Quantitative forecasting of PTSD from early trauma responses: a Machine Learning application

Supervised Machine Learning: A Brief Primer

Machine learning is increasingly used in mental health research and has the potential to advance our understanding of how to characterize, predict, and treat mental disorders and associated adverse health outcomes (e.g., suicidal behavior). Machine learning offers new tools to overcome challenges for which traditional statistical methods are not well-suited. This paper provides an overview of machine learning with a specific focus on supervised learning (i.e., methods that are designed to predict or classify an outcome of interest). Several common supervised learning methods are described, along with applied examples from the published literature. We also provide an overview of supervised learning model building, validation, and performance evaluation. Finally, challenges in creating robust and generalizable machine learning algorithms are discussed.

Riemannian classifier enhances the accuracy of machine-learning-based diagnosis of PTSD using resting EEG

Recently, objective and automated methods for the diagnosis of post-traumatic stress disorder (PTSD) have attracted increasing attention. However, previous studies on machine-learning-based diagnosis of PTSD with resting-state electroencephalogram (EEG) have reported poor accuracies of as low as 60%. Here, a Riemannian geometry-based classifier, the Fisher geodesic minimum distance to the mean (FgMDM), was employed for PTSD classification for the first time. Eyes-closed resting-state EEG data of 39 healthy individuals and 42 PTSD patients were used for the analysis. EEG source activities in 148 cortical regions were parcellated based on the Destrieux atlas, and their covariances were evaluated for each individual. Thirty epochs of preprocessed EEG were employed to calculate source activities. In addition, the FgMDM approach was applied to each EEG source covariance to construct the classifier. For a comparison, linear discriminant analysis (LDA), support vector machine (SVM), and random forest (RF) classifiers employing source band powers and network features as feature candidates were also tested. The FgMDM classifier showed an average classification accuracy of 75.240.80%. In contrast, the maximum accuracies of LDA, SVM, and RF classifiers were 66.54 ± 2.99%, 61.11 ± 2.98%, and 60.99 ± 2.19%, respectively. Our study demonstrated that the diagnostic accuracy of PTSD with resting-state EEG could be significantly improved by employing the FgMDM framework, which is a type of Riemannian geometry-based classifier.

Real-Time Stress Assessment Using Sliding Window Based Convolutional Neural Network

Mental stress has been identified as a significant cause of several bodily disorders, such as depression, hypertension, neural and cardiovascular abnormalities. Conventional stress assessment methods are highly subjective and tedious and tend to lack accuracy. Machine-learning (ML)-based computer-aided diagnosis systems can be used to assess the mental state with reasonable accuracy, but they require offline processing and feature extraction, rendering them unsuitable for real-time applications. This paper presents a real-time mental stress assessment approach based on convolutional neural networks (CNNs). The CNN-based approach afforded real-time mental stress assessment with an accuracy as high as 96%, the sensitivity of 95%, and specificity of 97%. The proposed approach is compared with state-of-the-art ML techniques in terms of accuracy, time utilisation, and quality of features.

Applicability of an Automated Model and Parameter Selection in the Prediction of Screening-Level PTSD in Danish Soldiers Following Deployment: Development Study of Transferable Predictive Models Using Automated Machine Learning

BACKGROUND

Posttraumatic stress disorder (PTSD) is a relatively common consequence of deployment to war zones. Early postdeployment screening with the aim of identifying those at risk for PTSD in the years following deployment will help deliver interventions to those in need but have so far proved unsuccessful.

OBJECTIVE

This study aimed to test the applicability of automated model selection and the ability of automated machine learning prediction models to transfer across cohorts and predict screening-level PTSD 2.5 years and 6.5 years after deployment.

METHODS

Automated machine learning was applied to data routinely collected 6-8 months after return from deployment from 3 different cohorts of Danish soldiers deployed to Afghanistan in 2009 (cohort 1, N=287 or N=261 depending on the timing of the outcome assessment), 2010 (cohort 2, N=352), and 2013 (cohort 3, N=232).

RESULTS

Models transferred well between cohorts. For screening-level PTSD 2.5 and 6.5 years after deployment, random forest models provided the highest accuracy as measured by area under the receiver operating characteristic curve (AUC): 2.5 years, AUC=0.77, 95% CI 0.71-0.83; 6.5 years, AUC=0.78, 95% CI 0.73-0.83. Linear models performed equally well. Military rank, hyperarousal symptoms, and total level of PTSD symptoms were highly predictive.

CONCLUSIONS

Automated machine learning provided validated models that can be readily implemented in future deployment cohorts in the Danish Defense with the aim of targeting postdeployment support interventions to those at highest risk for developing PTSD, provided the cohorts are deployed on similar missions.

A validated predictive algorithm of post-traumatic stress course following emergency department admission after a traumatic stressor

Annually, approximately 30 million patients are discharged from the emergency department (ED) after a traumatic event¹. These patients are at substantial psychiatric risk, with approximately 10-20% developing one or more disorders, including anxiety, depression or post-traumatic stress disorder (PTSD)^2-4. At present, no accurate method exists to predict the development of PTSD symptoms upon ED admission after trauma⁵. Accurate risk identification at the point of treatment by ED services is necessary to inform the targeted deployment of existing treatment^6-9 to mitigate subsequent psychopathology in high-risk populations^10,11. This work reports the development and validation of an algorithm for prediction of post-traumatic stress course over 12 months using two independently collected prospective cohorts of trauma survivors from two level 1 emergency trauma centers, which uses routinely collectible data from electronic medical records, along with brief clinical assessments of the patient's immediate stress reaction. Results demonstrate externally validated accuracy to discriminate PTSD risk with high precision. While the predictive algorithm yields useful reproducible results on two independent prospective cohorts of ED patients, future research should extend the generalizability to the broad, clinically heterogeneous ED population under conditions of routine medical care.

Multi-domain potential biomarkers for post-traumatic stress disorder (PTSD) severity in recent trauma survivors

Contemporary symptom-based diagnosis of post-traumatic stress disorder (PTSD) largely overlooks related neurobehavioral mechanisms and relies entirely on subjective interpersonal reporting. Previous studies associating biomarkers with PTSD have mostly used symptom-based diagnosis as the main outcome measure, disregarding the wide variability and richness of PTSD phenotypical features. Here, we aimed to computationally derive potential biomarkers that could efficiently differentiate PTSD subtypes among recent trauma survivors. A three-staged semi-unsupervised method ("3C") was used to firstly categorize individuals by current PTSD symptom severity, then derive clusters based on clinical features related to PTSD (e.g. anxiety and depression), and finally to classify participants' cluster membership using objective multi-domain features. A total of 256 features were extracted from psychometrics, cognitive functioning, and both structural and functional MRI data, obtained from 101 adult civilians (age = 34.80 ± 11.95; 51 females) evaluated within 1 month of trauma exposure. The features that best differentiated cluster membership were assessed by importance analysis, classification tree, and ANOVA. Results revealed that entorhinal and rostral anterior cingulate cortices volumes (structural MRI domain), in-task amygdala's functional connectivity with the insula and thalamus (functional MRI domain), executive function and cognitive flexibility (cognitive testing domain) best differentiated between two clusters associated with PTSD severity. Cross-validation established the results' robustness and consistency within this sample. The neural and cognitive potential biomarkers revealed by the 3C analytics offer objective classifiers of post-traumatic morbidity shortly following trauma. They also map onto previously documented neurobehavioral mechanisms associated with PTSD and demonstrate the usefulness of standardized and objective measurements as differentiating clinical sub-classes shortly after trauma.

Utilization of machine learning to test the impact of cognitive processing and emotion recognition on the development of PTSD following trauma exposure

BACKGROUND

Though lifetime exposure to traumatic events is significant, only a minority of individuals develops symptoms of posttraumatic stress disorder (PTSD). Post-trauma alterations in neurocognitive and affective functioning are likely to reflect changes in underlying brain networks that are predictive of PTSD. These constructs are assumed to interact in a highly complex way. The aim of this exploratory study was to apply machine learning models to investigate the contribution of these interactions on PTSD symptom development and identify measures indicative of circuit related dysfunction.

METHODS

N = 94 participants admitted to the emergency room of an inner-city hospital after trauma exposure completed a battery of neurocognitive and emotional tests 1 month after the incident. Different machine learning algorithms were applied to predict PTSD symptom severity and clusters after 3 months based.

RESULTS

Overall, model accuracy did not differ between PTSD clusters, though the importance of cognitive and emotional domains demonstrated both key differences and overlap. Alterations in higher-order executive functioning, speed of information processing, and processing of emotionally incongruent cues were the most important predictors.

CONCLUSIONS

Data-driven approaches are a powerful tool to investigate complex interactions and can enhance the mechanistic understanding of PTSD. The study identifies important relationships between cognitive processing and emotion recognition that may be valuable to predict and understand mechanisms of risk and resilience responses to trauma prospectively.

The use of machine learning techniques in trauma-related disorders: a systematic review

Establishing the diagnosis of trauma-related disorders such as Acute Stress Disorder (ASD) and Posttraumatic Stress Disorder (PTSD) have always been a challenge in clinical practice and in academic research, due to clinical and biological heterogeneity. Machine learning (ML) techniques can be applied to improve classification of disorders, to predict outcomes or to determine person-specific treatment selection. We aim to review the existing literature on the use of machine learning techniques in the assessment of subjects with ASD or PTSD. We systematically searched PubMed, Embase and Web of Science for articles published in any language up to May 2019. We found 806 abstracts and included 49 studies in our review. Most of the included studies used multiple levels of biological data to predict risk factors or to identify early symptoms related to PTSD. Other studies used ML classification techniques to distinguish individuals with ASD or PTSD from other psychiatric disorder or from trauma-exposed and healthy controls. We also found studies that attempted to define outcome profiles using clustering techniques and studies that assessed the relationship among symptoms using network analysis. Finally, we proposed a quality assessment in this review, evaluating methodological and technical features on machine learning studies. We concluded that etiologic and clinical heterogeneity of ASD/PTSD patients is suitable to machine learning techniques and a major challenge for the future is to use it in clinical practice for the benefit of patients in an individual level.

Developing algorithms to predict adult onset internalizing disorders: An ensemble learning approach

A growing literature is utilizing machine learning methods to develop psychopathology risk algorithms that can be used to inform preventive intervention. However, efforts to develop algorithms for internalizing disorder onset have been limited. The goal of this study was to utilize prospective survey data and ensemble machine learning to develop algorithms predicting adult onset internalizing disorders. The data were from Waves 1-2 of the National Epidemiological Survey on Alcohol and Related Conditions (n = 34,653). Outcomes were incident occurrence of DSM-IV generalized anxiety, panic, social phobia, depression, and mania between Waves 1-2. In total, 213 risk factors (features) were operationalized based on their presence/occurrence at the time of or before Wave 1. For each of the five internalizing disorder outcomes, super learning was used to generate a composite algorithm from several linear and non-linear classifiers (e.g., random forests, k-nearest neighbors). AUCs achieved by the cross-validated super learner ensembles were in the range of 0.76 (depression) to 0.83 (mania), and were higher than AUCs achieved by the individual algorithms. Individuals in the top 10% of super learner predicted risk accounted for 37.97% (depression) to 53.39% (social anxiety) of all incident cases. Thus, the algorithms achieved acceptable-to-excellent prediction accuracy with a high concentration of incident cases observed among individuals predicted to be highest risk. In parallel with the development of effective preventive interventions, further validation, expansion, and dissemination of algorithms predicting internalizing disorder onset/trajectory could be of great value.

A decennial review of psychotraumatology: what did we learn and where are we going?

On 6 December 2019 we start the 10th year of the European Journal of Psychotraumatogy (EJPT), a full Open Access journal on psychotrauma. This editorial is part of a special issue/collection celebrating the 10 years anniversary of the journal where we will acknowledge some of our most impactful articles of the past decade (also discussed below and marked with * in the reference list). In this editorial the editors present a decennial review of the field addressing a range of topics that are core to both the journal and to psychotraumatology as a discipline. These include neurobiological developments (genomics, neuroimaging and neuroendocrine research), forms of trauma exposure and impact across the lifespan, mass trauma and early interventions, work-related trauma, trauma in refugee populations, and the potential consequences of trauma such as PTSD or Complex PTSD, but also resilience. We address innovations in psychological, medication (enhanced) and technology-assisted treatments, mediators and moderators like social support and finally how new research methods help us to gain insights in symptom structures or to better predict symptom development or treatment success. We aimed to answer three questions 1. Where did we stand in 2010? 2. What did we learn in the past 10 years? 3. What are our knowledge gaps? We conclude with a number of recommendations concerning top priorities for the future direction of the field of psychotraumatology and correspondingly the journal.

Screening and treating hospitalized trauma survivors for posttraumatic stress disorder and depression

Traumatic injury affects over 2.6 million U.S. adults annually and elevates risk for a number of negative health consequences. This includes substantial psychological harm, the most prominent being posttraumatic stress disorder (PTSD), with approximately 21% of traumatic injury survivors developing the disorder within the first year after injury. Posttraumatic stress disorder is associated with deficits in physical recovery, social functioning, and quality of life. Depression is diagnosed in approximately 6% in the year after injury and is also a predictor of poor quality of life. The American College of Surgeons Committee on Trauma suggests screening for and treatment of PTSD and depression, reflecting a growing awareness of the critical need to address patients' mental health needs after trauma. While some trauma centers have implemented screening and treatment or referral for treatment programs, the majority are evaluating how to best address this recommendation, and no standard approach for screening and treatment currently exists. Further, guidelines are not yet available with respect to resources that may be used to effectively screen and treat these disorders in trauma survivors, as well as who is going to bear the costs. The purpose of this review is: (1) to evaluate the current state of the literature regarding evidence-based screens for PTSD and depression in the hospitalized trauma patient and (2) summarize the literature to date regarding the treatments that have empirical support in treating PTSD and depression acutely after injury. This review also includes structural and funding information regarding existing postinjury mental health programs. Screening of injured patients and timely intervention to prevent or treat PTSD and depression could substantially improve health outcomes and improve quality of life for this high-risk population. LEVEL OF EVIDENCE: Review, level IV.

Predicting Posttraumatic Stress Disorder Risk: A Machine Learning Approach

BACKGROUND

A majority of adults in the United States are exposed to a potentially traumatic event but only a handful go on to develop impairing mental health conditions such as posttraumatic stress disorder (PTSD).

OBJECTIVE

Identifying those at elevated risk shortly after trauma exposure is a clinical challenge. The aim of this study was to develop computational methods to more effectively identify at-risk patients and, thereby, support better early interventions.

METHODS

We proposed machine learning (ML) induction of models to automatically predict elevated PTSD symptoms in patients 1 month after a trauma, using self-reported symptoms from data collected via smartphones.

RESULTS

We show that an ensemble model accurately predicts elevated PTSD symptoms, with an area under the curve (AUC) of .85, using a bag of support vector machines, naive Bayes, logistic regression, and random forest algorithms. Furthermore, we show that only 7 self-reported items (features) are needed to obtain this AUC. Most importantly, we show that accurate predictions can be made 10 to 20 days posttrauma.

CONCLUSIONS

These results suggest that simple smartphone-based patient surveys, coupled with automated analysis using ML-trained models, can identify those at risk for developing elevated PTSD symptoms and thus target them for early intervention.

The replicability of ICD-11 complex post-traumatic stress disorder symptom networks in adults

BACKGROUND

The ICD-11 includes a new disorder, complex post-traumatic stress disorder (CPTSD). A network approach to CPTSD will enable investigation of the structure of the disorder at the symptom level, which may inform the development of treatments that target specific symptoms to accelerate clinical outcomes.AimsWe aimed to test whether similar networks of ICD-11 CPTSD replicate across culturally different samples and to investigate possible differences, using a network analysis.

METHOD

We investigated the network models of four nationally representative, community-based cross-sectional samples drawn from Germany, Israel, the UK, and the USA (total N = 6417). CPTSD symptoms were assessed with the International Trauma Questionnaire in all samples. Only those participants who reported significant functional impairment by CPTSD symptoms were included (N = 1591 included in analysis; mean age 43.55 years, s.d. 15.10, range 14-99, 67.7% women). Regularised partial correlation networks were estimated for each sample and the resulting networks were compared.

RESULTS

Despite differences in traumatic experiences, symptom severity and symptom profiles, the networks were very similar across the four countries. The symptoms within dimensions were strongly associated with each other in all networks, except for the two symptom indicators assessing aspects of affective dysregulation. The most central symptoms were 'feelings of worthlessness' and 'exaggerated startle response'.

CONCLUSIONS

The structure of CPTSD symptoms appears very similar across countries. Addressing symptoms with the strongest associations in the network, such as negative self-worth and startle reactivity, will likely result in rapid treatment response.Declaration of interestA.M. and M.C. were members of the World Health Organization (WHO) ICD-11 Working Group on the Classification of Disorders Specifically Associated with Stress, reporting to the WHO International Advisory Group for the Revision of ICD-10 Mental and Behavioural Disorders. The views expressed in this article are those of the authors and do not represent the official policies or positions of the International Advisory Group or the WHO.

Machine Learning for Prediction of Posttraumatic Stress and Resilience Following Trauma: An Overview of Basic Concepts and Recent Advances

Posttraumatic stress responses are characterized by a heterogeneity in clinical appearance and etiology. This heterogeneity impacts the field's ability to characterize, predict, and remediate maladaptive responses to trauma. Machine learning (ML) approaches are increasingly utilized to overcome this foundational problem in characterization, prediction, and treatment selection across branches of medicine that have struggled with similar clinical realities of heterogeneity in etiology and outcome, such as oncology. In this article, we review and evaluate ML approaches and applications utilized in the areas of posttraumatic stress, stress pathology, and resilience research, and present didactic information and examples to aid researchers interested in the relevance of ML to their own research. The examined studies exemplify the high potential of ML approaches to build accurate predictive and diagnostic models of posttraumatic stress and stress pathology risk based on diverse sources of available information. The use of ML approaches to integrate high-dimensional data demonstrates substantial gains in risk prediction even when the sources of data are the same as those used in traditional predictive models. This area of research will greatly benefit from collaboration and data sharing among researchers of posttraumatic stress disorder, stress pathology, and resilience.

Machine-learning prediction of adolescent alcohol use: a cross-study, cross-cultural validation

BACKGROUND AND AIMS

The experience of alcohol use among adolescents is complex, with international differences in age of purchase and individual differences in consumption and consequences. This latter underlines the importance of prediction modeling of adolescent alcohol use. The current study (a) compared the performance of seven machine-learning algorithms to predict different levels of alcohol use in mid-adolescence and (b) used a cross-cultural cross-study scheme in the training-validation-test process to display the predictive power of the best performing machine-learning algorithm.

DESIGN

A comparison of seven machine-learning algorithms: logistic regression, support vector machines, random forest, neural network, lasso regression, ridge regression and elastic-net.

SETTING

Canada and Australia.

PARTICIPANTS

The Canadian sample is part of a 4-year follow-up (2012-16) of the Co-Venture cohort (n = 3826, baseline age 12.8 ± 0.4, 49.2% girls). The Australian sample is part of a 3-year follow-up (2012-15) of the Climate Schools and Preventure (CAP) cohort (n = 2190, baseline age 13.3 ± 0.3, 43.7% girls).

MEASUREMENTS

The algorithms used several prediction indices, such as F₁ prediction score, accuracy, precision, recall, negative predictive value and area under the curve (AUC).

FINDINGS

Based on prediction indices, the elastic-net machine-learning algorithm showed the best predictive performance in both Canadian (AUC = 0.869 ± 0.066) and Australian (AUC = 0.855 ± 0.072) samples. Domain contribution analysis showed that the highest prediction accuracy indices yielded from models with only psychopathology (AUC = 0.816 ± 0.044/0.790 ± 0.071 in Canada/Australia) and only personality clusters (AUC = 0.776 ± 0.063/0.796 ± 0.066 in Canada/Australia). Similarly, regardless of the level of alcohol use, in both samples, externalizing psychopathologies, alcohol use at baseline and the sensation-seeking personality profile contributed to the prediction.

CONCLUSIONS

Computerized screening software shows promise in predicting the risk of alcohol use among adolescents.

Childhood Trauma in Schizophrenia: Current Findings and Research Perspectives

Schizophrenia is a severe neuropsychiatric disorder with persistence of symptoms throughout adult life in most of the affected patients. This unfavorable course is associated with multiple episodes and residual symptoms, mainly negative symptoms and cognitive deficits. The neural diathesis-stress model proposes that psychosocial stress acts on a pre-existing vulnerability and thus triggers the symptoms of schizophrenia. Childhood trauma is a severe form of stress that renders individuals more vulnerable to developing schizophrenia; neurobiological effects of such trauma on the endocrine system and epigenetic mechanisms are discussed. Childhood trauma is associated with impaired working memory, executive function, verbal learning, and attention in schizophrenia patients, including those at ultra-high risk to develop psychosis. In these patients, higher levels of childhood trauma were correlated with higher levels of attenuated positive symptoms, general symptoms, and depressive symptoms; lower levels of global functioning; and poorer cognitive performance in visual episodic memory end executive functions. In this review, we discuss effects of specific gene variants that interact with childhood trauma in patients with schizophrenia and describe new findings on the brain structural and functional level. Additive effects between childhood trauma and brain-derived neurotrophic factor methionine carriers on volume loss of the hippocampal subregions cornu ammonis (CA)4/dentate gyrus and CA2/3 have been reported in schizophrenia patients. A functional magnetic resonance imaging study showed that childhood trauma exposure resulted in aberrant function of parietal areas involved in working memory and of visual cortical areas involved in attention. In a theory of mind task reflecting social cognition, childhood trauma was associated with activation of the posterior cingulate gyrus, precuneus, and dorsomedial prefrontal cortex in patients with schizophrenia. In addition, decreased connectivity was shown between the posterior cingulate/precuneus region and the amygdala in patients with high levels of physical neglect and sexual abuse during childhood, suggesting that disturbances in specific brain networks underlie cognitive abilities. Finally, we discuss some of the questionnaires that are commonly used to assess childhood trauma and outline possibilities to use recent biostatistical methods, such as machine learning, to analyze the resulting datasets.

Modelling PTSD diagnosis using sleep, memory, and adrenergic metabolites: An exploratory machine-learning study

OBJECTIVE

Features of posttraumatic stress disorder (PTSD) typically include sleep disturbances, impaired declarative memory, and hyperarousal. This study evaluated whether these combined features may accurately delineate pathophysiological changes associated with PTSD.

METHOD

We recruited a cohort of PTSD-diagnosed individuals (N = 20), trauma survivors without PTSD (TE; N = 20), and healthy controls (HC; N = 20). Analyses of between-group differences and support vector machine (SVM)-learning were applied to participant features.

RESULTS

Analyses of between-group differences replicated previous findings, indicating that PTSD-diagnosed individuals self-reported poorer sleep quality, objectively demonstrated less sleep depth, and evidenced declarative memory deficits in comparison to HC. Integrative SVM-learning distinguished HC from trauma participants with 80% accuracy using a combination of five features, including subjective and objective sleep, neutral declarative memory, and metabolite variables. PTSD and TE participants could be distinguished with 70% accuracy using a combination of subjective and objective sleep variables but not by metabolite or declarative memory variables.

CONCLUSION

From among a broad range of sleep, cognitive, and biochemical variables, sleep characteristics were the primary features that could differentiate those with PTSD from those without. Our exploratory SVM-learning analysis establishes a framework for future sleep- and memory-based PTSD investigations that could drive improvements in diagnostic accuracy and treatment.

Investigation of the impact of total sleep deprivation at home on the number of intrusive memories to an analogue trauma

Sleep enhances the consolidation of memory; however, this property of sleep may be detrimental in situations where memories of an event can lead to psychopathology, such as following a traumatic event. Intrusive memories of trauma are emotional memories that spring to mind involuntarily and are a core feature of post-traumatic stress disorder. Total sleep deprivation in a hospital setting on the first night after an analogue trauma (a trauma film) led to fewer intrusive memories compared to sleep as usual in one study. The current study aimed to test an extension of these findings: sleep deprivation under more naturalistic conditions-at home. Polysomnographic recordings show inconsistent sleep deprivation was achieved at home. Fewer intrusive memories were reported on day 1 after the trauma film in the sleep-deprived condition. On day 2 the opposite was found: more intrusive memories in the sleep-deprived condition. However, no significant differences were found with the removal of two participants with extreme values and no difference was found in the total number of intrusive memories reported in the week following the trauma film. Voluntary memory of the trauma film was found to be slightly impaired in the sleep deprivation condition. In conclusion, compared to our eariler findings using total sleep deprivation in a hospital setting, in the current study the use of inconsistent sleep deprivation at home does not replicate the pattern of results on reducing the number of intrusive memories. Considering the conditions under which sleep deprivation (naturalistic versus hospital) was achieved requires further examination.

Sex differences in global and local connectivity of adolescent posttraumatic stress disorder symptoms

BACKGROUND

Sex differences in youth's posttraumatic stress disorder (PTSD) symptomatology have not been well studied.

METHODS

Based on a recently burgeoning theory of psychopathology networks, this study conducted sex comparisons of global and local connectivity of PTSD symptoms in a sample of 868 disaster-exposed adolescents (57.0% girls; a mean age of 13.4 ± 0.8 years) with significant PTSD symptomatology evaluated by the UCLA PTSD Reaction Index for DSM-IV.

RESULTS

The results revealed that global connectivity was stronger in girls' network than in boys', and individual symptoms' connectivity and its rankings differed by sex. Intrusive recollections, flashbacks, avoiding activities/people, and detachment were the most strongly connected symptoms in girls, whereas flashbacks, physiological cue reactivity, diminished interest, and foreshortened future were the most strongly connected symptoms in boys. Several symptoms were identified as featuring large connectivity differences across sex.

CONCLUSIONS

These findings provide novel insights into sex differential risk and features of youth's PTSD symptomatology. Sex differences reflected in the co-occurrence of PTSD symptoms may merit more consideration in research and clinical practice.

Posttraumatic stress disorder: a model of the longitudinal course and the role of risk factors

Posttraumatic stress disorder (PTSD) differs from other anxiety disorders in that experience of a traumatic event is necessary for the onset of the disorder. The condition runs a longitudinal course, involving a series of transitional states, with progressive modification occurring with time. Notably, only a small percentage of people that experience trauma will develop PTSD. Risk factors, such as prior trauma, prior psychiatric history, family psychiatric history, peritraumatic dissociation, acute stress symptoms, the nature of the biological response, and autonomic hyperarousal, need to be considered when setting up models to predict the course of the condition. These risk factors influence vulnerability to the onset of PTSD and its spontaneous remission. In the majority of cases, PTSD is accompanied by another condition, such as major depression, an anxiety disorder, or substance abuse. This comorbidity can also complicate the course of the disorder and raises questions about the role of PTSD in other psychiatric conditions. This article reviews what is known about the emergence of PTSD following exposure to a traumatic event using data from clinical studies.

Ensemble machine learning prediction of posttraumatic stress disorder screening status after emergency room hospitalization

Posttraumatic stress disorder (PTSD) develops in a substantial minority of emergency room admits. Inexpensive and accurate person-level assessment of PTSD risk after trauma exposure is a critical precursor to large-scale deployment of early interventions that may reduce individual suffering and societal costs. Toward this aim, we applied ensemble machine learning to predict PTSD screening status three months after severe injury using cost-effective and minimally invasive data. Participants (N = 271) were recruited at a Level 1 Trauma Center where they provided variables routinely collected at the hospital, including pulse, injury severity, and demographics, as well as psychological variables, including self-reported current depression, psychiatric history, and social support. Participant zip codes were used to extract contextual variables including population total and density, average annual income, and health insurance coverage rates from publicly available U.S. Census data. Machine learning yielded good prediction of PTSD screening status 3 months post-hospitalization, AUC = 0.85 95% CI [0.83, 0.86], and significantly outperformed all benchmark comparison models in a cross-validation procedure designed to yield an unbiased estimate of performance. These results demonstrate that good prediction can be attained from variables that individually have relatively weak predictive value, pointing to the promise of ensemble machine learning approaches that do not rely on strong isolated risk factors.

A Novel Machine Learning Algorithm to Automatically Predict Visual Outcomes in Intravitreal Ranibizumab-Treated Patients with Diabetic Macular Edema

Purpose: Artificial neural networks (ANNs) are one type of artificial intelligence. Here, we use an ANN-based machine learning algorithm to automatically predict visual outcomes after ranibizumab treatment in diabetic macular edema. Methods: Patient data were used to optimize ANNs for regression calculation. The target was established as the final visual acuity at 52, 78, or 104 weeks. The input baseline variables were sex, age, diabetes type or condition, systemic diseases, eye status and treatment time tables. Three groups were randomly devised to build, test and demonstrate the accuracy of the algorithms. Results: At 52, 78 and 104 weeks, 512, 483 and 464 eyes were included, respectively. For the training group, testing group and validation group, the respective correlation coefficients were 0.75, 0.77 and 0.70 (52 weeks); 0.79, 0.80 and 0.55 (78 weeks); and 0.83, 0.47 and 0.81 (104 weeks), while the mean standard errors of final visual acuity were 6.50, 6.11 and 6.40 (52 weeks); 5.91, 5.83 and 7.59; (78 weeks); and 5.39, 8.70 and 6.81 (104 weeks). Conclusions: Machine learning had good correlation coefficients for predicating prognosis with ranibizumab with just baseline characteristics. These models could be the useful clinical tools for prediction of success of the treatments.

An exhaustive review and analysis on applications of statistical forecasting in hospital emergency departments

Emergency departments (EDs) provide medical treatment for a broad spectrum of illnesses and injuries to patients who arrive at all hours of the day. The quality and efficient delivery of health care in EDs are associated with a number of factors, such as patient overall length of stay (LOS) and admission, prompt ambulance diversion, quick and accurate triage, nurse and physician assessment, diagnostic and laboratory services, consultations and treatment. One of the most important ways to plan the healthcare delivery efficiently is to make forecasts of ED processes. The aim this study is thus to provide an exhaustive review for ED stakeholders interested in applying forecasting methods to their ED processes. A categorisation, analysis and interpretation of 102 papers is performed for review. This exhaustive review provides an insight for researchers and practitioners about forecasting in EDs in terms of showing current state and potential areas for future attempts.

DiscMLA: An Efficient Discriminative Motif Learning Algorithm over High-Throughput Datasets

The transcription factors (TFs) can activate or suppress gene expression by binding to specific sites, hence are crucial regulatory elements for transcription. Recently, series of discriminative motif finders have been tailored to offering promising strategy for harnessing the power of large quantities of accumulated high-throughput experimental data. However, in order to achieve high speed, these algorithms have to sacrifice accuracy by employing simplified statistical models during the searching process. In this paper, we propose a novel approach named Discriminative Motif Learning via AUC (DiscMLA) to discover motifs on high-throughput datasets. Unlike previous approaches, DiscMLA tries to optimize with a more comprehensive criterion (AUC) during motifs searching. In addition, based on an experimental observation of motif identification on large-scale datasets, some novel procedures are designed to accelerate DiscMLA. The experimental results on 52 real-world datasets demonstrate that our approach substantially outperforms previous methods on discriminative motif learning problems. DiscMLA' stability, discriminability, and validity will help to exploit high-throughput datasets and answer many fundamental biological questions.

Effect of acute alarm odor exposure and biological sex on generalized avoidance and glutamatergic signaling in the hippocampus of Wistar rats

Post-traumatic stress disorder (PTSD) is characterized by the development of paradoxical memory disturbances including intrusive memories and amnesia for specific details of the traumatic experience. Despite evidence that women are at higher risk to develop PTSD, most animal research has focused on the processes by which male rodents develop adaptive fear memory. As such, the mechanisms contributing to sex differences in the development of PTSD-like memory disturbances are poorly understood. In this investigation, we exposed adult male and female Wistar rats to the synthetic alarm odor 2,4,5-trimethylthiazole (TMT) to assess development of generalized fear behavior and rapid modulation of glutamate uptake and signaling cascades associated with hippocampus-dependent long-term memory. We report that female Wistar rats exposed to alarm odor exhibit context discrimination impairments relative to TMT-exposed male rats, suggesting the intriguing possibility that females are at greater risk in developing generalized fear memories. Mechanistically, alarm odor exposure rapidly modulated signaling cascades consistent with activation of the CREB shut-off cascade in the male, but not the female hippocampus. Moreover, TMT exposure dampened glutamate uptake and affected expression of the glutamate transporter, GLT-1 in the hippocampus. Taken together, these results provide evidence for rapid sex-dependent modulation of CREB signaling in the hippocampus by alarm odor exposure which may contribute to the development of generalized fear.

Machine Learning for Predicting Outcomes in Trauma

To date, there are no reviews on machine learning (ML) for predicting outcomes in trauma. Consequently, it remains unclear as to how ML-based prediction models compare in the triage and assessment of trauma patients. The objective of this review was to survey and identify studies involving ML for predicting outcomes in trauma, with the hypothesis that models predicting similar outcomes may share common features but the performance of ML in these studies will differ greatly. MEDLINE and other databases were searched for studies involving trauma and ML. Sixty-five observational studies involving ML for the prediction of trauma outcomes met inclusion criteria. In total 2,433,180 patients were included in the studies. The studies focused on prediction of the following outcome measures: survival/mortality (n = 34), morbidity/shock/hemorrhage (n = 12), hospital length of stay (n = 7), hospital admission/triage (n = 6), traumatic brain injury (n = 4), life-saving interventions (n = 5), post-traumatic stress disorder (n = 4), and transfusion (n = 1). Six studies were prospective observational studies. Of the 65 studies, 33 used artificial neural networks for prediction. Importantly, most studies demonstrated the benefits of ML models. However, algorithm performance was assessed differently by different authors. Sensitivity-specificity gap values varied greatly from 0.035 to 0.927. Notably, studies shared many features for model development. A common ML feature base may be determined for predicting outcomes in trauma. However, the impact of ML will require further validation in prospective observational studies and randomized clinical trials, establishment of common performance criteria, and high-quality evidence about clinical and economic impacts before ML can be widely accepted in practice.

Trajectories of resilience and dysfunction following potential trauma: A review and statistical evaluation

Given the rapid proliferation of trajectory-based approaches to study clinical consequences to stress and potentially traumatic events (PTEs), there is a need to evaluate emerging findings. This review examined convergence/divergences across 54 studies in the nature and prevalence of response trajectories, and determined potential sources of bias to improve future research. Of the 67 cases that emerged from the 54 studies, the most consistently observed trajectories following PTEs were resilience (observed in: n = 63 cases), recovery (n = 49), chronic (n = 47), and delayed onset (n = 22). The resilience trajectory was the modal response across studies (average of 65.7% across populations, 95% CI [0.616, 0.698]), followed in prevalence by recovery (20.8% [0.162, 0.258]), chronicity (10.6%, [0.086, 0.127]), and delayed onset (8.9% [0.053, 0.133]). Sources of heterogeneity in estimates primarily resulted from substantive population differences rather than bias, which was observed when prospective data is lacking. Overall, prototypical trajectories have been identified across independent studies in relatively consistent proportions, with resilience being the modal response to adversity. Thus, trajectory models robustly identify clinically relevant patterns of response to potential trauma, and are important for studying determinants, consequences, and modifiers of course following potential trauma.

A Clinician's Guide to PTSD Biomarkers and Their Potential Future Use

No clinically validated biomarkers have yet been found to assist in the diagnosis and treatment of posttraumatic stress disorder (PTSD). Innovation in clinical trial design, however, has led to the study of biomarkers as part of testing new medications and psychotherapies. There may soon be viable biomarkers to assist in diagnosis of PTSD and prediction of illness trajectory, severity, and functional outcomes; subtyping; and treatment selection. Processes for the identification and validation of biomarker findings are complex, involving several stages of clinical testing before use. The authors provide an overview of issues regarding the clinical use of PTSD biomarkers and examine a set of genetic, epigenetic, and other blood-based markers along with physiological markers currently proposed as candidate tests for PTSD. Studies that have identified candidate biomarkers with relevance to treatment selection in PTSD are discussed as a promising area of research that may lead to changes in clinical practice.

Multiple memory systems, multiple time points: how science can inform treatment to control the expression of unwanted emotional memories

Memories that have strong emotions associated with them are particularly resilient to forgetting. This is not necessarily problematic, however some aspects of memory can be. In particular, the involuntary expression of those memories, e.g. intrusive memories after trauma, are core to certain psychological disorders. Since the beginning of this century, research using animal models shows that it is possible to change the underlying memory, for example by interfering with its consolidation or reconsolidation. While the idea of targeting maladaptive memories is promising for the treatment of stress and anxiety disorders, a direct application of the procedures used in non-human animals to humans in clinical settings is not straightforward. In translational research, more attention needs to be paid to specifying what aspect of memory (i) can be modified and (ii) should be modified. This requires a clear conceptualization of what aspect of memory is being targeted, and how different memory expressions may map onto clinical symptoms. Furthermore, memory processes are dynamic, so procedural details concerning timing are crucial when implementing a treatment and when assessing its effectiveness. To target emotional memory in its full complexity, including its malleability, science cannot rely on a single method, species or paradigm. Rather, a constructive dialogue is needed between multiple levels of research, all the way 'from mice to mental health'.This article is part of a discussion meeting issue 'Of mice and mental health: facilitating dialogue between basic and clinical neuroscientists'.

Preventing intrusive memories after trauma via a brief intervention involving Tetris computer game play in the emergency department: a proof-of-concept randomized controlled trial

After psychological trauma, recurrent intrusive visual memories may be distressing and disruptive. Preventive interventions post trauma are lacking. Here we test a behavioural intervention after real-life trauma derived from cognitive neuroscience. We hypothesized that intrusive memories would be significantly reduced in number by an intervention involving a computer game with high visuospatial demands (Tetris), via disrupting consolidation of sensory elements of trauma memory. The Tetris-based intervention (trauma memory reminder cue plus c. 20 min game play) vs attention-placebo control (written activity log for same duration) were both delivered in an emergency department within 6 h of a motor vehicle accident. The randomized controlled trial compared the impact on the number of intrusive trauma memories in the subsequent week (primary outcome). Results vindicated the efficacy of the Tetris-based intervention compared with the control condition: there were fewer intrusive memories overall, and time-series analyses showed that intrusion incidence declined more quickly. There were convergent findings on a measure of clinical post-trauma intrusion symptoms at 1 week, but not on other symptom clusters or at 1 month. Results of this proof-of-concept study suggest that a larger trial, powered to detect differences at 1 month, is warranted. Participants found the intervention easy, helpful and minimally distressing. By translating emerging neuroscientific insights and experimental research into the real world, we offer a promising new low-intensity psychiatric intervention that could prevent debilitating intrusive memories following trauma.

Derivation and validation of different machine-learning models in mortality prediction of trauma in motorcycle riders: a cross-sectional retrospective study in southern Taiwan

OBJECTIVES

This study aimed to build and test the models of machine learning (ML) to predict the mortality of hospitalised motorcycle riders.

SETTING

The study was conducted in a level-1 trauma centre in southern Taiwan.

PARTICIPANTS

Motorcycle riders who were hospitalised between January 2009 and December 2015 were classified into a training set (n=6306) and test set (n=946). Using the demographic information, injury characteristics and laboratory data of patients, logistic regression (LR), support vector machine (SVM) and decision tree (DT) analyses were performed to determine the mortality of individual motorcycle riders, under different conditions, using all samples or reduced samples, as well as all variables or selected features in the algorithm.

PRIMARY AND SECONDARY OUTCOME MEASURES

The predictive performance of the model was evaluated based on accuracy, sensitivity, specificity and geometric mean, and an analysis of the area under the receiver operating characteristic curves of the two different models was carried out.

RESULTS

In the training set, both LR and SVM had a significantly higher area under the receiver operating characteristic curve (AUC) than DT. No significant difference was observed in the AUC of LR and SVM, regardless of whether all samples or reduced samples and whether all variables or selected features were used. In the test set, the performance of the SVM model for all samples with selected features was better than that of all other models, with an accuracy of 98.73%, sensitivity of 86.96%, specificity of 99.02%, geometric mean of 92.79% and AUC of 0.9517, in mortality prediction.

CONCLUSION

ML can provide a feasible level of accuracy in predicting the mortality of motorcycle riders. Integration of the ML model, particularly the SVM algorithm in the trauma system, may help identify high-risk patients and, therefore, guide appropriate interventions by the clinical staff.

Predicting posttraumatic stress disorder following a natural disaster

Earthquakes are a common and deadly natural disaster, with roughly one-quarter of survivors subsequently developing posttraumatic stress disorder (PTSD). Despite progress identifying risk factors, limited research has examined how to combine variables into an optimized post-earthquake PTSD prediction tool that could be used to triage survivors to mental health services. The current study developed a post-earthquake PTSD risk score using machine learning methods designed to optimize prediction. The data were from a two-wave survey of Chileans exposed to the 8.8 magnitude earthquake that occurred in February 2010. Respondents (n = 23,907) were interviewed roughly three months prior to and again three months after the earthquake. Probable post-earthquake PTSD was assessed using the Davidson Trauma Scale. We applied super learning, an ensembling machine learning method, to develop the PTSD risk score from 67 risk factors that could be assessed within one week of earthquake occurrence. The super learner algorithm had better cross-validated performance than the 39 individual algorithms from which it was developed, including conventional logistic regression. The super learner also had a better area under the receiver operating characteristic curve (0.79) than existing post-disaster PTSD risk tools. Individuals in the top 5%, 10%, and 20% of the predicted risk distribution accounted for 17.5%, 32.2%, and 51.4% of all probable cases of PTSD, respectively. In addition to developing a risk score that could be implemented in the near future, these results more broadly support the utility of super learning to develop optimized prediction functions for mental health outcomes.

Cognitive Flexibility Predicts PTSD Symptoms: Observational and Interventional Studies

Introduction: Post-Traumatic Stress Disorder (PTSD) is a prevalent, severe and tenacious psychopathological consequence of traumatic events. Neurobehavioral mechanisms underlying PTSD pathogenesis have been identified, and may serve as risk-resilience factors during the early aftermath of trauma exposure. Longitudinally documenting the neurobehavioral dimensions of early responses to trauma may help characterize survivors at risk and inform mechanism-based interventions. We present two independent longitudinal studies that repeatedly probed clinical symptoms and neurocognitive domains in recent trauma survivors. We hypothesized that better neurocognitive functioning shortly after trauma will be associated with less severe PTSD symptoms a year later, and that an early neurocognitive intervention will improve cognitive functioning and reduce PTSD symptoms. Methods: Participants in both studies were adult survivors of traumatic events admitted to two general hospitals' emergency departments (EDs) in Israel. The studies used identical clinical and neurocognitive tools, which included assessment of PTSD symptoms and diagnosis, and a battery of neurocognitive tests. The first study evaluated 181 trauma-exposed individuals one-, six-, and 14 months following trauma exposure. The second study evaluated 97 trauma survivors 1 month after trauma exposure, randomly allocated to 30 days of web-based neurocognitive intervention (n = 50) or control tasks (n = 47), and re-evaluated all subjects three- and 6 months after trauma exposure. Results: In the first study, individuals with better cognitive flexibility at 1 month post-trauma showed significantly less severe PTSD symptoms after 13 months (p = 0.002). In the second study, the neurocognitive training group showed more improvement in cognitive flexibility post-intervention (p = 0.019), and lower PTSD symptoms 6 months post-trauma (p = 0.017), compared with controls. Intervention- induced improvement in cognitive flexibility positively correlated with clinical improvement (p = 0.002). Discussion: Cognitive flexibility, shortly after trauma exposure, emerged as a significant predictor of PTSD symptom severity. It was also ameliorated by a neurocognitive intervention and associated with a better treatment outcome. These findings support further research into the implementation of mechanism-driven neurocognitive preventive interventions for PTSD.

Post-traumatic stress disorder associated with sexual assault among women in the WHO World Mental Health Surveys

BACKGROUND

Sexual assault is a global concern with post-traumatic stress disorder (PTSD), one of the common sequelae. Early intervention can help prevent PTSD, making identification of those at high risk for the disorder a priority. Lack of representative sampling of both sexual assault survivors and sexual assaults in prior studies might have reduced the ability to develop accurate prediction models for early identification of high-risk sexual assault survivors.

METHODS

Data come from 12 face-to-face, cross-sectional surveys of community-dwelling adults conducted in 11 countries. Analysis was based on the data from the 411 women from these surveys for whom sexual assault was the randomly selected lifetime traumatic event (TE). Seven classes of predictors were assessed: socio-demographics, characteristics of the assault, the respondent's retrospective perception that she could have prevented the assault, other prior lifetime TEs, exposure to childhood family adversities and prior mental disorders.

RESULTS

Prevalence of Diagnostic and Statistical Manual of Mental Disorders IV (DSM-IV) PTSD associated with randomly selected sexual assaults was 20.2%. PTSD was more common for repeated than single-occurrence victimization and positively associated with prior TEs and childhood adversities. Respondent's perception that she could have prevented the assault interacted with history of mental disorder such that it reduced odds of PTSD, but only among women without prior disorders (odds ratio 0.2, 95% confidence interval 0.1-0.9). The final model estimated that 40.3% of women with PTSD would be found among the 10% with the highest predicted risk.

CONCLUSIONS

Whether counterfactual preventability cognitions are adaptive may depend on mental health history. Predictive modelling may be useful in targeting high-risk women for preventive interventions.

Data Science in the Research Domain Criteria Era: Relevance of Machine Learning to the Study of Stress Pathology, Recovery, and Resilience

Diverse environmental and biological systems interact to influence individual differences in response to environmental stress. Understanding the nature of these complex relationships can enhance the development of methods to: (1) identify risk, (2) classify individuals as healthy or ill, (3) understand mechanisms of change, and (4) develop effective treatments. The Research Domain Criteria (RDoC) initiative provides a theoretical framework to understand health and illness as the product of multiple inter-related systems but does not provide a framework to characterize or statistically evaluate such complex relationships. Characterizing and statistically evaluating models that integrate multiple levels (e.g. synapses, genes, environmental factors) as they relate to outcomes that a free from prior diagnostic benchmarks represents a challenge requiring new computational tools that are capable to capture complex relationships and identify clinically relevant populations. In the current review, we will summarize machine learning methods that can achieve these goals.

Applying network analysis to psychological comorbidity and health behavior: Depression, PTSD, and sexual risk in sexual minority men with trauma histories

OBJECTIVE

High rates of depression and posttraumatic stress disorder (PTSD) contribute to sexual risk, particularly in men who have sex with men (MSM) who have experienced childhood sexual abuse. The comorbidity between depression and PTSD and mechanisms by which they contribute to sexual risk in MSM remain unclear. This study sought to demonstrate the feasibility and utility of a network approach to (a) characterize symptom interconnections between depression and PTSD in MSM, (b) identify specific symptoms related to sexual risk behavior, and (c) compare symptom networks across groups at different levels of risk.

METHOD

Cross-sectional baseline data were collected from 296 HIV-negative urban MSM as part of a multisite randomized intervention trial. Symptoms of depression and PTSD were self-reported along with sexual risk behavior. Analyses were performed in R using regularized partial correlation network modeling.

RESULTS

Network analyses revealed complex associations between depression and PTSD symptoms and in relation to sexual risk behavior. While symptoms clustered within their respective disorders, depression and PTSD were connected at key symptom nodes (e.g., sleep, concentration). Specific symptoms (e.g., avoiding thoughts and feelings) were linked to sexual risk behavior. Network comparisons across risk groups suggested avoidant processes could be more readily activated in higher-risk individuals, whereas hyperarousal symptoms may be more salient and protective for lower-risk individuals.

CONCLUSIONS

This study is one of the earliest network analyses of depression and PTSD, and first to extend this inquiry to health behavior. Symptom-level investigations may clarify mechanisms underlying psychological comorbidity and behavioral risk in MSM and refine targets for intervention/prevention. (PsycINFO Database Record

Machine learning methods to predict child posttraumatic stress: a proof of concept study

BACKGROUND

The care of traumatized children would benefit significantly from accurate predictive models for Posttraumatic Stress Disorder (PTSD), using information available around the time of trauma. Machine Learning (ML) computational methods have yielded strong results in recent applications across many diseases and data types, yet they have not been previously applied to childhood PTSD. Since these methods have not been applied to this complex and debilitating disorder, there is a great deal that remains to be learned about their application. The first step is to prove the concept: Can ML methods - as applied in other fields - produce predictive classification models for childhood PTSD? Additionally, we seek to determine if specific variables can be identified - from the aforementioned predictive classification models - with putative causal relations to PTSD.

METHODS

ML predictive classification methods - with causal discovery feature selection - were applied to a data set of 163 children hospitalized with an injury and PTSD was determined three months after hospital discharge. At the time of hospitalization, 105 risk factor variables were collected spanning a range of biopsychosocial domains.

RESULTS

Seven percent of subjects had a high level of PTSD symptoms. A predictive classification model was discovered with significant predictive accuracy. A predictive model constructed based on subsets of potentially causally relevant features achieves similar predictivity compared to the best predictive model constructed with all variables. Causal Discovery feature selection methods identified 58 variables of which 10 were identified as most stable.

CONCLUSIONS

In this first proof-of-concept application of ML methods to predict childhood Posttraumatic Stress we were able to determine both predictive classification models for childhood PTSD and identify several causal variables. This set of techniques has great potential for enhancing the methodological toolkit in the field and future studies should seek to replicate, refine, and extend the results produced in this study.

Reducing intrusive traumatic memories after emergency caesarean section: A proof-of-principle randomized controlled study

Preventative psychological interventions to aid women after traumatic childbirth are needed. This proof-of-principle randomized controlled study evaluated whether the number of intrusive traumatic memories mothers experience after emergency caesarean section (ECS) could be reduced by a brief cognitive intervention. 56 women after ECS were randomized to one of two parallel groups in a 1:1 ratio: intervention (usual care plus cognitive task procedure) or control (usual care). The intervention group engaged in a visuospatial task (computer-game 'Tetris' via a handheld gaming device) for 15 min within six hours following their ECS. The primary outcome was the number of intrusive traumatic memories related to the ECS recorded in a diary for the week post-ECS. As predicted, compared with controls, the intervention group reported fewer intrusive traumatic memories (M = 4.77, SD = 10.71 vs. M = 9.22, SD = 10.69, d = 0.647 [95% CI: 0.106, 1.182]) over 1 week (intention-to-treat analyses, primary outcome). There was a trend towards reduced acute stress re-experiencing symptoms (d = 0.503 [95% CI: -0.032, 1.033]) after 1 week (intention-to-treat analyses). Times series analysis on daily intrusions data confirmed the predicted difference between groups. 72% of women rated the intervention "rather" to "extremely" acceptable. This represents a first step in the development of an early (and potentially universal) intervention to prevent postnatal posttraumatic stress symptoms that may benefit both mother and child.

CLINICAL TRIAL REGISTRATION

ClinicalTrials.gov, www.clinicaltrials.gov, NCT02502513.

Diagnostic potential of multimodal neuroimaging in posttraumatic stress disorder

Despite accumulating evidence of physiological abnormalities related to posttraumatic stress disorder (PTSD), the current diagnostic criteria for PTSD still rely on clinical interviews. In this study, we investigated the diagnostic potential of multimodal neuroimaging for identifying posttraumatic symptom trajectory after trauma exposure. Thirty trauma-exposed individuals and 29 trauma-unexposed healthy individuals were followed up over a 5-year period. Three waves of assessments using multimodal neuroimaging, including structural magnetic resonance imaging (MRI) and diffusion-weighted MRI, were performed. Based on previous findings that the structural features of the fear circuitry-related brain regions may dynamically change during recovery from the trauma, we employed a machine learning approach to determine whether local, connectivity, and network features of brain regions of the fear circuitry including the amygdala, orbitofrontal and ventromedial prefrontal cortex (OMPFC), hippocampus, insula, and thalamus could distinguish trauma-exposed individuals from trauma-unexposed individuals at each recovery stage. Significant improvement in PTSD symptoms was observed in 23%, 52%, and 88% of trauma-exposed individuals at 1.43, 2.68, and 3.91 years after the trauma, respectively. The structural features of the amygdala were found as major classifiers for discriminating trauma-exposed individuals from trauma-unexposed individuals at 1.43 years after the trauma, but these features were nearly normalized at later phases when most of the trauma-exposed individuals showed clinical improvement in PTSD symptoms. Additionally, the structural features of the OMPFC showed consistent predictive values throughout the recovery period. In conclusion, the current study provides a promising step forward in the development of a clinically applicable predictive model for diagnosing PTSD and predicting recovery from PTSD.

Heart rate-based window segmentation improves accuracy of classifying posttraumatic stress disorder using heart rate variability measures

OBJECTIVE

Heart rate variability (HRV) characterizes changes in autonomic nervous system function and varies with posttraumatic stress disorder (PTSD). In this study we developed a classifier based on heart rate (HR) and HRV measures, and improved classifier performance using a novel HR-based window segmentation.

APPROACH

Single-channel ECG data were collected from 23 subjects with current PTSD, and 25 control subjects with no history of PTSD over 24 h. RR intervals were derived from these data, cleaned, and used to calculate HR and HRV metrics. These metrics were used as features in a logistic regression classifier. Performance was assessed via repeated random sub-sampling validation. To reduce noise and activity-related effects, we calculated features from five non-overlapping ten-minute quiescent segments of RR intervals defined by lowest HR, as well as random ten-minute segments as a control.

MAIN RESULTS

Using a combination of the four most predictive features derived from quiescent segments we achieved a median area under the receiver operating curve (AUC) of 0.86 on out-of-sample test set data. This was significantly higher than the AUC using 24 h of data (0.72) or random segments (0.67).

SIGNIFICANCE

These results demonstrate our segmentation approach improves the classification of PTSD from HR and HRV measures, and suggest the potential for tracking PTSD illness severity via objective physiological monitoring. Future studies should prospectively evaluate if classifier output changes significantly with worsening or effective treatment of PTSD.

Development and preliminary performance of a risk factor screen to predict posttraumatic psychological disorder after trauma exposure

OBJECTIVE

We examined data from a prospective study of risk factors that increase vulnerability or resilience, exacerbate distress, or foster recovery to determine whether risk factors accurately predict which individuals will later have high posttraumatic (PT) symptom levels and whether brief measures of risk factors also accurately predict later symptom elevations.

METHOD

Using data from 129 adults exposed to traumatic injury of self or a loved one, we conducted receiver operating characteristic (ROC) analyses of 14 risk factors assessed by full-length measures, determined optimal cutoff scores, and calculated predictive performance for the nine that were most predictive. For five risk factors, we identified sets of items that accounted for 90% of variance in total scores and calculated predictive performance for sets of brief risk measures.

RESULTS

A set of nine risk factors assessed by full measures identified 89% of those who later had elevated PT symptoms (sensitivity) and 78% of those who did not (specificity). A set of four brief risk factor measures assessed soon after injury identified 86% of those who later had elevated PT symptoms and 72% of those who did not.

CONCLUSIONS

Use of sets of brief risk factor measures shows promise of accurate prediction of PT psychological disorder and probable PTSD or depression. Replication of predictive accuracy is needed in a new and larger sample.

The association between posttraumatic stress symptoms, depression, and length of hospital stay following traumatic injury

OBJECTIVE

The present study examined the relationship between posttraumatic stress symptoms (PTSS) and depression symptoms with hospital outcome measures to explore how psychiatric factors relate to hospital length of stay (LOS).

METHOD

Participants were adults admitted to a large Level I Trauma Center for longer than 24h. Depression was assessed at hospitalization using the Patient Health Questionnaire (PHQ-8), and PTSS was measured by the Primary Care PTSD Screen (PC-PTSD). Hospital outcome information was collected from the hospital's trauma registry. Pearson correlations were performed.

RESULTS

460 participants (mean age=44years, SD=16.8; 65.4% male) completed the study. Baseline PTSS and depression were significantly correlated with longer hospital LOS while controlling for demographics and injury severity (p=0.026; p=0.023). Both PTSS-positive and depression-positive groups had an average increased hospital LOS of two days.

CONCLUSIONS

A significant proportion of individuals who are admitted to the hospital following trauma may be at risk for depression and PTSS, which may then increase hospital LOS. As national attention turns to reducing healthcare costs, early screenings and interventions may aid in minimizing psychiatric symptoms in trauma patients, in turn reducing the cost and outcomes associated with total hospital LOS.

Computational neuroscience approach to biomarkers and treatments for mental disorders

Psychiatry research has long experienced a stagnation stemming from a lack of understanding of the neurobiological underpinnings of phenomenologically defined mental disorders. Recently, the application of computational neuroscience to psychiatry research has shown great promise in establishing a link between phenomenological and pathophysiological aspects of mental disorders, thereby recasting current nosology in more biologically meaningful dimensions. In this review, we highlight recent investigations into computational neuroscience that have undertaken either theory- or data-driven approaches to quantitatively delineate the mechanisms of mental disorders. The theory-driven approach, including reinforcement learning models, plays an integrative role in this process by enabling correspondence between behavior and disorder-specific alterations at multiple levels of brain organization, ranging from molecules to cells to circuits. Previous studies have explicated a plethora of defining symptoms of mental disorders, including anhedonia, inattention, and poor executive function. The data-driven approach, on the other hand, is an emerging field in computational neuroscience seeking to identify disorder-specific features among high-dimensional big data. Remarkably, various machine-learning techniques have been applied to neuroimaging data, and the extracted disorder-specific features have been used for automatic case-control classification. For many disorders, the reported accuracies have reached 90% or more. However, we note that rigorous tests on independent cohorts are critically required to translate this research into clinical applications. Finally, we discuss the utility of the disorder-specific features found by the data-driven approach to psychiatric therapies, including neurofeedback. Such developments will allow simultaneous diagnosis and treatment of mental disorders using neuroimaging, thereby establishing 'theranostics' for the first time in clinical psychiatry.

Examining the disconnect between psychometric models and clinical reality of posttraumatic stress disorder

There have been many factor analytic studies aimed at testing alternative latent structures of DSM-IV and DSM-5 posttraumatic stress disorder (PTSD) symptoms. The primary rationale for such studies is that determining the 'best' factor analytic model will result in better diagnoses if that structure is the basis for diagnostic decisions. However, there appears to be a disconnect between the factor analytic modelling and the diagnostic implications. In this study, we derived prevalence rates based on commonly reported models of PTSD, based on data from two clinical samples (N=434), and also assessed if the different models generated consistent risk estimates in relation to the effects of childhood maltreatment. We found that the different models produced different prevalence rates, ranging from 64.5% to 83.9%. Furthermore, we found that the relationship between childhood maltreatment and 'diagnosis' varied considerably depending upon which latent symptom profile was adopted. It is argued that, given the maturity of this area of research, factor analytic studies of PTSD should now include information on the diagnostic implications of their findings.