A pilot study of an online group-based Internal Family Systems intervention for comorbid posttraumatic stress disorder and substance use

Abstract

Individuals with comorbid posttraumatic stress disorder (PTSD) and substance use disorder (SUD) present with a diversity of symptoms. Current interventions show minimal efficacy differences and have high attrition. Offering a variety of treatment options, including virtual ones, ensures treatment access that is appropriate and acceptable to individual needs. We conducted a single-arm pilot study to examine the acceptability and feasibility of an online intervention based on Internal Family Systems (IFS), called the Program for Alleviating and Reducing Trauma, Stress, and Substance Use (PARTS-SUD). Ten adults (N=10), with comorbid diagnoses of PTSD and SUD, were allocated to 12 weekly groups with 6 individual counseling sessions. Our pre-specified aims were acceptability (70% overall acceptability, 75% willingness to refer a friend), and feasibility (70% completion), with key exploratory clinical outcomes (PTSD symptom severity and craving). Participants rated the intervention with a mean score of 86% on acceptability, 92% on willingness to refer a friend, retaining 70% of participants at 12 weeks. Furthermore, PTSD symptoms reduced by 1.7 points/week (95% CI: -2.45, -0.93, p=0.002) with 54% of the sample achieving a minimally important difference in PCL-5 scores. Craving Scale scores were reduced by 0.25 points/week (95% CI: -0.45, -0.06, p=0.014). An online IFS intervention was a feasible and acceptable way to provide whole-person treatment for people with PTSD-SUD within a diverse community mental health center setting. Despite being a small pilot study, decreases in both PTSD symptom severity and craving indicate the need for a randomized controlled trial with a large, diverse sample.

Keywords: Internal Family Systems; PTSD-SUD; addiction; community mental health; telehealth; trauma; whole-person treatment.

Figure 1

CONSORT diagram.

Figure 2

Global effects for the reduction in symptom severity of PTSD and craving were observed over the course of the intervention. A linear mixed-effects analysis was conducted to examine the effect of intervention time on two clinical outcomes of interest (PCL-5 and Craving Scale). Treatment Time was treated as a continuous fixed effect (marginal), allowing for the estimation of a global trend (solid black line) across participants. To account for individual differences among participants (N=10), both at baseline and in the rate of change over time random intercepts and random slopes (dotted grey lines in (A, C) were included in the model. To understand the uncertainty around the marginal effect of intervention time (β, solid black line), a non-parametric bootstrapping procedure (N=1000) with replacement (including attrition) was performed (B, D). (A) PCL-5: There is a global effect of Treatment Time (β = -1.7 ± 0.4, p = 0.002). (B) PCL-5: In 52% of the shuffled datasets we found a marginal effect at least as extreme as in the observed sample. (C) Craving Scale: A global effect of Treatment Time (β = -0.3 ± 0.1, p=0.014) with higher participant variability in rates of change over time. (D) Craving Scale: In 30% of the shuffled datasets we found a marginal effect at least as extreme as in the observed sample.