Date of Degree


Document Type


Degree Name





Denise Hien

Committee Members

Eric Fertuck

Paul Wachtel

Lesia Ruglass

Teresa Lopez-Castro

Subject Categories

Clinical Psychology


posttraumatic stress disorder, substance use disorder, therapeutic alliance, prolonged exposure


Prolonged exposure has emerged as a highly effective method of treating PTSD (Foa, Hembree, & Rothbaum, 2007; Ruzek et al., 2014), yet for some individuals it has proven to be anxiety-arousing to the point of being contraindicated (Morris, 2015). As a treatment, prolonged exposure can be a challenging procedure to implement given the high levels of avoidance, fear, and resistance that often accompany a diagnosis of PTSD (Arntz, Tiesema, & Kindt, 2007; Hoffart et al., 2013; McLaughlin et al., 2014). When it is used to treat individuals who carry an additional diagnosis of SUD, further complications may arise with respect to avoidance and tolerating the temporarily anxiety-provoking nature of the treatment. To date, few studies have considered how the use of prolonged exposure in treating PTSD may impact the strength of the therapeutic alliance, and therefore additionally impact treatment outcomes. No studies have examined how prolonged exposure may be related to therapeutic alliance among comorbid substance use with PTSD. As a secondary analysis, this dissertation drew upon data from a randomized clinical trial conducted by Hien and colleagues assessing the relative efficacy of two active treatment conditions for individuals with substance dependence who additionally meet criteria for DSM-IV PTSD. This study explored the predictive capacity of early therapeutic alliance on treatment outcomes, changes in therapeutic alliance over the course of treatment and on the intra-session level when imaginal exposures were introduced, and the moderating impact of baseline PTSD and SUD severity on alliance as a predictor of treatment outcomes. Results: Multiple linear regression identified no relationship between early therapeutic alliance and treatment outcomes across measures of both PTSD and SUD. Parametric and non-parametric analyses demonstrated a significant change in the strength of therapeutic alliance from the middle phase of treatment to the end phase of treatment. Paired samples t-tests and Wilcoxon signed-rank tests identified a significant strengthening of alliance following the introduction of imaginal exposure in both the middle and end phases of treatment. Correlational analyses demonstrated that changes in alliance did not correlate with concurrent self-reported symptom changes, and regression analyses revealed baseline symptomatology did not moderate the impact of early alliance on treatment outcomes. Altogether, this study offers a valuable contribution to the literature exploring the relationship between therapeutic alliance and prolonged exposure in the comorbid PTSD and SUD population, offering findings suggesting that rather than harming the alliance, prolonged exposure techniques strengthen the alliance instead.