Predicting Postraumatic Stress Disorder in Single-Incident Trauma Survivors with an Acute Injury

Abstract

The objective of this study was to create a brief and easily administered screen that can be used by hospital staff to identify those at risk for the later development of PTSD. Utilizing previous research examining pretrauma, peritrauma, and posttrauma risk factors for the development of PTSD among single-incident trauma survivors with an acute injury, an item pool was created and reviewed by experts in the field. This item pool along with a previously created screen were given to patients admitted to two level 1 trauma centers in the U.S. A follow-up was conducted at one month in which participants were administered two psychometrically valid PTSD diagnostic tools. A stepwise bivariate logistic regression was used to determine the items from the item pool that were most strongly associated with PTSD diagnosis at approximately one month post injury. The logistic regression yielded a five item model which outperformed a previously created screen. ROC curve analysis was used to determine sensitivity (100%), specificity (81.33%), negative predictive value (100%), positive predictive value (66.7%), and to yield an optimal cutoff score (≥1). The clinical implications of this tool along with a rationale for item retention is provided.