Neurocognitive predictors of treatment outcomes in psychotherapy for comorbid PTSD and substance use disorders

Abstract

Objective: Comorbidity between posttraumatic stress disorder (PTSD) and substance use disorders (SUD) is common, and both are associated with cognitive dysfunction. However, few studies examine the impact of cognitive deficits on treatment outcomes. Here, we leverage data from a randomized clinical trial of integrated versus phased psychotherapy for SUD and PTSD to examine the relation of cognitive functioning to treatment response.

Method: One-hundred and thirteen veterans with co-occurring PTSD and SUD completed Penn Computerized Neurocognitive Battery tests assessing attention, executive control, memory, and spatial processing. Linear mixed-effects models examined interactions between cognitive functioning and time in predicting primary PTSD and SUD outcomes across both treatments.

Results: Significant verbal immediate memory by time interactions were found for both PTSD symptoms (p = .01, f 2 = 0.020) and percent heavy drinking or drug use days (p = .004, f 2 = 0.020). There was a significant working memory by time interaction for percent heavy drinking or drug use days (p = .007, f 2 = 0.016). Participants with better verbal memory had greater reductions across time in PTSD symptoms and drinking/drug use, while those with better working memory had lesser reductions in their drinking/drug use across time.

Conclusions: Individuals with lower verbal memory functioning had less robust PTSD and SUD symptom reductions in PTSD/SUD psychotherapy, with differences that were generally small in magnitude. Those with better working memory functioning had worse SUD outcomes. Together with prior literature, findings suggest that neurocognitive functioning may impact the effectiveness of PTSD and SUD treatment. (PsycInfo Database Record (c) 2021 APA, all rights reserved).

Trial registration: ClinicalTrials.gov NCT01211106.

Conflict of interest statement

The authors declare no conflicts of interest.

Figures

Figure 1.

Working memory and verbal memory predicting the time course of PTSD symptoms. Note. Although continuous variables were used in primary analyses, tertiles were created from these scores for visualization of these interactions. Panel A shows the sample participants categorized into three (approximately equal sized) groups on the basis of the tertiles of their baseline working memory scores, with the first-tertile group comprising the participants in the lowest third, the second-tertile group comprising the participants in the middle third, and the third-tertile group comprising the participants with the highest third of scores. Panel B was created in the same way, this time using the tertiles of the baseline immediate verbal memory scores.

Figure 2.

Working memory and verbal memory predicting the time course of percent drug use or heavy drinking days. Note. Although continuous variables were used in primary analyses, tertiles were created from these scores for visualization of these interactions. Panel A shows the sample participants categorized into three (approximately equal sized) groups on the basis of the tertiles of their baseline working memory scores, with the first-tertile group comprising the participants in the lowest third, the second-tertile group comprising the participants in the middle third, and the third-tertile group comprising the participants with the highest third of scores. Panel B was created in the same way, this time using the tertiles of the baseline immediate verbal memory scores.