Karen-Inge Karstoft

Latent trajectories of trauma symptoms and resilience: the 3-year longitudinal prospective USPER study of Danish veterans deployed in Afghanistan.

OBJECTIVE To identify trajectories of posttraumatic stress disorder (PTSD) symptoms from before to 2.5 years after deployment and to assess risk factors for symptom fluctuations and late-onset PTSD. METHOD 743 soldiers deployed to Afghanistan in 2009 were assessed for PTSD symptoms using the PTSD Checklist (PCL) at 6 occasions from predeployment to 2.5 years postdeployment (study sample = 561). Predeployment vulnerabilities and deployment and postdeployment stressors were also assessed. RESULTS Six trajectories were identified: a resilient trajectory with low symptom levels across all assessments (78.1%) and 5 trajectories showing symptom fluctuations. These included a trajectory of late onset (5.7%), independently predicted by earlier emotional problems (OR = 5.59; 95% CI, 1.57-19.89) and predeployment and postdeployment traumas (OR = 1.10; 95% CI, 1.04-1.17 and OR = 1.13; 95% CI, 1.00-1.26). Two trajectories of symptom fluctuations in the low-to-moderate range (7.5% and 4.1%); a trajectory of symptom relief during deployment, but with a drastic increase at the final assessments (2.0%); and a trajectory with mild symptom increase during deployment followed by relief at return (2.7%) were also found. Symptom fluctuation was predicted independently by predeployment risk factors (depression [OR = 1.27; 95% CI, 1.16-1.39], neuroticism [OR = 1.10; 95% CI, 1.00-1.21], and earlier traumas [OR = 1.09; 95% CI, 1.03-1.16]) and deployment-related stressors (danger/injury exposure [OR = 1.20; 95% CI, 1.04-1.40]), but not by postdeployment stressors. DISCUSSION The results confirm earlier findings of stress response heterogeneity following military deployment and highlight the impact of predeployment, perideployment, and postdeployment risk factors in predicting PTSD symptomatology and late-onset PTSD symptoms.

Bridging a translational gap: using machine learning to improve the prediction of PTSD

BackgroundPredicting 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.MethodsData 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.ResultsThe 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.ConclusionsOur 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.

Diagnostic Accuracy of the Posttraumatic Stress Disorder Checklist–Civilian Version in a Representative Military Sample

This study aimed to assess the diagnostic accuracy of the Posttraumatic Stress Disorder Checklist-Civilian Version (PCL-C; Weathers, Litz, Herman, Huska, & Keane, 1993) and to establish the most accurate cutoff for prevalence estimation of posttraumatic stress disorder (PTSD) in a representative military sample compared to a clinical interview. Danish soldiers (N = 415; 94.4% male, mean age 26.6 years) were assessed with the PCL-C and the Structured Clinical Interview for the DSM-IV (SCID; First, Spitzer, Gibbon, & Williams, 2002) 2.5 years after their return from deployment to Afghanistan. Diagnostic accuracy of the PCL-C was assessed through receiver operating characteristic curve analysis. The PCL-C displayed high overall accuracy (area under the curve = .95, confidence interval [.92, .98]) and performed well (sensitivity > .70 and specificity ≥ .90), with cutoff scores ranging from 37 to 44. When including sensitivity values a little below .70 (.69), the PCL-C performed well for cutoff levels up to 53. Prevalence of PTSD varied considerably with the application of different cutoff values and scoring methods. Our results show that the PCL-C is a relevant and valid tool for screening for probable PTSD in active military samples. However, it is of great importance that cutoff scores be chosen based on the sample and the purpose of the particular study or screening.

The role of locus of control and coping style in predicting longitudinal PTSD-trajectories after combat exposure

While longitudinal posttraumatic stress responses are known to be heterogeneous, little is known about predictors of those responses. We investigated if locus of control (LOC) and coping style are associated with long-term PTSD-trajectories after exposure to combat. Six hundred and seventy five Israeli soldiers with or without combat stress reaction (CSR) from the Lebanon war were assessed 1, 2, and 20 years after the war. Combat exposure, LOC, and coping style were then investigated as covariates of the trajectories of resilience, recovery, delayed onset, and chronicity. Symptomatic trajectories in the CSR and the non-CSR group were significantly associated to varying degrees with perceived life threat during combat (ORs: 1.76-2.53), internal LOC (0.77-0.87), emotional coping style (0.28-0.34), and low use of problem-focused coping (2.12-3.11). In conclusion, assessment of LOC and coping can aid prediction of chronic PTSD outcomes of combat exposure.

Long-Term Trajectories of Posttraumatic Stress Disorder in Veterans: The Role of Social Resources

OBJECTIVE To (1) identify long-term trajectories of combat-induced posttraumatic stress disorder (PTSD) symptoms over a 20-year period from 1983 to 2002 in veterans with and without combat stress reaction (CSR) and (2) identify social predictors of these trajectories. METHOD A latent growth mixture modeling analysis on PTSD symptoms was conducted to identify PTSD trajectories and predictors. PTSD was defined according to DSM-III and assessed through the PTSD Inventory. Israeli male veterans with (n = 369) and without (n = 306) CSR were queried at 1, 2, and 20 years after war about combat exposure, military unit support, family environment, and social reintegration. RESULTS For both study groups, we identified 4 distinct trajectories with varying prevalence across groups: resilience (CSR = 34.4%, non-CSR = 76.5%), recovery (CSR = 36.3%, non-CSR = 10.5%), delayed onset (CSR = 8.4%, non-CSR = 6.9%), and chronicity (CSR = 20.9%, non-CSR = 6.2%). Predictors of trajectories in both groups included perception of war threat (ORs = 1.59-2.47, P values ≤ .30), and negative social reintegration (ORs = 0.24-0.51, P values ≤ .047). Social support was associated with symptomatology only in the CSR group (ORs = 0.40-0.61, P values ≤ .045), while family coherence was predictive of symptomatology in the non-CSR group (OR = 0.76, P = .015) but not in the CSR group. CONCLUSIONS Findings confirmed heterogeneity of long-term sequelae of combat, revealing 4 trajectories of resilience, recovery, delay, and chronicity in veterans with and without CSR. Symptomatic trajectories were more prevalent for the CSR group, suggesting that acute functional impairment predicts pathological outcomes. Predictors of symptomatic trajectories included perceived threat and social resources at the family, network, and societal levels.

Predicting criminality from child maltreatment typologies and posttraumatic stress symptoms

Background The associations between childhood abuse and subsequent criminality and posttraumatic stress disorder (PTSD) are well known. However, a major limitation of research related to childhood abuse and its effects is the focus on one particular type of abuse at the expense of others. Recent work has established that childhood abuse rarely occurs as a unidimensional phenomenon. Therefore, a number of studies have investigated the existence of abuse typologies. Methods The study is based on a Danish stratified random probability survey including 2980 interviews of 24-year-old people. The sample was constructed to include an oversampling of child protection cases. Building on a previous latent class analysis of four types of childhood maltreatment, three maltreatment typologies were used in the current analyses. A criminality scale was constructed based on seven types of criminal behavior. PTSD symptoms were assessed by the PC-PTSD Screen. Results Significant differences were found between the two genders with males reporting heightened rates of criminality. Furthermore, all three maltreatment typologies were associated with criminal behavior with odds ratios (ORs) from 2.90 to 5.32. Female gender had an OR of 0.53 and possible PTSD an OR of 1.84. Conclusion The independent association of participants at risk for PTSD and three types of maltreatment with criminality should be studied to determine if it can be replicated, and considered in social policy and prevention and rehabilitation interventions.

Exploring the benefit of synaesthetic colours: Testing for “pop-out” in individuals with grapheme–colour synaesthesia

In grapheme–colour synaesthesia, letters, numbers, and words elicit involuntary colour experiences. Recently, there has been much emphasis on individual differences and possible subcategories of synaesthetes with different underlying mechanisms. In particular, there are claims that for some, synaesthesia occurs prior to attention and awareness of the inducing stimulus. We first characterized our sample using two versions of the “Synaesthetic Congruency Task” to distinguish “projector” and “associator” synaesthetes who may differ in the extent to which their synaesthesia depends on attention and awareness. We then used a novel modification of the “Embedded Figures Task” that included a set-size manipulation to look for evidence of preattentive “pop-out” from synaesthetic colours, at both a group and an individual level. We replicate an advantage for synaesthetes over nonsynaesthetic controls on the Embedded Figures Task in accuracy, but find no support for pop-out of synaesthetic colours. We conclude that grapheme–colour synaesthetes are fundamentally similar in their visual processing to the general population, with the source of their unusual conscious colour experiences occurring late in the cognitive hierarchy.

Replicability and Generalizability of Posttraumatic Stress Disorder (PTSD) Networks : A Cross-Cultural Multisite Study of PTSD Symptoms in Four Trauma Patient Samples

The growing literature conceptualizing mental disorders like posttraumatic stress disorder (PTSD) as networks of interacting symptoms faces three key challenges. Prior studies predominantly used (a) small samples with low power for precise estimation, (b) nonclinical samples, and (c) single samples. This renders network structures in clinical data, and the extent to which networks replicate across data sets, unknown. To overcome these limitations, the present cross-cultural multisite study estimated regularized partial correlation networks of 16 PTSD symptoms across four data sets of traumatized patients receiving treatment for PTSD (total N = 2,782). Despite differences in culture, trauma type, and severity of the samples, considerable similarities emerged, with moderate to high correlations between symptom profiles (0.43–0.82), network structures (0.62–0.74), and centrality estimates (0.63–0.75). We discuss the importance of future replicability efforts to improve clinical psychological science and provide code, model output, and correlation matrices to make the results of this article fully reproducible.

Early identification of posttraumatic stress following military deployment: Application of machine learning methods to a prospective study of Danish soldiers

BACKGROUND Pre-deployment identification of soldiers at risk for long-term posttraumatic stress psychopathology after home coming is important to guide decisions about deployment. Early post-deployment identification can direct early interventions to those in need and thereby prevents the development of chronic psychopathology. Both hold significant public health benefits given large numbers of deployed soldiers, but has so far not been achieved. Here, we aim to assess the potential for pre- and early post-deployment prediction of resilience or posttraumatic stress development in soldiers by application of machine learning (ML) methods. METHODS ML feature selection and prediction algorithms were applied to a prospective cohort of 561 Danish soldiers deployed to Afghanistan in 2009 to identify unique risk indicators and forecast long-term posttraumatic stress responses. RESULTS Robust pre- and early postdeployment risk indicators were identified, and included individual PTSD symptoms as well as total level of PTSD symptoms, previous trauma and treatment, negative emotions, and thought suppression. The predictive performance of these risk indicators combined was assessed by cross-validation. Together, these indicators forecasted long term posttraumatic stress responses with high accuracy (pre-deployment: AUC = 0.84 (95% CI = 0.81-0.87), post-deployment: AUC = 0.88 (95% CI = 0.85-0.91)). LIMITATIONS This study utilized a previously collected data set and was therefore not designed to exhaust the potential of ML methods. Further, the study relied solely on self-reported measures. CONCLUSIONS Pre-deployment and early post-deployment identification of risk for long-term posttraumatic psychopathology are feasible and could greatly reduce the public health costs of war.

Stress Detection for PTSD via the StartleMart Game

Computer games have recently shown promise as a diagnostic and treatment tool for psychiatric rehabilitation. This paper examines the positive impact of affect detection and advanced game technology on the treatment of mental diagnoses such as Post Traumatic Stress Disorder (PTSD). For that purpose, we couple game design and game technology with stress detection for the automatic profiling and the personalized treatment of PTSD via game-based exposure therapy and stress inoculation training. The PTSD treatment game we designed forces the player to go through various stressful experiences while a stress detection mechanism profiles the severity and type of PTSD via skin conductance responses to those in-game stress elicitors. The initial study and analysis of 14 PTSD-diagnosed veteran soldiers presented in this paper reveals clear correspondence between diagnostic standard measures of PTSD severity and skin conductance responses. Significant correlations between physiological responses and subjective evaluations of the stressfulness of experiences, represented as pair wise preferences, are also found. We conclude that this supports the use of the simulation as a relevant treatment tool for stress inoculation training. This points to future avenues of research toward discerning between degrees and types of PTSD using game-based diagnostic and treatment tools.