Symposia
Trauma and Stressor Related Disorders and Disasters
Skye Fitzpatrick, Ph.D.
Professor
York University
Toronto, Ontario, Canada
Antonio Morgan-Lopez, PhD
Fellow
RTI International
Research Triangle Park, North Carolina
Alex Kline, PhD
Postdoctoral Fellow
University of California- San diego
La Jolla, California
Teresa López-Castro, PhD
Assistant Professor
City College of New York
new York, New York
Sonya Norman, Ph.D.
Director, PTSD Consultation Program
NCPTSD
San Diego, CA
Lissette Saavedra, PhD
Senior Research Psychologist
RTI International
research Triangle, North Carolina
Lesia Ruglass, PhD
Professor
City College of New York
New York, New York
Therese Killeen, Ph.D.
Research Professor
Medical University of South Carolina
Charleston, SC
Sudie Back, Ph.D.
Professor
Medical University of South Carolina/Ralph H. Johnson VA Medical Center
Charleston, SC
Chantel Ebrahimi, MA
PhD Student
The New School
New York, New York
Denise Hien, PhD
Vice Provost for Research
Rutgers University
piscataway, New Jersey
Posttraumatic Stress Disorder and Alcohol/Other Drug Disorders (PTSD+AOD) are highly comorbid and mutually exacerbate the functioning, severity, and treatment outcomes of each other. A range of treatments for PTSD+AOD have been developed that vary in the extent to which they are behavioural versus pharmacologic; focus on one disorder versus the other; involve trauma processing (i.e., trauma-focus); and are integrated (i.e., focus on both PTSD and AOD) versus not. Understanding the impact of treatment dose on PTSD+AOD outcomes across these interventions is important optimize treatment selection and implementation. However, existing findings regarding dose response in this area are problematic because treatment dosage is often non-randomized and self-selected. Thus, the intervention dosage that individuals receive is confounded with individual characteristics that are related both to self-selected dosage and treatment outcomes. Recent advances in causal moderation analysis, using propensity score weighting and “parallel (multilevel) regression” models can yield results akin to virtual randomization of individuals to both interventions and intervention dosage. The present study used causal moderation analysis with data collected as part of Project Harmony, an initiative that synthesized raw data from 36 randomized controlled trials (RCTs) of PTSD+AOD interventions (N = 4046). Significant dose × treatment interaction effects were observed only for non-integrated, trauma-focused interventions. Specifically, increases in dose in these treatments were associated with greater improvements in PTSD (b = -1.08 [CI = -1.40, -.74]), alcohol use, b = -.77 [CI = -.96, -.56]), and drug use (b = -.55 [CI = -.77, -.32]), compared to treatment-as-usual at 12 month follow-up. Findings suggest that, in general, yielding full or adequate PTSD+AOD treatment doses are particularly important when clients are receiving trauma-focused interventions. Clinicians should emphasize the importance of dose early and utilize strategies to boost patient attendance in such interventions, and potentially select these treatments particularly for those with a low likelihood of dropping out.