Reliability and validity of the PDS and PSS-I among participants with PTSD and alcohol dependence

Abstract

The prevalence of alcohol use disorder (e.g., alcohol dependence; AD) among individuals with posttraumatic stress disorder (PTSD) is quite high, with estimates of 52% for men and 30% for women (Kessler, Sonnega, Bromet, Hughes, & Nelson, 1995). There are several interviews and self-report measures of PTSD with good published psychometric properties, and they are routinely used with comorbid AD and PTSD. However, none of these instruments was validated with this population. The current study fills this gap by examining the psychometric properties of the PTSD Symptom Scale-Interview (PSS-I) and the self-report PTSD Diagnostic Scale (PDS) in individuals diagnosed with current PTSD and AD. Both scales comprised of 17 items provide diagnostic and symptom severity information according to DSM-IV-TR criteria. Participants were 167 individuals who were diagnosed with AD and chronic PTSD and were enrolled in a randomized controlled treatment study. Results revealed excellent internal consistency of both the PSS-I and the PDS, good test–retest reliability over a 1-month period, and good convergent validity with the SCID. The specificity of the PSS-I diagnosis of PTSD was better than the PDS diagnosis, the latter exhibiting a greater percentage of false positives. In sum, the results showed that the PSS-I and PDS performed well in this population and can be used with confidence to assess PTSD diagnosis and symptom severity.