Optimizing Outcomes in Psychotherapy for Anxiety Disorders Using Smartphone-Based and Passive Sensing Features: Protocol for a Randomized Controlled Trial.

BACKGROUND: Psychotherapies, such as cognitive behavioral therapy (CBT), currently have the strongest evidence of durable symptom changes for most psychological disorders, such as anxiety disorders. Nevertheless, only about half of individuals treated with CBT benefit from it. Predictive algorithms, including digital assessments and passive sensing features, could better identify patients who would benefit from CBT, and thus, improve treatment choices. OBJECTIVE: This study aims to establish predictive features that forecast responses to transdiagnostic CBT in anxiety disorders and to investigate key mechanisms underlying treatment responses. METHODS: This study is a 2-armed randomized controlled clinical trial. We include patients with anxiety disorders who are randomized to either a transdiagnostic CBT group or a waitlist (referred to as WAIT). We index key features to predict responses prior to starting treatment using subjective self-report questionnaires, experimental tasks, biological samples, ecological momentary assessments, activity tracking, and smartphone-based passive sensing to derive a multimodal feature set for predictive modeling. Additional assessments take place weekly at mid- and posttreatment and at 6- and 12-month follow-ups to index anxiety and depression symptom severity. We aim to include 150 patients, randomized to CBT versus WAIT at a 3:1 ratio. The data set will be subject to full feature and important features selected by minimal redundancy and maximal relevance feature selection and then fed into machine leaning models, including eXtreme gradient boosting, pattern recognition network, and k-nearest neighbors to forecast treatment response. The performance of the developed models will be evaluated. In addition to predictive modeling, we will test specific mechanistic hypotheses (eg, association between self-efficacy, daily symptoms obtained using ecological momentary assessments, and treatment response) to elucidate mechanisms underlying treatment response. RESULTS: The trial is now completed. It was approved by the Cantonal Ethics Committee, Zurich. The results will be disseminated through publications in scientific peer-reviewed journals and conference presentations. CONCLUSIONS: The aim of this trial is to improve current CBT treatment by precise forecasting of treatment response and by understanding and potentially augmenting underpinning mechanisms and personalizing treatment. TRIAL REGISTRATION: ClinicalTrials.gov NCT03945617; https://clinicaltrials.gov/ct2/show/results/NCT03945617. INTERNATIONAL REGISTERED REPORT IDENTIFIER (IRRID): DERR1-10.2196/42547.

Resilience and Disaster: Flexible Adaptation in the Face of Uncertain Threat.

Disasters cause sweeping damage, hardship, and loss of life. In this article, we first consider the dominant psychological approach to disasters and its narrow focus on psychopathology (e.g., posttraumatic stress disorder). We then review research on a broader approach that has identified heterogeneous, highly replicable trajectories of outcome, the most common being stable mental health or resilience. We review trajectory research for different types of disasters, including the COVID-19 pandemic. Next, we consider correlates of the resilience trajectory and note their paradoxically limited ability to predict future resilient outcomes. Research using machine learning algorithms improved prediction but has not yet illuminated the mechanism behind resilient adaptation. To that end, we propose a more direct psychological explanation for resilience based on research on the motivational and mechanistic components of regulatory flexibility. Finally, we consider how future research might leverage new computational approaches to better capture regulatory flexibility in real time.

Machine Learning and the Digital Measurement of Psychological Health.

Since its inception, the discipline of psychology has utilized empirical epistemology and mathematical methodologies to infer psychological functioning from direct observation. As new challenges and technological opportunities emerge, scientists are once again challenged to define measurement paradigms for psychological health and illness that solve novel problems and capitalize on new technological opportunities. In this review, we discuss the theoretical foundations of and scientific advances in remote sensor technology and machine learning models as they are applied to quantify psychological functioning, draw clinical inferences, and chart new directions in treatment.

The Heterogeneity of Posttraumatic Stress Disorder in DSM-5.

This diagnostic study reports patterns of DSM-5 posttraumatic stress disorder diagnostic presentations from multiple cohorts.

Modern views of machine learning for precision psychiatry.

In light of the National Institute of Mental Health (NIMH)'s Research Domain Criteria (RDoC), the advent of functional neuroimaging, novel technologies and methods provide new opportunities to develop precise and personalized prognosis and diagnosis of mental disorders. Machine learning (ML) and artificial intelligence (AI) technologies are playing an increasingly critical role in the new era of precision psychiatry. Combining ML/AI with neuromodulation technologies can potentially provide explainable solutions in clinical practice and effective therapeutic treatment. Advanced wearable and mobile technologies also call for the new role of ML/AI for digital phenotyping in mobile mental health. In this review, we provide a comprehensive review of ML methodologies and applications by combining neuroimaging, neuromodulation, and advanced mobile technologies in psychiatry practice. We further review the role of ML in molecular phenotyping and cross-species biomarker identification in precision psychiatry. We also discuss explainable AI (XAI) and neuromodulation in a closed human-in-the-loop manner and highlight the ML potential in multi-media information extraction and multi-modal data fusion. Finally, we discuss conceptual and practical challenges in precision psychiatry and highlight ML opportunities in future research.

Associations among civilian mild traumatic brain injury with loss of consciousness, posttraumatic stress disorder symptom trajectories, and structural brain volumetric data.

Posttraumatic stress disorder (PTSD) is prevalent and associated with significant morbidity. Mild traumatic brain injury (mTBI) concurrent with psychiatric trauma may be associated with PTSD. Prior studies of PTSD-related structural brain alterations have focused on military populations. The current study examined correlations between PTSD, acute mTBI, and structural brain alterations longitudinally in civilian patients (N = 504) who experienced a recent Criterion A traumatic event. Participants who reported loss of consciousness (LOC) were characterized as having mTBI; all others were included in the control group. PTSD symptoms were assessed at enrollment and over the following year; a subset of participants (n = 89) underwent volumetric brain MRI (M = 53 days posttrauma). Classes of PTSD symptom trajectories were modeled using latent growth mixture modeling. Associations between PTSD symptom trajectories and cortical thicknesses or subcortical volumes were assessed using a moderator-based regression. mTBI with LOC during trauma was positively correlated with the likelihood of developing a chronic PTSD symptom trajectory. mTBI showed significant interactions with cortical thickness in the rostral anterior cingulate cortex (rACC) in predicting PTSD symptoms, r = .461-.463. Bilateral rACC thickness positively predicted PTSD symptoms but only among participants who endorsed LOC, p < .001. The results demonstrate positive correlations between mTBI with LOC and PTSD symptom trajectories, and findings related to mTBI with LOC and rACC thickness interactions in predicting subsequent chronic PTSD symptoms suggest the importance of further understanding the role of mTBI in the context of PTSD to inform intervention and risk stratification.

Predicting non-response to multimodal day clinic treatment in severely impaired depressed patients: a machine learning approach.

A considerable number of depressed patients do not respond to treatment. Accurate prediction of non-response to routine clinical care may help in treatment planning and improve results. A longitudinal sample of N = 239 depressed patients was assessed at admission to multi-modal day clinic treatment, after six weeks, and at discharge. First, patient's treatment response was modelled by identifying longitudinal trajectories using the Hamilton Depression Rating Scale (HDRS-17). Then, individual items of the HDRS-17 at admission as well as individual patient characteristics were entered as predictors of response/non-response trajectories into the binary classification model (eXtremeGradient Boosting; XGBoost). The model was evaluated on a hold-out set and explained in human-interpretable form by SHapley Additive explanation (SHAP) values. The prediction model yielded a multi-class AUC = 0.80 in the hold-out set. The predictive power for the binary classification yielded an AUC = 0.83 (sensitivity = .80, specificity = .77). Most relevant predictors for non-response were insomnia symptoms, younger age, anxiety symptoms, depressed mood, being unemployed, suicidal ideation and somatic symptoms of depressive disorder. Non-responders to routine treatment for depression can be identified and screened for potential next-generation treatments. Such predictors may help personalize treatment and improve treatment response.

Pre-trauma predictors of severe psychiatric comorbidity 5 years following traumatic experiences.

BACKGROUND: A minority of persons who have traumatic experiences go on to develop post-traumatic stress disorder (PTSD), leading to interest in who is at risk for psychopathology after these experiences. Complicating this effort is the observation that post-traumatic psychopathology is heterogeneous. The goal of this nested case-control study was to identify pre-trauma predictors of severe post-traumatic psychiatric comorbidity, using data from Danish registries. METHODS: The source population for this study was the population of Denmark from 1994 through 2016. Cases had received three or more psychiatric diagnoses (across all ICD-10 categories) within 5 years of a traumatic experience (n = 20 361); controls were sampled from the parent cohort using risk-set sampling (n = 81 444). Analyses were repeated in samples stratified by pre-trauma psychiatric diagnoses. We used machine learning methods (classification and regression trees and random forest) to determine the important predictors of severe post-trauma psychiatric comorbidity from among hundreds of pre-trauma predictor variables spanning demographic and social variables, psychiatric and somatic diagnoses and filled medication prescriptions. RESULTS: In the full sample, pre-trauma psychiatric diagnoses (e.g. stress disorders, alcohol-related disorders, personality disorders) were the most important predictors of severe post-trauma psychiatric comorbidity. Among persons with no pre-trauma psychiatric diagnoses, demographic and social variables (e.g. marital status), type of trauma, medications used primarily to treat psychiatric symptomatology, anti-inflammatory medications and gastrointestinal distress were important to prediction. Results among persons with pre-trauma psychiatric diagnoses were consistent with the overall sample. CONCLUSIONS: This study builds on the understanding of pre-trauma factors that predict psychopathology following traumatic experiences, by examining a broad range of predictors of post-trauma psychopathology and comorbidity beyond PTSD.

Facial and Vocal Markers of Schizophrenia Measured Using Remote Smartphone Assessments: Observational Study.

BACKGROUND: Machine learning-based facial and vocal measurements have demonstrated relationships with schizophrenia diagnosis and severity. Demonstrating utility and validity of remote and automated assessments conducted outside of controlled experimental or clinical settings can facilitate scaling such measurement tools to aid in risk assessment and tracking of treatment response in populations that are difficult to engage. OBJECTIVE: This study aimed to determine the accuracy of machine learning-based facial and vocal measurements acquired through automated assessments conducted remotely through smartphones. METHODS: Measurements of facial and vocal characteristics including facial expressivity, vocal acoustics, and speech prevalence were assessed in 20 patients with schizophrenia over the course of 2 weeks in response to two classes of prompts previously utilized in experimental laboratory assessments: evoked prompts, where subjects are guided to produce specific facial expressions and speech; and spontaneous prompts, where subjects are presented stimuli in the form of emotionally evocative imagery and asked to freely respond. Facial and vocal measurements were assessed in relation to schizophrenia symptom severity using the Positive and Negative Syndrome Scale. RESULTS: Vocal markers including speech prevalence, vocal jitter, fundamental frequency, and vocal intensity demonstrated specificity as markers of negative symptom severity, while measurement of facial expressivity demonstrated itself as a robust marker of overall schizophrenia symptom severity. CONCLUSIONS: Established facial and vocal measurements, collected remotely in schizophrenia patients via smartphones in response to automated task prompts, demonstrated accuracy as markers of schizophrenia symptom severity. Clinical implications are discussed.

Using Danish national registry data to understand psychopathology following potentially traumatic experiences.

Research on posttraumatic psychopathology has focused primarily on posttraumatic stress disorder (PTSD); other posttraumatic psychiatric diagnoses are less well documented. The present study aimed to (a) develop a methodology to derive a cohort of individuals who experienced potentially traumatic events (PTEs) from registry-based data and (b) examine the risk of psychopathology within 5 years of experiencing a PTE. Using data from Danish national registries, we created a cohort of individuals with no age restrictions (range: 0-108 years) who experienced at least one of eight possible PTEs between 1994 and 2016 (N = 1,406,637). We calculated the 5-year incidence of nine categories of ICD-10 psychiatric disorders among this cohort and examined standardized morbidity ratios (SMRs) comparing the incidence of psychopathology in this group to the incidence in a nontraumatic stressor cohort (i.e., nonsuicide death of a relative; n = 423,270). Stress disorders (2.5%), substance use disorders (4.1%), and depressive disorders (3.0%) were the most common diagnoses following PTEs. Overall, the SMRs for the associations between any PTE and psychopathology varied from 1.9, 95% CI [1.9, 2.0], for stress disorders to 5.2, 95% CI [5.1. 5.3], for personality disorders. All PTEs except pregnancy-related trauma were associated with all forms of psychopathology. Associations were consistent regardless of whether a stress disorder was present. Traumatic experiences have a broad impact on psychiatric health. The present findings demonstrate one approach to capturing trauma exposure in medical record registry data. Increased traumatic experience characterization across studies will help improve the field's understanding of posttraumatic psychopathology.

Deep learning-based classification of posttraumatic stress disorder and depression following trauma utilizing visual and auditory markers of arousal and mood.

BACKGROUND: Visual and auditory signs of patient functioning have long been used for clinical diagnosis, treatment selection, and prognosis. Direct measurement and quantification of these signals can aim to improve the consistency, sensitivity, and scalability of clinical assessment. Currently, we investigate if machine learning-based computer vision (CV), semantic, and acoustic analysis can capture clinical features from free speech responses to a brief interview 1 month post-trauma that accurately classify major depressive disorder (MDD) and posttraumatic stress disorder (PTSD). METHODS: N = 81 patients admitted to an emergency department (ED) of a Level-1 Trauma Unit following a life-threatening traumatic event participated in an open-ended qualitative interview with a para-professional about their experience 1 month following admission. A deep neural network was utilized to extract facial features of emotion and their intensity, movement parameters, speech prosody, and natural language content. These features were utilized as inputs to classify PTSD and MDD cross-sectionally. RESULTS: Both video- and audio-based markers contributed to good discriminatory classification accuracy. The algorithm discriminates PTSD status at 1 month after ED admission with an AUC of 0.90 (weighted average precision = 0.83, recall = 0.84, and f1-score = 0.83) as well as depression status at 1 month after ED admission with an AUC of 0.86 (weighted average precision = 0.83, recall = 0.82, and f1-score = 0.82). CONCLUSIONS: Direct clinical observation during post-trauma free speech using deep learning identifies digital markers that can be utilized to classify MDD and PTSD status.

Remote Digital Measurement of Facial and Vocal Markers of Major Depressive Disorder Severity and Treatment Response: A Pilot Study.

Objectives: Multiple machine learning-based visual and auditory digital markers have demonstrated associations between major depressive disorder (MDD) status and severity. The current study examines if such measurements can quantify response to antidepressant treatment (ADT) with selective serotonin reuptake inhibitors (SSRIs) and serotonin-norepinephrine uptake inhibitors (SNRIs). Methods: Visual and auditory markers were acquired through an automated smartphone task that measures facial, vocal, and head movement characteristics across 4 weeks of treatment (with time points at baseline, 2 weeks, and 4 weeks) on ADT (n = 18). MDD diagnosis was confirmed using the Mini-International Neuropsychiatric Interview (MINI), and the Montgomery-Åsberg Depression Rating Scale (MADRS) was collected concordantly to assess changes in MDD severity. Results: Patient responses to ADT demonstrated clinically and statistically significant changes in the MADRS [F (2, 34) = 51.62, p < 0.0001]. Additionally, patients demonstrated significant increases in multiple digital markers including facial expressivity, head movement, and amount of speech. Finally, patients demonstrated significantly decreased frequency of fear and anger facial expressions. Conclusion: Digital markers associated with MDD demonstrate validity as measures of treatment response.

Sex Differences in Peritraumatic Inflammatory Cytokines and Steroid Hormones Contribute to Prospective Risk for Nonremitting Posttraumatic Stress Disorder.

Women are at higher risk for developing posttraumatic stress disorder (PTSD) compared to men, yet little is known about the biological contributors to this sex difference. One possible mechanism is differential immunological and neuroendocrine responses to traumatic stress exposure. In the current prospective study, we aimed to identify whether sex is indirectly associated with the probability of developing nonremitting PTSD through pro-inflammatory markers and whether steroid hormone concentrations influence this effect. Female (n = 179) and male (n = 197) trauma survivors were recruited from an emergency department and completed clinical assessment within 24 h and blood samples within ∼three hours of trauma exposure. Pro-inflammatory cytokines (IL-6, IL-1 ß , TNF, IFNγ), and steroid hormone (estradiol, testosterone, progesterone, cortisol) concentrations were quantified in plasma. Compared to men, women had a higher probability of developing nonremitting PTSD after trauma (p = 0.04), had lower pro-inflammatory cytokines and testosterone (p's<0.001), and had higher cortisol and progesterone (p's<0.001) concentrations. Estradiol concentrations were not different between the sexes (p = 0.24). Pro-inflammatory cytokines were a significant mediator in the relationship between sex and probability of developing nonremitting PTSD (p < 0.05), such that men had higher concentrations of pro-inflammatory cytokines which were associated with lower risk of nonremitting PTSD development. This effect was significantly moderated by estradiol (p < 0.05), as higher estradiol levels in men were associated with higher pro-inflammatory cytokine concentrations and lower risk for developing nonremitting PTSD. The current results suggest that sex differences in the pro-inflammatory cytokine response to trauma exposure partially mediate the probability of developing nonremitting PTSD, and that the protective ability to mount an pro-inflammatory cytokine response in men may depend on higher estradiol levels in the aftermath of trauma exposure.

SIMON: A Digital Protocol to Monitor and Predict Suicidal Ideation.

Each year, more than 800,000 persons die by suicide, making it a leading cause of death worldwide. Recent innovations in information and communication technology may offer new opportunities in suicide prevention in individuals, hereby potentially reducing this number. In our project, we design digital indices based on both self-reports and passive mobile sensing and test their ability to predict suicidal ideation, a major predictor for suicide, and psychiatric hospital readmission in high-risk individuals: psychiatric patients after discharge who were admitted in the context of suicidal ideation or a suicidal attempt, or expressed suicidal ideations during their intake. Specifically, two smartphone applications -one for self-reports (SIMON-SELF) and one for passive mobile sensing (SIMON-SENSE)- are installed on participants' smartphones. SIMON-SELF uses a text-based chatbot, called Simon, to guide participants along the study protocol and to ask participants questions about suicidal ideation and relevant other psychological variables five times a day. These self-report data are collected for four consecutive weeks after study participants are discharged from the hospital. SIMON-SENSE collects behavioral variables -such as physical activity, location, and social connectedness- parallel to the first application. We aim to include 100 patients over 12 months to test whether (1) implementation of the digital protocol in such a high-risk population is feasible, and (2) if suicidal ideation and psychiatric hospital readmission can be predicted using a combination of psychological indices and passive sensor information. To this end, a predictive algorithm for suicidal ideation and psychiatric hospital readmission using various learning algorithms (e.g., random forest and support vector machines) and multilevel models will be constructed. Data collected on the basis of psychological theory and digital phenotyping may, in the future and based on our results, help reach vulnerable individuals early and provide links to just-in-time and cost-effective interventions or establish prompt mental health service contact. The current effort may thus lead to saving lives and significantly reduce economic impact by decreasing inpatient treatment and days lost to inability.

Validation of Visual and Auditory Digital Markers of Suicidality in Acutely Suicidal Psychiatric Inpatients: Proof-of-Concept Study.

BACKGROUND: Multiple symptoms of suicide risk have been assessed based on visual and auditory information, including flattened affect, reduced movement, and slowed speech. Objective quantification of such symptomatology from novel data sources can increase the sensitivity, scalability, and timeliness of suicide risk assessment. OBJECTIVE: We aimed to examine measurements extracted from video interviews using open-source deep learning algorithms to quantify facial, vocal, and movement behaviors in relation to suicide risk severity in recently admitted patients following a suicide attempt. METHODS: We utilized video to quantify facial, vocal, and movement markers associated with mood, emotion, and motor functioning from a structured clinical conversation in 20 patients admitted to a psychiatric hospital following a suicide risk attempt. Measures were calculated using open-source deep learning algorithms for processing facial expressivity, head movement, and vocal characteristics. Derived digital measures of flattened affect, reduced movement, and slowed speech were compared to suicide risk with the Beck Scale for Suicide Ideation controlling for age and sex, using multiple linear regression. RESULTS: Suicide severity was associated with multiple visual and auditory markers, including speech prevalence (ß=-0.68, P=.02, r2=0.40), overall expressivity (ß=-0.46, P=.10, r2=0.27), and head movement measured as head pitch variability (ß=-1.24, P=.006, r2=0.48) and head yaw variability (ß=-0.54, P=.06, r2=0.32). CONCLUSIONS: Digital measurements of facial affect, movement, and speech prevalence demonstrated strong effect sizes and linear associations with the severity of suicidal ideation.

Transcriptome-wide association study of post-trauma symptom trajectories identified GRIN3B as a potential biomarker for PTSD development.

Biomarkers that predict symptom trajectories after trauma can facilitate early detection or intervention for posttraumatic stress disorder (PTSD) and may also advance our understanding of its biology. Here, we aimed to identify trajectory-based biomarkers using blood transcriptomes collected in the immediate aftermath of trauma exposure. Participants were recruited from an Emergency Department in the immediate aftermath of trauma exposure and assessed for PTSD symptoms at baseline, 1, 3, 6, and 12 months. Three empirical symptom trajectories (chronic-PTSD, remitting, and resilient) were identified in 377 individuals based on longitudinal symptoms across four data points (1, 3, 6, and 12 months), using latent growth mixture modeling. Blood transcriptomes were examined for association with longitudinal symptom trajectories, followed by expression quantitative trait locus analysis. GRIN3B and AMOTL1 blood mRNA levels were associated with chronic vs. resilient post-trauma symptom trajectories at a transcriptome-wide significant level (N = 153, FDR-corrected p value = 0.0063 and 0.0253, respectively). We identified four genetic variants that regulate mRNA blood expression levels of GRIN3B. Among these, GRIN3B rs10401454 was associated with PTSD in an independent dataset (N = 3521, p = 0.04). Examination of the BrainCloud and GTEx databases revealed that rs10401454 was associated with brain mRNA expression levels of GRIN3B. While further replication and validation studies are needed, our data suggest that GRIN3B, a glutamate ionotropic receptor NMDA type subunit-3B, may be involved in the manifestation of PTSD. In addition, the blood mRNA level of GRIN3B may be a promising early biomarker for the PTSD manifestation and development.

Predicting Sex-Specific Nonfatal Suicide Attempt Risk Using Machine Learning and Data From Danish National Registries.

Suicide attempts are a leading cause of injury globally. Accurate prediction of suicide attempts might offer opportunities for prevention. This case-cohort study used machine learning to examine sex-specific risk profiles for suicide attempts in Danish nationwide registry data. Cases were all persons who made a nonfatal suicide attempt between 1995 and 2015 (n = 22,974); the subcohort was a 5% random sample of the population at risk on January 1, 1995 (n = 265,183). We developed sex-stratified classification trees and random forests using 1,458 predictors, including demographic factors, family histories, psychiatric and physical health diagnoses, surgery, and prescribed medications. We found that substance use disorders/treatment, prescribed psychiatric medications, previous poisoning diagnoses, and stress disorders were important factors for predicting suicide attempts among men and women. Individuals in the top 5% of predicted risk accounted for 44.7% of all suicide attempts among men and 43.2% of all attempts among women. Our findings illuminate novel risk factors and interactions that are most predictive of nonfatal suicide attempts, while consistency between our findings and previous work in this area adds to the call to move machine learning suicide research toward the examination of high-risk subpopulations.

Discriminating Heterogeneous Trajectories of Resilience and Depression After Major Life Stressors Using Polygenic Scores.

Importance: Major life stressors, such as loss and trauma, increase the risk of depression. It is known that individuals show heterogeneous trajectories of depressive symptoms following major life stressors, including chronic depression, recovery, and resilience. Although common genetic variation has been associated with depression risk, genomic factors that could help discriminate trajectories of risk vs resilience following adversity have not been identified. Objective: To assess the discriminatory accuracy of a deep neural net combining joint information from 21 psychiatric and health-related multiple polygenic scores (PGSs) for discriminating resilience vs other longitudinal symptom trajectories with use of longitudinal, genetically informed data on adults exposed to major life stressors. Design, Setting, and Participants: The Health and Retirement Study is a longitudinal panel cohort study in US citizens older than 50 years, with data being collected once every 2 years between 1992 and 2010. A total of 2071 participants who were of European ancestry with available depressive symptom trajectory information after experiencing an index depressogenic major life stressor were included. Latent growth mixture modeling identified heterogeneous trajectories of depressive symptoms before and after major life stressors, including stable low symptoms (ie, resilience), as well as improving, emergent, and preexisting/chronic symptom patterns. Twenty-one PGSs were examined as factors distinctively associated with these heterogeneous trajectories. Local interpretable model-agnostic explanations were applied to examine PGSs associated with each trajectory. Data were analyzed using the DNN model from June to July 2020. Exposures: Development of depression and resilience were examined in older adults after a major life stressor, such as bereavement, divorce, and job loss, or major health events, such as myocardial infarction and cancer. Main Outcomes and Measures: Discriminatory accuracy of a deep neural net model trained for the multinomial classification of 4 distinct trajectories of depressive symptoms (Center for Epidemiologic Studies-Depression scale) based on 21 PGSs using supervised machine learning. Results: Of the 2071 participants, 1329 were women (64.2%); mean (SD) age was 55.96 (8.52) years. Of these, 1638 (79.1%) were classified as resilient, 160 (7.75) in recovery (improving), 159 (7.7%) with emerging depression, and 114 (5.5%) with preexisting/chronic depression symptoms. Deep neural nets distinguished these 4 trajectories with high discriminatory accuracy (multiclass micro-average area under the curve, 0.88; 95% CI, 0.87-0.89; multiclass macro-average area under the curve, 0.86; 95% CI, 0.85-0.87). Discriminatory accuracy was highest for preexisting/chronic depression (AUC 0.93), followed by emerging depression (AUC 0.88), recovery (AUC 0.87), resilience (AUC 0.75). Conclusions and Relevance: The results of the longitudinal cohort study suggest that multivariate PGS profiles provide information to accurately distinguish between heterogeneous stress-related risk and resilience phenotypes.

Forecasting individual risk for long-term Posttraumatic Stress Disorder in emergency medical settings using biomedical data: A machine learning multicenter cohort study.

The necessary requirement of a traumatic event preceding the development of Posttraumatic Stress Disorder, theoretically allows for administering preventive and early interventions in the early aftermath of such events. Machine learning models including biomedical data to forecast PTSD outcome after trauma are highly promising for detection of individuals most in need of such interventions. In the current study, machine learning was applied on biomedical data collected within 48 h post-trauma to forecast individual risk for long-term PTSD, using a multinominal approach including the full spectrum of common PTSD symptom courses within one prognostic model for the first time. N = 417 patients (37.2% females; mean age 46.09 ± 15.88) admitted with (suspected) serious injury to two urban Academic Level-1 Trauma Centers were included. Routinely collected biomedical information (endocrine measures, vital signs, pharmacotherapy, demographics, injury and trauma characteristics) upon ED admission and subsequent 48 h was used. Cross-validated multi-nominal classification of longitudinal self-reported symptom severity (IES-R) over 12 months and bimodal classification of clinician-rated PTSD diagnosis (CAPS-IV) at 12 months post-trauma was performed using extreme Gradient Boosting and evaluated on hold-out sets. SHapley Additive exPlanations (SHAP) values were used to explain the derived models in human-interpretable form. Good prediction of longitudinal PTSD symptom trajectories (multiclass AUC = 0.89) and clinician-rated PTSD at 12 months (AUC = 0.89) was achieved. Most relevant prognostic variables to forecast both multinominal and dichotomous PTSD outcomes included acute endocrine and psychophysiological measures and hospital-prescribed pharmacotherapy. Thus, individual risk for long-term PTSD was accurately forecasted from biomedical information routinely collected within 48 h post-trauma. These results facilitate future targeted preventive interventions by enabling future early risk detection and provide further insights into the complex etiology of PTSD.