighborhood cohesion) are more strongly associated with subsequent loneliness. Yet, objective ratings (e.g. specific chronic health conditions, living status) show less evidence of associations with subsequent loneliness. The current study identified potentially modifiable predictors of subsequent loneliness that may be important targets for interventions aimed at reducing loneliness.

Research using latent variable models demonstrates that pre-attentive measures of early auditory processing (EAP) and cognition may initiate a cascading effect on daily functioning in schizophrenia. However, such models fail to account for relationships among individual measures of cognition and EAP, thereby limiting their utility. Hence, EAP and cognition may function as complementary and interacting measures of brain function rather than independent stages of information processing. Here, we apply a data-driven approach to identifying directional relationships among neurophysiologic and cognitive variables.

Using data from the Consortium on the Genetics of Schizophrenia 2, we estimated Gaussian Graphical Models and Bayesian networks to examine undirected and directed connections between measures of EAP, including mismatch negativity and P3a, and cognition in 663 outpatients with schizophrenia and 630 control participants.

Chain structures emerged among EAP and attention/vigilance measures in schizophrenia and control groups. Concerning differences between the groups, object memory was an influential variable in schizophrenia upon which other cognitive domains depended, and working memory was an influential variable in controls.

Measures of EAP and attention/vigilance are conditionally independent of other cognitive domains that were used in this study. Findings also revealed additional causal assumptions among measures of cognition that could help guide statistical control and ultimately help identify early-stage targets or surrogate endpoints in schizophrenia.

Individuals with bipolar disorder are commonly correctly diagnosed a decade after symptom onset. Machine learning techniques may aid in early recognition and reduce the disease burden. As both individuals at risk and those with a manifest disease display structural brain markers, structural magnetic resonance imaging may provide relevant classification features.

Following a pre-registered protocol, we trained linear support vector machine (SVM) to classify individuals according to their estimated risk for bipolar disorder using regional cortical thickness of help-seeking individuals from seven study sites (N = 276). We estimated the risk using three state-of-the-art assessment instruments (BPSS-P, BARS, EPIbipolar).

For BPSS-P, SVM achieved a fair performance of Cohen's κ of 0.235 (95% CI 0.11–0.361) and a balanced accuracy of 63.1% (95% CI 55.9–70.3) in the 10-fold cross-validation. In the leave-one-site-out cross-validation, the model performed with a Cohen's κ of 0.128 (95% CI −0.069 to 0.325) and a balanced accuracy of 56.2% (95% CI 44.6–67.8). BARS and EPIbipolar could not be predicted. In post hoc analyses, regional surface area, subcortical volumes as well as hyperparameter optimization did not improve the performance.

Individuals at risk for bipolar disorder, as assessed by BPSS-P, display brain structural alterations that can be detected using machine learning. The achieved performance is comparable to previous studies which attempted to classify patients with manifest disease and healthy controls. Unlike previous studies of bipolar risk, our multicenter design permitted a leave-one-site-out cross-validation. Whole-brain cortical thickness seems to be superior to other structural brain features.

Although numerous studies have examined the effects of psychological treatments for obsessive-compulsive disorder (OCD), their overall effectiveness remains unclear. We aimed to estimate their overall effect by combining all available randomized controlled trials (RCTs) comparing psychological treatments to control groups for OCD.

We conducted a meta-analysis of 48 RCTs with 55 comparisons published between 1992 and 1 January 2023. The primary outcome was OCD symptom severity, with Hedges' g calculated at post-treatment and follow-up. Random-effects models were employed for all analyses, and the risk of bias was assessed.

In general, psychological treatments demonstrated a significantly large effect (g = −1.14; 95% CI [−1.31 to −0.97]; I2 = 72.23%) on reducing OCD symptom severity post-treatment, this finding remained consistent across measures and after excluding outliers, but lost significance in the sensitivity analysis for only studies with low risk of bias. Type of treatment, control group and treatment format were associated with treatment effects. Moreover, more severe baseline OCD symptom severity predicted higher degree of treatment efficacy. No significant differences were observed in dropout rates between the treatment and control groups. Treatment effects lost significance at 3–6 and 6–12 month follow-ups. 87% of RCTs were rated at high risk of bias.

Psychological treatments are effective in reducing OCD symptom severity. However, caution should be exercised when interpreting these results due to the high heterogeneity and risk of bias across RCTs. Future studies with more rigorous methodology are required, as well as studies examining their long-term effectiveness.