Faculty Bibliography

PTSD is distressful and debilitating, following a non-remitting course in about 10% to 20% of trauma survivors. Numerous risk indicators of PTSD have been identified, but individual level prediction remains elusive. As an effort to bridge the gap between scientific discovery and practical application, we designed and implemented a clinical decision support pipeline to provide clinically relevant recommendation for trauma survivors. To meet the specific challenge of early prediction, this work uses data obtained within ten days of a traumatic event. The pipeline creates personalized predictive model for each individual, and computes quality metrics for each predictive model. Clinical recommendations are made based on both the prediction of the model and its quality, thus avoiding making potentially detrimental recommendations based on insufficient information or suboptimal model. The current pipeline outperforms the acute stress disorder, a commonly used clinical risk factor for PTSD development, both in terms of sensitivity and specificity.

Objectives: Peri-traumatic reactions to potentially traumatic events are likely to play a part in the development of posttraumatic stress disorder. In addition they are the focus of psychological first aid. Most studies have retrospective data, sometimes gathered months after the event. This study reports on data collected in the Emergency Department following a traumatic event. Methods: Consecutive admissions to a Level I trauma center following motor vehicle accident or terror attacks were assessed for objective aspects of the event, peri-traumatic distress and dissociation, while still in the Emergency Department. Results: These show that different interventions are associated with different event types. Motor vehicle accidents appear to contain fewer objectively difficult aspects, and survivors report corresponding lower distress and dissociation. Conclusions: These data give preliminary evidence for levels of distress and dissociation found in the first hours following a traumatic event, following different event types

Importance: The long-term course of readjustment problems in military personnel has not been evaluated in a nationally representative sample. The National Vietnam Veterans Longitudinal Study (NVVLS) is a congressionally mandated assessment of Vietnam veterans who underwent previous assessment in the National Vietnam Veterans Readjustment Study (NVVRS). Objective: To determine the prevalence, course, and comorbidities of war-zone posttraumatic stress disorder (PTSD) across a 25-year interval. Design, Setting, and Participants: The NVVLS survey consisted of a self-report health questionnaire (n = 1409), a computer-assisted telephone survey health interview (n = 1279), and a telephone clinical interview (n = 400) in a representative national sample of veterans who served in the Vietnam theater of operations (theater veterans) from July 3, 2012, through May 17, 2013. Of 2348 NVVRS participants, 1920 were alive at the outset of the NVVLS, and 81 died during recruitment; 1450 of the remaining 1839 (78.8%) participated in at least 1 NVVLS study phase. Data analysis was performed from May 18, 2013, through January 9, 2015, with further analyses continued through April 13, 2015. Main Outcomes and Measures: Study instruments included the Mississippi Scale for Combat-Related PTSD, PTSD Checklist for DSM-IV supplemented with PTSD Checklist for DSM-5 items (PCL-5+), Clinician-Administered PTSD Scale for DSM-5 (CAPS-5), and Structured Clinical Interview for DSM-IV, Nonpatient Version. Results: Among male theater veterans, we estimated a prevalence (95% CI) of 4.5% (1.7%-7.3%) based on CAPS-5 criteria for a current PTSD diagnosis; 10.8% (6.5%-15.1%) based on CAPS-5 full plus subthreshold PTSD; and 11.2% (8.3%-14.2%) based on PCL-5+ criteria for current war-zone PTSD. Among female veterans, estimates were 6.1% (1.8%-10.3%), 8.7% (3.8%-13.6%), and 6.6% (3.5%-9.6%), respectively. The PCL-5+ prevalence (95% CI) of current non-war-zone PTSD was 4.6% (2.6%-6.6%) in male and 5.1% (2.3%-8.0%) in female theater veterans. Comorbid major depression occurred in 36.7% (95% CI, 6.2%-67.2%) of veterans with current war-zone PTSD. With regard to the course of PTSD, 16.0% of theater veterans reported an increase and 7.6% reported a decrease of greater than 20 points in Mississippi Scale for Combat-Related PTSD symptoms. The prevalence (95% CI) of current PCL-5+-derived PTSD in study respondents was 1.2% (0.0%-3.0%) for male and 3.9% (0.0%-8.1%) for female Vietnam veterans. Conclusions and Relevance: Approximately 271000 Vietnam theater veterans have current full PTSD plus subthreshold war-zone PTSD, one-third of whom have current major depressive disorder, 40 or more years after the war. These findings underscore the need for mental health services for many decades for veterans with PTSD symptoms.

UNLABELLED:Prevalence findings for the elderly are artificially low, most likely due to insufficient consideration of age-related cognitive abilities in diagnostic interviews. AIMS/OBJECTIVE:(1) To describe the rationale for the development of an age-adapted Composite International Diagnostic Interview (CIDI65+) for use in a European project (MentDis_ICF65+). (2) To examine its test-retest reliability. METHODS:Based on substantive pilot work the CIDI standard questions were shortened, broken down into shorter subsets and combined with sensitization questions and dimensional measures. Test-retest was determined in N = 68 subjects aged 60-79 years via two independent examinations by clinical interviewers using kappa (sensitivity, specificity) for categorical and intraclass correlation (ICC) coefficients for dimensional measures. RESULTS:Test-retest reliability was good for any mental disorder (κ = 0.63), major depression (κ = 0.55), anxiety (κ = 0.62, range = 0.30-0.78), substance (κ = 0.77, range = 0.71-0.82), obsessive-compulsive disorder (κ = 1.00) and most core symptoms/syndromes (κ range = 0.48-1.00). Agreement for some disorders (i.e. somatoform/pain) attenuated, partly due to time lapse effects. ICC for age of onset, recency, quantity, frequency and duration questions ranged between κ = 0.60-0.90. Dimensional agreement measures were not consistently higher. CONCLUSION/CONCLUSIONS:The age-adapted CIDI65+ is reliable for assessing most mental disorders, distress, impairment and time-related information in the elderly, prompting the need to examine validity.

BACKGROUND: Predicting Posttraumatic Stress Disorder (PTSD) is a pre-requisite for targeted prevention. Current research has identified group-level risk-indicators, many of which (e.g., head trauma, receiving opiates) concern but a subset of survivors. Identifying interchangeable sets of risk indicators may increase the efficiency of early risk assessment. The study goal is to use supervised machine learning (ML) to uncover interchangeable, maximally predictive combinations of early risk indicators. METHODS: Data variables (features) reflecting event characteristics, emergency department (ED) records and early symptoms were collected in 957 trauma survivors within ten days of ED admission, and used to predict PTSD symptom trajectories during the following fifteen months. A Target Information Equivalence Algorithm (TIE*) identified all minimal sets of features (Markov Boundaries; MBs) that maximized the prediction of a non-remitting PTSD symptom trajectory when integrated in a support vector machine (SVM). The predictive accuracy of each set of predictors was evaluated in a repeated 10-fold cross-validation and expressed as average area under the Receiver Operating Characteristics curve (AUC) for all validation trials. RESULTS: The average number of MBs per cross validation was 800. MBs' mean AUC was 0.75 (95% range: 0.67-0.80). The average number of features per MB was 18 (range: 12-32) with 13 features present in over 75% of the sets. CONCLUSIONS: Our findings support the hypothesized existence of multiple and interchangeable sets of risk indicators that equally and exhaustively predict non-remitting PTSD. ML's ability to increase prediction versatility is a promising step towards developing algorithmic, knowledge-based, personalized prediction of post-traumatic psychopathology.

BACKGROUND: Impaired social relationships are linked with higher levels of posttraumatic stress disorder (PTSD), but the association's underlying dynamics are unknown. PTSD may impair social relationships, and, vice versa, poorer relationship quality may interfere with the recovery from PTSD. OBJECTIVE: This work longitudinally evaluates the simultaneous progression of PTSD symptoms and social relationship satisfaction (SRS) in a large cohort of recent trauma survivors. It also explores the effect of cognitive behavior therapy (CBT) on the association between the two. METHOD: Consecutive emergency department trauma admissions with qualifying PTSD symptoms (n=501) were assessed 3 weeks and 5 months after trauma admission. The World Health Organization Quality of Life evaluated SRS and the Clinician Administered PTSD Scale evaluated PTSD symptom severity. Ninety-eight survivors received CBT between measurement sessions. We used Structural Equation Modeling to evaluate cross-lagged effects between the SRS and PTSD symptoms. RESULTS: The cross-lagged effect of SRS on PTSD was statistically significant (beta=-0.12, p=0.01) among survivors who did not receive treatment whilst the effect of PTDS on SRS was nil (beta=-0.02, p=0.67). Both relationships were non-significant among survivors who received CBT. DISCUSSION: SRS and PTSD are highly associated, and this study shows that changes in SRS in the early aftermath of traumatic events contribute to changes in PTSD, rather than vice versa. SRS impacts natural recovery, but not effective treatment. This study suggests that being satisfied with one's relationships might be considered as an important factor in natural recovery from trauma, as well as in intervention.

Abnormality in the "fear circuitry" has been known as a major neural characteristic of posttraumatic stress disorder (PTSD). Recent studies also revealed decreased functional connectivity in the default mode network in PTSD. The present study aims to investigate, in war-zone-related PTSD, the spontaneous activity and functional connectivity of the amygdala and the precuneus, which are two representative brain regions of the two networks, respectively. Two groups of 52 male US Operation Iraqi Freedom (OIF) and Operation Enduring Freedom (OEF) veterans (PTSD vs. controls), well matched on age and ethnicity, were clinically assessed and then studied in a resting state functional magnetic resonance imaging (fMRI) procedure. Functional connectivity analysis was conducted on the resting state fMRI data with the amygdala and precuneus as seeds. Compared with controls, veterans with PTSD had lower functional connectivity in the default mode network, as well as lower amygdala-frontal functional connectivity. Both the PTSD and the control group had a significant positive precuneal-amygdala functional connectivity without a significant group difference. The magnitudes of spontaneous activity of the amygdala and the precuneus were negatively correlated in the PTSD group and showed significant quadratic relationships with the amount of emotional abuse in early life trauma. These findings may improve our understanding about the relationships between fear circuitry and the default mode network in the context of war-zone-related PTSD.

There is broad interest in predicting the clinical course of mental disorders from early, multimodal clinical and biological information. Current computational models, however, constitute a significant barrier to realizing this goal. The early identification of trauma survivors at risk of post-traumatic stress disorder (PTSD) is plausible given the disorder's salient onset and the abundance of putative biological and clinical risk indicators. This work evaluates the ability of Machine Learning (ML) forecasting approaches to identify and integrate a panel of unique predictive characteristics and determine their accuracy in forecasting non-remitting PTSD from information collected within10 days of a traumatic event. Data on event characteristics, emergency department observations, and early symptoms were collected in 957 trauma survivors, followed for fifteen months. An ML feature selection algorithm identified a set of predictors that rendered all others redundant. Support Vector Machines (SVMs) as well as other ML classification algorithms were used to evaluate the forecasting accuracy of i) ML selected features, ii) all available features without selection, and iii) Acute Stress Disorder (ASD) symptoms alone. SVM also compared the prediction of a) PTSD diagnostic status at 15 months to b) posterior probability of membership in an empirically derived non-remitting PTSD symptom trajectory. Results are expressed as mean Area Under Receiver Operating Characteristics Curve (AUC). The feature selection algorithm identified 16 predictors, present in >/=95% cross-validation trials. The accuracy of predicting non-remitting PTSD from that set (AUC = .77) did not differ from predicting from all available information (AUC = .78). Predicting from ASD symptoms was not better then chance (AUC = .60). The prediction of PTSD status was less accurate than that of membership in a non-remitting trajectory (AUC = .71). ML methods may fill a critical gap in forecasting PTSD. The ability to identify and integrate unique risk indicators makes this a promising approach for developing algorithms that infer probabilistic risk of chronic posttraumatic stress psychopathology based on complex sources of biological, psychological, and social information.