Predicting posttraumatic stress disorder following a natural disaster

Abstract

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.

Untitled

“…This is added with the advancement of technology that facilitated to identify trauma risk [29]. A research prepared by other researcher in western countries such as in United States are much more advanced by developing tools that has been tested to identify mental health stage of an individual after the event of natural disaster that led to trauma amongst its people called a super learner algorithm [30]. However the, tools used can only measure the stage of seriousness of one mental health but not identifying what are the factors that can reduce trauma.…”

Internalizing Symptoms as Predictors of School Absenteeism Severity at Multiple Levels: Ensemble and Classification and Regression Tree Analysis

“…Ensemble analysis is the combination of multiple algorithmic models or classifiers to produce one, best model that can be applied to the data (Berk, 2006). These models have been shown to outperform standard parametric methods, primarily due to the automation of identifying interactions and non-linearities and the reduction of overestimations of a model's predictive ability (Rosellini et al, 2018). Ensemble analysis can include many different statistical methods; the present study utilized CHAID decision trees, support vector machines, and neural network analyses.…”

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

“…The most frequently used ML approach was SVM (Galatzer‐Levy, Karstoft, Statnikov, & Shalev, ; Galatzer‐Levy et al., ; Jin et al., ; Karstoft, Galatzer‐Levy, Statnikov, Li, & Shalev, ; Karstoft, Statnikov, Andersen, Madsen, & Galatzer‐Levy, ; Liu et al., ; Tylee et al., ; Zhang et al., ), which is a classification algorithm that transforms the data in a multidimensional space to search for a hyperplane that linearly separates the instances that belong to the same class (e.g., PTSD vs. no PTSD). In addition, many studies have used ensemble techniques, such as boosted trees (Galatzer‐Levy, Karstoft et al., ), random forests (Galatzer‐Levy, Karstoft et al., ; Kessler et al., ; Marinić et al., ; Reece & Danforth, ; Rosellini, Dussaillant, Zubizarreta, Kessler, & Rose, ; Saxe, Ma, Ren, & Aliferis, ), multilayer perceptron (Omurca & Ekinci, ), and Bayesian additive regression trees (Rosellini et al., ). The super learner algorithm outperformed other algorithms in two studies and was used for developing a risk score for PTSD (Kessler et al., ; Rosellini et al., ).…”

“…In addition, many studies have used ensemble techniques, such as boosted trees (Galatzer‐Levy, Karstoft et al., ), random forests (Galatzer‐Levy, Karstoft et al., ; Kessler et al., ; Marinić et al., ; Reece & Danforth, ; Rosellini, Dussaillant, Zubizarreta, Kessler, & Rose, ; Saxe, Ma, Ren, & Aliferis, ), multilayer perceptron (Omurca & Ekinci, ), and Bayesian additive regression trees (Rosellini et al., ). The super learner algorithm outperformed other algorithms in two studies and was used for developing a risk score for PTSD (Kessler et al., ; Rosellini et al., ).…”