Imaginal exposure processing during Concurrent Treatment of PTSD and Substance Use Disorders using Prolonged Exposure (COPE) therapy: Examination of linguistic markers of cohesiveness

Abstract

Concurrent Treatment of PTSD and Substance Use Disorders Using Prolonged Exposure (COPE) is an integrated, evidence-based treatment that results in significant reductions in posttraumatic stress disorder (PTSD) and substance use disorder (SUD) severity. Emotional processing theory suggests that successful prolonged exposure-based treatments should result in more cohesive trauma narratives due to better integration and organization of trauma memory into cognitive conceptualizations of fear. Therefore, we hypothesized that language used by patients would become more cohesive over time and increased language cohesion would be related to larger reductions in PTSD and SUD outcomes. Broadly, language cohesion refers to several linguistic devices that help establish and cohere meaning throughout spoken and written discourse (e.g., increased use of transition words like "and," "then," and "but"). This was the first known study to examine changes in language related to both PTSD and SUD severity during COPE treatment. The sample included 28 military veterans with current comorbid PTSD/SUD enrolled in a larger COPE study. A text analysis program, Coh-Metrix, was used to analyze language cohesiveness. No language cohesion variables significantly changed over time. Narrativity levels significantly moderated change in PTSD outcomes, $R_{\beta }^2$ = 0.11. Adversative connectives significantly moderated change in SUD outcomes, $R_{\beta }^2$ = 0.26. The findings illuminate potential processes underlying successful COPE treatment. Less use of language conveying a narrative and more use of contrast-indicative words (e.g., but, whereas) was associated with larger reductions in PTSD and SUD outcomes during treatment. These results contribute to the extant literature on associations between trauma exposure, language, and emotional processing.