Identifying patients with first-episode psychosis (FEP) who are unlikely to achieve early clinical recovery (ECR) is critical for personalised intervention and resource allocation. ECR – defined as the concurrent achievement of symptomatic and functional remission – represents a clinically meaningful outcome that captures both illness control and functional reintegration.

To develop and externally validate prediction models for ECR using clinical, cognitive and genetic data.

We analysed two large, independent Spanish cohorts: the primeros episodios psicóticos cohort (N = 335), for model development and internal validation, and the Programa Asistencial a las Fases Iniciales de Psicosis cohort (N = 668), for external validation. Forty-seven baseline clinical and cognitive variables and 87 polygenic risk scores (PRSs) were examined. Predictors were selected using penalised logistic regression. Logistic regression and three machine learning algorithms were compared for discrimination, calibration and clinical utility.

The best-performing model was a logistic regression using six routinely collected clinical and cognitive predictors (duration of untreated psychosis, days of treated psychosis, baseline functioning, insight, executive function and cognitive reserve), with an optimism-corrected area under the receiver operating characteristic curve of 0.73 in development and 0.63 in external validation. PRS models showed limited external generalisability and did not improve prediction. Machine learning algorithms offered no advantage over regression models.

A simple, interpretable logistic regression model based on routine clinical and cognitive variables can predict early recovery in FEP with acceptable generalisability. These findings support the use of transparent, clinically grounded models in early psychosis care and highlight the current limitations of genetic predictors for individualised treatment.

Machine learning (ML) models show promise in predicting post-traumatic stress disorder (PTSD) treatment outcomes, but it is unknown how their predictions compare to those of clinicians. This study directly compared the accuracy of clinicians’ predictions of patient treatment outcomes with those of three ML models.

Twenty clinicians providing cognitive processing therapy repeatedly predicted outcomes for 194 veterans. We compared their accuracy against three ML models on two key endpoints: clinically meaningful symptom reduction (≥10-point PCL-5 decrease) and posttreatment severity (final PCL-5 < 33). Clinician predictions were compared against a recurrent neural network, a mixed-effects random forest, and a generalized linear mixed-effects model. We analyzed prediction accuracy and the association between clinician confidence and accuracy using logistic mixed-effects models.

ML models were significantly more accurate than clinicians at predicting whether a patient’s posttreatment PCL-5 score would be below 33 (p < .001). However, no significant difference in accuracy was found for predicting a ≥10-point symptom reduction (p = .734). Clinician confidence increased throughout treatment and was significantly associated with greater prediction accuracy for both outcomes (ORs = 1.06, ps < .001).

ML models can outperform clinicians in predicting posttreatment symptom severity, particularly early in treatment, suggesting they could be a useful tool for identifying patients at risk for suboptimal outcomes. However, ML models were not superior in predicting symptom reduction, where clinicians also performed at a high level. Findings support the selective use of ML to enhance, rather than replace, clinical judgment in PTSD treatment.

Epigenetic mechanisms might play a role in modulating susceptibility to bipolar disorder (BD) and response to lithium, the mainstay treatment for BD. Additionally, individuals with BD experience accelerated biological aging.

We compared blood DNA methylation profiles measured with EPIC v.2.0 arrays between patients with BD (33 lithium responders and 31 nonresponders) and nonpsychiatric controls (n = 32), as well as based on long-term lithium response. In addition, we compared cellular aging between these groups using epigenetic age, pace of aging, and, for the first time, transcriptional age acceleration based on bulk RNA sequencing in 93 patients and 56 controls.

We identified 191 differentially methylated positions (DMPs) and 8 differentially methylated regions between patients with BD and controls, located in genes enriched for “Postsynaptic Density” (odds ratio = 6.81, p = 0.001). No DMP was significantly associated with lithium response after multiple testing correction. Patients showed a significantly higher biological age acceleration than controls based on two epigenetic clocks (GrimAge, Mann–Whitney U = 551, p = 0.0009; GrimAge2: U = 477, p = 9.0E-05) and pace of aging (DunedinPACE, t = 3.01, p = 0.003), but not on transcriptional age. While we observed no significant difference in epigenetic aging based on lithium response, lithium responders showed lower epigenetic acceleration using all clocks, with a trend observed using the PhenoAge clock (t = 1.97, p = 0.053).

Our findings point to methylation patterns characterizing BD and support the hypothesis of accelerated cellular aging in BD.

Network analysis is a promising approach for elucidating the dynamics of the transition from psychopathology to well-being. Recently, symptom connectivity strength has been proposed as a measure of plasticity – the capacity to change disease severity. Yet, empirical findings remain inconsistent. We propose that this inconsistency can be resolved by recognizing that the interpretation of connectivity strength varies along the recovery process from depression, whether at baseline or during clinical change.

We analyzed 2,710 depressed patients from the STAR*D dataset, grouped by the magnitude of change in depressive score. Symptom network connectivity was estimated from QIDS-C items at three time points: (i) baseline, (ii) change – defined as when clinical change in depression score occurs, (iii) post-change - corresponding to when the maximum clinical change is achieved.

At baseline, connectivity strength predicts the maximum clinical change, inversely correlating with its magnitude (ρ = −0.95, p = 0.001). At the change time point, connectivity strength parallels clinical change (ρ = 0.92, p = 0.002). A direct and significant association between connectivity strength and depression severity emerges only at the change (ρ = 0.98, p = 0.0003) and post-change (ρ = 0.95, p = 0.001) time points.

The interpretation of connectivity strength for predicting depression trajectories varies by timepoint: at baseline, it measures plasticity -- the capacity for change -- whereas during clinical change, it indicates the magnitude of change in symptom severity. This framework supports the reliability of this prognostic marker for designing personalized therapeutic interventions in psychiatry.

Depressive symptoms are highly prevalent in first-episode psychosis (FEP) and worsen clinical outcomes. It is currently difficult to determine which patients will have persistent depressive symptoms based on a clinical assessment. We aimed to determine whether depressive symptoms and post-psychotic depressive episodes can be predicted from baseline clinical data, quality of life, and blood-based biomarkers, and to assess the geographical generalizability of these models.

Two FEP trials were analyzed: European First-Episode Schizophrenia Trial (EUFEST) (n = 498; 2002–2006) and Recovery After an Initial Schizophrenia Episode Early Treatment Program (RAISE-ETP) (n = 404; 2010–2012). Participants included those aged 15–40 years, meeting Diagnostic and Statistical Manual of Mental Disorders IV criteria for schizophrenia spectrum disorders. We developed support vector regressors and classifiers to predict changes in depressive symptoms at 6 and 12 months and depressive episodes within the first 6 months. These models were trained in one sample and externally validated in another for geographical generalizability.

A total of 320 EUFEST and 234 RAISE-ETP participants were included (mean [SD] age: 25.93 [5.60] years, 56.56% male; 23.90 [5.27] years, 73.50% male). Models predicted changes in depressive symptoms at 6 months with balanced accuracy (BAC) of 66.26% (RAISE-ETP) and 75.09% (EUFEST), and at 12 months with BAC of 67.88% (RAISE-ETP) and 77.61% (EUFEST). Depressive episodes were predicted with BAC of 66.67% (RAISE-ETP) and 69.01% (EUFEST), showing fair external predictive performance.

Predictive models using clinical data, quality of life, and biomarkers accurately forecast depressive events in FEP, demonstrating generalization across populations.

Functional impairment is a major concern among those presenting to youth mental health services and can have a profound impact on long-term outcomes. Early recognition and prevention for those at risk of functional impairment is essential to guide effective youth mental health care. Yet, identifying those at risk is challenging and impacts the appropriate allocation of indicated prevention and early intervention strategies.

We developed a prognostic model to predict a young person’s social and occupational functional impairment trajectory over 3 months. The sample included 718 young people (12–25 years) engaged in youth mental health care. A Bayesian random effects model was designed using demographic and clinical factors and model performance was evaluated on held-out test data via 5-fold cross-validation.

Eight factors were identified as the optimal set for prediction: employment, education, or training status; self-harm; psychotic-like experiences; physical health comorbidity; childhood-onset syndrome; illness type; clinical stage; and circadian disturbances. The model had an acceptable area under the curve (AUC) of 0.70 (95% CI, 0.56–0.81) overall, indicating its utility for predicting functional impairment over 3 months. For those with good baseline functioning, it showed excellent performance (AUC = 0.80, 0.67–0.79) for identifying individuals at risk of deterioration.

We developed and validated a prognostic model for youth mental health services to predict functional impairment trajectories over a 3-month period. This model serves as a foundation for further tool development and demonstrates its potential to guide indicated prevention and early intervention for enhancing functional outcomes or preventing functional decline.

International guidelines present overall symptom severity as the key dimension for clinical characterisation of major depressive disorder (MDD). However, differences may reside within severity levels related to how symptoms interact in an individual patient, called symptom dynamics.

To investigate these individual differences by estimating the proportion of patients that display differences in their symptom dynamics while sharing the same overall symptom severity.

Participants with MDD (n = 73; mean age 34.6 years, s.d. = 13.1; 56.2% female) rated their baseline symptom severity using the Inventory for Depressive Symptomatology Self-Report (IDS-SR). Momentary indicators for depressive symptoms were then collected through ecological momentary assessments five times per day for 28 days; 8395 observations were conducted (average per person: 115; s.d. = 16.8). Each participant's symptom dynamics were estimated using person-specific dynamic network models. Individual differences in these symptom relationship patterns in groups of participants sharing the same symptom severity levels were estimated using individual network invariance tests. Subsequently, the overall proportion of participants that displayed differential symptom dynamics while sharing the same symptom severity was calculated. A supplementary simulation study was conducted to investigate the accuracy of our methodology against false-positive results.

Differential symptom dynamics were identified across 63.0% (95% bootstrapped CI 41.0–82.1) of participants within the same severity group. The average false detection of individual differences was 2.2%.

The majority of participants within the same depressive symptom severity group displayed differential symptom dynamics. Examining symptom dynamics provides information about person-specific psychopathological expression beyond severity levels by revealing how symptoms aggravate each other over time. These results suggest that symptom dynamics may be a promising new dimension for clinical characterisation, warranting replication in independent samples. To inform personalised treatment planning, a next step concerns linking different symptom relationship patterns to treatment response and clinical course, including patterns related to spontaneous recovery and forms of disorder progression.

Irremediability is a key requirement for euthanasia and assisted suicide for psychiatric disorders (psychiatric EAS). Countries like the Netherlands and Belgium ask clinicians to assess irremediability in light of the patient's diagnosis and prognosis and ‘according to current medical understanding’. Clarifying the relevance of a default objective standard for irremediability when applied to psychiatric EAS is crucial for solid policymaking. Yet so far, a thorough examination of this standard is lacking.

Using treatment-resistant depression (TRD) as a test case, through a scoping review in PubMed, we analyzed the state-of-the-art evidence for whether clinicians can accurately predict individual long-term outcome and single out irremediable cases, by examining the following questions: (1) What is the definition of TRD; (2) What are group-level long-term outcomes of TRD; and (3) Can clinicians make accurate individual outcome predictions in TRD?

A uniform definition of TRD is lacking, with over 150 existing definitions, mostly focused on psychopharmacological research. Available yet limited studies about long-term outcomes indicate that a majority of patients with long-term TRD show significant improvement over time. Finally, evidence about individual predictions in TRD using precision medicine is growing, but methodological shortcomings and varying predictive accuracies pose important challenges for its implementation in clinical practice.

Our findings support the claim that, as per available evidence, clinicians cannot accurately predict long-term chances of recovery in a particular patient with TRD. This means that the objective standard for irremediability cannot be met, with implications for policy and practice of psychiatric EAS.

Involuntary admissions are increasing in numbers across Europe.1 They can be traumatic for the patients2 and are associated with large societal costs.3 Individuals with psychotic disorder are at particularly elevated risk of involuntary admission.

This study aims to investigate whether machine learning methods including natural language processing can predict involuntary admission among patients with psychotic disorder.

We have obtained a dataset based on electronic health records for all patients having had at least one contact with the psychiatric services in the Central Denmark Region from 2011 to 2021. This dataset covers more than 120,000 patients, of which approximately 10,000 have been diagnosed with a psychotic disorder. The dataset contains both structured data, such as diagnoses, blood tests etc., as well as unstructured data (text). We will train machine learning models, basic logistic regression-models as well as state-of-the-art neural networks, to predict involuntary admission after contacts to the psychiatric services.

As the machine learning models are under development, no results are available at this time. Preliminary results are expected in spring 2022.

If involuntary admission can be predicted among patients with psychotic disorder based on data from electronic health records, it will pave the way for potentially preventive interventions. References: 1. Sheridans-Rains, L et al., 2019 2. Frueh, B.C et al., 2005 3. Smith,S., 2020

No significant relationships.

Despite the widespread use of long-acting injectable (LAI) antipsychotics in schizophrenia and other disorders, there is a lack of longitudinal studies evaluating prescription trends and the usefulness of therapeutic drug monitoring (TDM) to inform dosing. Indeed, LAI prescription varies greatly among different areas of the world and over the years.

Assess trends in LAI prescription in 2013-2020 at the Psychiatry Department of Bozen, Italy, and (2) analyze the correlation between dose of prescribed LAIs and blood levels measured via TDM.

Parametric statistics.

LAIs were administered to 471 patients (x̅ age±SD=47.2±16.3 years; 56.3% men). The pie chart shows LAI treatment duration, i.e., from 1 to 7 consecutive years. The most used LAIs were haloperidol in 2013-2104 (26.5-31.8%) and paliperidone in 2015-2020 (22.5-25.7%). Dose adjustments were rather frequent, whereas the switch between LAI, due to moderate-to-side effects or unsatisfactory improvement of clinical status, was infrequent (41 cases/8 years). LAI interruption for the same reasons or non-compliance was even more infrequent (10 cases), but in 8 cases it happened for opposite reasons, i.e., achievement of patients’ stabilization and good compliance. The Table shows doses and plasma levels of LAIs. Correlations between doses and plasma levels were: haloperidol: r=-0.037, p=0.620; paliperidone: r=0.290; p=0.000; risperidone: r=0.219, p=0.006; fluphenazine r=0.358, p=0.000; aripiprazole: r=-0.051, p=0.610; olanzapine: r=-0.090, p=0.634.

Haloperidol and paliperidone were the most used LAIs. Drug prescription trends and doses were stable over time. A significant positive correlation between dose and plasma level was found for paliperidone, fluphenazine, and aripiprazole.

No significant relationships.

The term “personalised therapy” refers to the use of genetic data to better treat or determine the predisposition to a specific genetic disease, with the ultimate goal of improving quality of life. Telepsychiatry and AI are key to support it..

Determine benefits of pharmacogenomic analysis (PGx) in CNS diseases regarding: - cost effectiveness - adverse drug reactions - reduced hospitalizations -drug interactions - efficacy - quality of life - “trial and error” approach avoidance

Questionnaires before and after the treatment provided using PGX tests Telepsychiatry for consultation along face to face sessions were conducted. Artificial intelligence in data analyses

Benefits of pharmacogenomic analysis (PGx) in CNS diseases: - cost effective savings - prediction and prevention of adverse drug reactions - reduced hospitalization due to ineffectiveness of medication - reduced risk of drug interactions - more effective treatments - better quality of life for the patient - with the analysis (PGx) the “trial and error” approach is avoided

In a number of studies in patients with mental disorders, pharmacogenomic analysis (PGx) has led to an increase in both clinical response and remission, better tolerated treatments, fewer side effects, and reduced treatment costs. In conclusion, pharmacogenomic analysis is ideal for patients with CNS diseases: a) Not responding to treatment b) Who in their history have many relapses and hospitalizations c) They show serious side effects d) Who do not comply with the treatment e) Taking many medications and suffering from serious illnesses f) Who are wary of taking psychotropic drugs

No significant relationships.