Practice Profile: Psychotherapies for Victims of Sexual Assault

Taylor and Harvey (2009) set out to evaluate treatment outcome studies of different types of psychotherapeutic approaches for sexual assault victims experiencing post-traumatic stress disorder (PTSD) or rape trauma symptoms. Therapy was defined as any intervention designed to reduce psychological distress, symptoms of PTSD, and rape trauma through counseling, structured or unstructured interaction, training programs, or predetermined treatment plans. Studies were included in the meta-analysis based on the following criteria: (1) interventions met the definition of therapy; (2) the study examined treatment outcomes of people who had experienced sexual assault and PTSD or rape trauma; (3) the results were based on quantitative findings; (4) the study was written in English; (5) the study was not a case report; (6) the majority of the treated sample had been sexually assaulted; (7) sufficient data was available to calculate effect sizes; and (8) studies reported independent datasets to ensure that each study was coded only once.

A literature search was conducted using a wide variety of sources, including studies that were published in peer-reviewed journals, as well as other studies not published in journals (such as government or private agency sources or dissertations). Of the 89 studies identified, 23 met the criteria for inclusion in the meta-analysis. Several studies with the same datasets were linked to preserve the independence of the research. In total, 15 outcome studies representing diverse treatment approaches were selected for inclusion in the meta-analysis; these studies comprised 25 treatment conditions. Of the 15 total studies, two meta-analyses were conducted based on study design: one meta-analysis for independent samples designs that included comparison groups (eight studies) and one meta-analysis for repeated measures designs that did not include comparison groups (seven studies). The outcomes reported here are only for the meta-analysis of the independent samples designs.

The studies that used independent samples designs were published between 1991 and 2005. They examined a variety of treatment types, including prolonged exposure, stress inoculation training, supportive counseling, brief prevention program, imagery rehearsal therapy, cognitive processing therapy, and eye movement desensitization reprocessing. Participants in the studies were drawn from community samples in the United States and Western Europe. There were 383 treatment participants and 311 control participants (total=694 study participants), all of whom were female. The majority of the sample population was white or Caucasian with an undergraduate education. Most participants were between the ages of 27.5 years and 33.5 years. 

Using a fixed effects model, independent samples treatment effects were represented by standardized mean difference scores. Because studies with small sample sizes were included in the analysis, an effect size statistic (Hedge’s g) was used that adjusted for small sample size. All studies used multiple outcome measures; therefore, effect sizes for each measure were averaged to give a single effect size per study. Each study contributed only one effect size to the overall analysis. One exception was when six measurement domains (PTSD/trauma, other anxiety, depression, social, self-concept/self-esteem, and overall/other pathology) were considered, given that only measures assessing the same psychological constructs could be pooled. In this instance, all outcome measures in a single domain were averaged to give a mean effect size.

Regehr and colleagues (2013) conducted a meta-analysis to examine the effectiveness of psychotherapeutic interventions in reducing symptoms of distress and trauma in victims of sexual assault and rape. The authors conducted a comprehensive search for studies that met the following eligibility criteria: (1) the allocation of study participants to experiment or control groups was by random allocation or quasi-experimental parallel cohort design; (2) participants were adults who had experienced sexual assault or rape as adults; and (3) the interventions specifically focused on victims of sexual assault or rape. Search methods to identify eligible studies included electronic searches, inspection of reference lists of all relevant articles, personal communication with experts in the field, and hand searches of journals. Special attention was also paid to identify relevant studies from within the grey literature.

The search process yielded the inclusion of six studies that included 405 participants. The average age of women included in the review was 32.2 years, although the age range was large. The race/ethnicity was reported in five studies, and approximately 75 percent of the women were white, 20 percent were African American, and the remainder was Hispanic or other. Four of the six included studies were randomized controlled trials and the other two studies were quasi-experimental designs. All six studies were conducted in large urban settings across the United States, including Philadelphia, Penn.; St. Louis, Mo.; and Atlanta, Ga. Across the six studies, six interventions were assessed including stress inoculation training, prolonged exposure, supportive counseling or supportive psychotherapy, cognitive processing therapy, assertiveness training, and eye movement desensitization and reprocessing.

Three primary outcomes were assessed in the six included studies: PTSD symptoms, depression symptoms, and anxiety symptoms. The standardized mean difference between the treatment and comparison groups was estimated based on the reported means and standard deviations for each group. Given the small sample sizes in both the treatment and comparison groups, the authors calculated Hedges’ g, which is similar to Cohen’s d but includes an adjustment for small sample bias. Inverse variance methods were used to weight each effect size by the inverse of its variance to obtain an overall estimate.

When examining the effect of psychotherapeutic treatment on PTSD symptoms, the authors examined the effect sizes of four randomized controlled trials (RCTs) separately from the effect sizes of two quasi-experimental designs (QEDs). However, CrimeSolutions Study Reviewers were able to combine the separate effect sizes and calculate an overall mean effect size for measures of PTSD symptoms. The Reviewers used the individual study-level effect sizes and confidence intervals that were reported by the authors in the review.