Proceedings of the 2018 IISE Annual Conference K. Barker, D. Berry, C. Rainwater, eds. Predictive Modeling of Post-Traumatic Stress Disorder Abstract ID: 1264 Ayca Erdogan , Minnie Patel, Yasser Dessouky San Jose State University Industrial and Systems Engineering Department San Jose, CA, 95192 , USA Steven Sanchez, Balmatee Bidassie VA Veterans Engineering Resource Center Indianapolis, IN, 46222, USA Abstract Posttraumatic Stress Disorder (PTSD) is a psychiatric disorder that can occur in people after experiencing or witnessing a life-threatening event such as a war/military-combat, terrorist attack, natural disaster, serious accident, sexual or other violent personal attack. It is one of the biggest problems facing veterans in the Veterans Health Administration. In this work, we present a methodology that is used to analyze a sample data set with dummy data created based on the findings in literature and build a prediction model. First, 2X2 Contingency tables are used to investigate the statistical independence of variables. Then, a Bayesian Model is developed to understand the underlying relationship between PTSD and various risk factors including but not limited to depression, substance abuse, bipolar disease, service length, combat experience along with varying demographic factors. Finally, logistic regression is used to model the relationship between PTSD and other variables that are found to be significantly related to PTSD. Our findings indicate that PTSD has significant relationship to depression and substance abuse. Keywords Post traumatic stress disorder, PTSD, Bayesian Modeling, Predictive modeling 1 Introduction According to American Psychiatric Association, “Posttraumatic stress disorder (PTSD) is a psychiatric disorder that can occur in people who have experienced or witnessed a traumatic event such as a natural disaster, a serious accident, a terrorist act, war/combat, rape or other violent personal assault” [1]. It approximately affects 3.5 million of adults in the US. While it occurs in people after a traumatic event, it should be noted that not everybody who experienced a trauma such as combat or violent assault develop PTSD, in fact most people don’t develop it. Certain risk factors may make a person more likely to develop PTSD. For example, PTSD and other mental diseases such as depression are often seen together. Results from a large national survey showed that depression is nearly 3 to 5 times more likely in those with PTSD than those without PTSD [5]. PTSD is also one of the largest medical problems facing Veterans in the VA health system [2-3]. It is estimated that approximately 1 in 12 veterans will experience it in their lifetime and this estimate increases for some specific cohorts [4]. About 11-20% of veterans served in Operations Iraqi Freedom, 12% of the veterans served in Gulf War, 15% of the veterans served in Vietnam War have had PTSD in their lifetime [5]. The objective of this work was to present a methodology that can help VA to identify veterans that are at high risk of developing PTSD such that a standardized referral process and appropriate treatments could be implemented. Section 2 includes information about the data and preprocessing methodology followed in this study. Sections 3 includes steps in methodology in analyzing the data set and building prediction model. Section 4 includes results and, finally Section 5 includes conclusions and discussion.