Treatment Attendance Among Veterans With Depression, Substance Use Disorder, and Trauma

Objective: Low attendance in psychotherapy, particularly among individuals with comorbid disorders, is a pervasive challenge. The present study examined predictors of treatment attendance in a sample of veterans with depression, substance use disorder, and trauma. Methods: This is an analysis of data collected as part of a larger clinical trial involving outpatients at a Veterans Administration dual diagnosis clinic. Individuals were excluded if they had significant memory deficits, schizophrenia, bipolar disorder, or substantial travel constraints. Participants (N = 146) received 12weeks of group-delivered integrated cognitive behavioral therapy for depression and substance use, followed by randomization to 12 additional weeks of individually delivered cognitive behavioral therapy (n = 62) or cognitive processing therapy (n = 61) modified to address substance use and trauma. Participants, therapists, and researchers were not blinded to group assignment. For this study, we included only the 123 participants who were randomized into the second phase, analyzing predictors of treatment attendance categorized into predisposing factors, enabling factors, and need factors. Results: Participants were primarily male (89%) and Caucasian (76%) and averaged 47years old (SD = 12). Forty-four percent had alcohol use disorder, 16% had drug use disorder, and 40% had polysubstance use disorder. Most met criteria for PTSD (82%), with 44% having combat-related trauma, 33% sexual trauma, and 28% other trauma. Treatment attendance did not differ between groups. More education was associated with increased group (r = .19, p = .04) and individual session attendance (r = .28, p = .002). Individuals with chronic housing problems attended fewer group sessions (r = -.19, p = .04), while individuals with sexual trauma, compared to those with other traumas, attended more individual sessions (r = .23, p = .01). Number of group sessions attended was positively associated with individual session attendance. Conclusions: Few variables were significantly predictive of treatment attendance, possibly due to the complex nature of comorbid disorders. Including a focus on trauma was not associated with lower attendance. Special consideration may need to be given to education level, homelessness, and trauma when trying to engage and retain patients with comorbid disorders in treatment.