(PS2-40) No Responder Left Behind: Assessing for PTSD in an Underreporting Population

Abstract Text: The Posttraumatic Stress Disorder (PTSD) Checklist for the DSM-5 (PCL-5) is widely used to assess PTSD symptoms in a variety of populations. While its intended use has been to determine the severity of trauma symptoms, it has been widely used as the primary tool to determine if someone meets the diagnostic criteria for PTSD (Blevins, et al., 2015). However, by using data collected through the clinical research protocol established within the UCF RESTORES trauma clinic, we have determined that 10 percent of patients diagnosed with PTSD had PCL-5 scores below the cutoff for probable PTSD (i.e., below 31). This study seeks to answer the question as to whether measures designed to assess anxiety (GAD-7), depression (PHQ-9), anger (DAR-5), and alcohol use (AUDIT) can be used to predict if someone will meet criteria for PTSD when their PCL-5 score is below the cutoff for probable PTSD. Data was taken from the UCF RESTORES treatment database. The diagnosis of PTSD was determined by trauma clinicians trained in administering the Clinician-Administered PTSD Scale for DSM-5 (CAPS-5). Due to unequal variance and non-normal distributions, Mann-Whitney U analyses were performed for scores on each of the measures. Of the measures, only symptom endorsement scores on the PHQ-9 (U=3891, p< .001) and the GAD-7 (U=4689, p= .006) were significantly different between groups. A logistic regression was conducted to determine if results from the PHQ-9, GAD-7, and the interaction of the scores predicted PTSD classifications. The overall model was statistically significant, 𝜒2(3) = 264, p = .002, with a Nagelkerke R2 of 0.09, suggesting a small effect. PHQ-9 scores were significant predictors of PTSD diagnosis (OR = 1.21, Β= 0.19, SE = 0.07, p = .007) meaning that for every 1-point increase on the PHQ-9 scale, the person’s odds of being diagnosed with PTSD goes up 21%. GAD-7 scores were also significant predictors of PTSD symptoms (OR = 1.20, Β= 0.19, SE = 0.07, p = .01) indicating that for every additional 1-point increase in symptom endorsements, the odds of also being diagnosed with PTSD increased by 20 %. Interestingly, the interaction between GAD-7 and PHQ-9 scores resulted in a decreased odds of being diagnosed with PTSD, OR = .99, B = -0.01, SE = 0.01, p = 0.02, for every 1 unit increase in the PHQ-9 and the GAD-7, the odds of being diagnosed with PTSD decreases by 1.48%. Results indicate that while the PHQ-9 and GAD-7 may offer some additional insight into a person’s PTSD symptoms, other measures should be evaluated for efficacy in detecting PTSD in those who under-report symptoms on the PCL-5.