Abstract
Linked article:
1. Background for the Review
Estimates of the incidence and prevalence of rape and other forms of sexual assault vary depending on how terms are defined, what types of sexual assaults are included, the time frame during which the data are collected, sampling methods used, age and gender of the population studied, and the location of the sample (Acierno, Resnick & Kilpatrick, 1997). Further, as few countries have undertaken studies that comprehensively document the prevalence of rape and sexual assault, worldwide incidence is difficult to determine (British Council, 2006). In the United Kingdom, Painter (1991), reported that 1 in 4 women experienced rape or attempted rape. A national random sample of 6,926 women in Sweden found that 1 in 6, or 16% of the respondents, had experienced sexual violence by a former husband or cohabitant and that 1 in 4, or 25% of the respondents, had been subject to sexual violence by a man outside a relationship (Lundgren, Heimer, Westerstrand & Kalliokoski, 2002). In a similar study in Ghana, Coker-Appiah and Cusack (1999) randomly sampled 2,069 women and adolescent girls and found that 1 in 3, or 33% of the respondents, had been touched against their will and that 1 in 5, or 21% of the respondents, had been raped. In the United States, a population-based study of 1,769 women in Virginia discovered that the lifetime prevalence of sexual assault was 27.6% and the prevalence of rape was 17.8%. In a survey of households in Los Angeles, the lifetime prevalence of sexual assault was 13.2% (16.7% for women and 9.4% for men; Burnham et al., 1988). A similar study in Northeastern United States, reported a lifetime prevalence of sexual assault of 7.3% for women and 1.3% for men (Norris, 1992). A national random sample of women found that 12.7% reported a history of rape and 14.3% reported other forms of sexual assault (Resnick et al., 1993). Other US studies have focused on college populations. Koss, Gidcyz and Widniewski (1987) found in their survey of 6,000 students from 32 colleges, that 50% of the respondents indicated having experienced some form of sexual violence after age 14 and 27.5% reported having been raped (Koss, Gidycz & Wisniewski, 1987). In a subsequent study with 2,700 college women, 15% reported rape and 12% reported attempted rape since age 14 (Koss & Dinero, 1989). Gross and colleagues (2006) reported that 27% of a sample of college women had experienced some form of unwanted sexual contact (ranging from kissing and petting to intercourse) since entering college. Thus, while estimates vary, sexual assault affects a non-trivial percentage of the population.
Distress often occurs after a person has been exposed to a traumatic event such as sexual assault. Resick (1983) for example found that victims of rape exhibited more depressive symptoms, more fear and anxiety, more problems with social and work adjustment, and more problems with sexual functioning than did a control group of non-victims. The increased prevalence of Post-Traumatic Stress Disorder (PTSD) in victims of rape is well documented. PTSD has been reported to affect between 9 and 15 percent of the general population and almost 50 percent of individuals who have been raped (Treadwell & Foa, 2004).
As the negative effects of sexual assault have become better recognized, there is increasing attention to the possibility that psychosocial interventions may reduce suffering and limit distress. There are now several treatments modalities available for victims of rape and sexual assault including: pharmacology; behavioural techniques such as flooding, systematic desensitization, eye movement desensitization and retraining (EMDR); cognitive behavioural therapy; cognitive therapy; relaxation; rational-emotive therapy; group therapy; hypnosis; family/couple therapy; existential therapy; humanistic approaches; and psychodynamic therapy. A critical review of this literature reveals many articles based on clinical impressions, but few studies measuring outcomes of treatment. Due to the well documented negative consequences of sexual assault such as posttraumatic stress disorder (PTSD), and secondary depression (Resnick, Acierno, Holmes, Kilpatrick, & Jager, 2006), it is imperative that effective and efficient treatment options for these victims are identified (Rothbaum, Astin & Mosteller, 2005).
One of the issues in evaluating post-traumatic stress interventions is the degree to which symptoms spontaneously remit. For instance, Rothbaum, Foa, Riggs, Murdock and Walsh (1992) reported that, while 94% of the 95 rape victims in their study met the criteria for PTSD at one week post-rape, this reduced to 47% at 94 days post-rape. Once the three month marker occurs, it has been suggested that symptoms of PTSD become relatively persistent. While there are hundreds of original reports describing the effectiveness of treatments for individuals who have been exposed to traumatic events, the vast majority are not empirically based studies (Solomon & Johnson, 2002). Yet, the natural diminishing of symptoms of PTSD requires that controlled studies be considered when discussing efficacy.
Stein, Ipser and Seedat's (2006) systematic review of pharmacotherapy for PTSD is the only systematic review to examine the reduction of distress for victims of rape. The reviewers concluded that while medication treatments can be effective in treating PTSD, there continues to be a need for more effective strategies. However, in this review victims of sexual violence were grouped with other populations of trauma sufferers, and no specific data synthesis was reported regarding those who had been victims of rape. Within the psychotherapy literature, there have been several non-systematic reviews of treatment for rape and sexual assault (Falsetti, 1997; Foa, Rothbaum & Steketee, 1993; Foa & Rothbaum, 1998). Other reviews have examined the reduction of distress in populations experiencing symptomology of PTSD but these reviews were not specific to adult victims of rape and sexual assault. Bisson and Andrew (2007), for example, completed a Cochrane Review of psychological treatments to reduce symptoms of post traumatic stress disorder (PTSD). Although they found evidence that individual Trauma-focused cognitive behavioural therapy/exposure therapy (TFCBT), eye movement desensitisation and reprocessing (EMDR), stress management and group TFCBT were effective in the treatment of PTSD, the authors made no restriction on the basis of severity of PTSD symptoms or type of traumatic event. Instead, their study population included war veterans, female assault (mainly sexual assault) survivors, refugees and police officers, and mixed groups of individuals who had experienced a variety of traumatic events including road traffic accidents, assaults, bereavement and industrial accidents. Likewise, Sherman (1998) completed a meta-analysis of controlled and clinical trials of psychotherapeutic treatments for posttraumatic stress disorder (PTSD) and included samples of combat veterans from the Vietnam and Lebanon Wars, crime-related victims, and severe bereavement sufferers and victims of rape. Sherman examined cognitive, and psychodynamic treatments, in group and individual settings and found the overall impact of psychotherapy on PTSD and psychiatric symptomatology was significant (d = .52, r = .25) with a non-zero, 95% confidence interval suggesting that the true effect lies between .39 and .68, however no data were available specifically for those who were victims of rape.
It is important to evaluate treatments specifically for victims rape because although there may appear to be similarity of response to different trauma including rape, accidents and disasters (Rothbaum, Ninan & Thomas, 1996), there is evidence that trauma associated with rape may be different than other forms of trauma in part due to the strong element of self-blame, the higher incidence of concurrent depression and the increased risk of suicide (Connor, Jonathan & Davidson, 1997; Kimerling, Ouimette & Wolfe, 2002). Using a representative sample to assess the differential risks of PTSD across types of trauma exposure, Breslau, Davis, Andreski and Peterson (1991) reported that the incidence of PTSD was highest after rape (49%, SE=12.2) followed by other forms of sexual assault (23%, SE=10.8). The incidence of PTSD after being shot or stabbed was 15% (SE=13.7), a serious car accident was 2.3% (SE=1.3), other kinds of serious accidents was 16.8% (SE=6.2) and following natural disaster was 3.8(SE= 3.0). Given the high risk of PTSD following rape and other forms of sexual assault, there have been a growing number of treatment programs that specifically target victims of rape and other forms of sexual assault.
While treatment modalities may not be explicitly feminist, due to the gendered nature of a majority of sexual assault their implementation is often influenced by feminist frameworks and feminist theory. Feminist frameworks pay heed to re-establishing appropriate boundaries in relationships, beginning with the therapeutic relationship, promoting self-determination in the victimized person, and empowerment of the victim to move from victim to survivor. Feminist informed approaches view sexual victimization as a crime against the self and highlight the contributions of a society that condones violence (Bass & Davies, 1992; Myers Avis, 1992; Solomon, 1992). Feminist therapeutic approaches emphasize integrating the social causes of rape into the client's world-view and reducing self-blame and guilt following sexual assault (Enns, 1993; Koss & Harvey, 1991). The goal of such analysis is to help the victim understand that such violence is a societal problem, not an individual problem (Enns, 1993). Although elements of feminist theories may underpin different models of treatment there is a lack of clarity in defining a “feminist treatment model” precluding its inclusion in this systematic review as a distinct and separate modality.
Models of therapy used to assist victims in recovery from sexual assault and rape are predominantly based on two theoretical frameworks: the psychodynamic model and the cognitive behavioral model.
A. Psychodynamic Approaches to Intervention
From the psychodynamic perspective, traumatic events call into question basic human relationships and breached attachments to family, friends and the community. They shatter the construction of self that is formed and sustained in relation to others and force a survivor to relive earlier struggles over autonomy, identity and intimacy. Recovery requires re-establishment of a sense of self and relationships with others (Herman, 1992, 1997; van der Kolk, 1994; van der Kolk, McFarlane & Weisaeth, 1996).
Psychodynamic approaches are perhaps the oldest method for dealing with trauma in various forms including, sexual assault and rape. This is the area, however, in which there is the least available empirical research. Rather, the literature focuses on theoretical papers (Bohleber, 2007; Boulanger, 2002; Evans, 1978; Rose, 1991; Straker, Watson & Robinson, 2002), case studies (Barnett, 2001; Fosha, 2006; Friedberg, 1997; Pole & Bloomberg-Fretter, 2006; Wren, 2003) and clinical reflections (Schootenbauer, Arnfoff, Glass & Gray, 2006).
Cognitive-Behavioural Models of Treatment
Cognitive therapists concentrate on how traumatic events confront and challenge an individual's conceptualization of themselves and the world. When this occurs, customary problem solving mechanisms become overwhelmed and the individual is unable to process and integrate the experience. Treatment and healing thus focus on mastery of intrusive and disturbing images (Resick & Schnicke, 1993; McCann & Pearlman, 1990). Cognitive-behavioural models of treatment cover a range of specific approaches including Exposure Therapy or Prolonged Exposure (ET/PE), Stress Inoculation Training (SIT), Cognitive Processing Therapy (CPT), and Eye Movement De-Sensitization and Reprocessing (EMDR). These models are based on the premise that cognitive factors play a significant role in the onset, severity and outcomes of PTSD symptoms after sexual assault (Foa & Meadows, 1997; Foa, Steketee & Rothbaum, 1989; Koss & Figuerdo, 2004; Jaycox, Zoellner & Foa, 2002). Cognitive therapies therefore focus on the modification of affective and cognitive responses associated with the traumatic memory which will ultimately lead to decreased autonomic arousal and decreased intrusive imagery. In general these approaches involve systematic exposure to traumatic memories and cognitive reinterpretation of these events (Foa, Riggs, & Gershuny, 1995) and didactic and behavioral therapy techniques, including educating survivors about rape myths and trauma, and teaching them anxiety reduction techniques (Foa, Hearst-Ikeda & Perry, 1995; Foa & Rothbaum, 1998; Lubin, Loris, Burt & Johnson, 1998).
Of all treatments for victims of rape and sexual assault, cognitive-behavioural models (CBT) have been most rigorously evaluated. Compared with other types of therapies such as supportive psychotherapy, psychodynamic psychotherapy, or supportive counselling hypnotherapy, CBT has been found to be effective in reducing symptoms of PTSD (Foa, Rothbaum, Riggs, Tamera & Murdock, 1991; Follette, Ruzek and Abueg, 1998; Harvey, Bryant and Tarrier, 2003; Resick &Schnicke, 1992; Rothbaum and Foa, 1996). Subsequent research has focused on differentiating which models of CBT treatment might be more effective than others.
Foa and colleagues (1991), for instance, compared SIT and PE treatment approaches. They found that while SIT was the most effective treatment in reducing PTSD symptomatology immediately after treatment, PE was found to be most effective treatment at 3-month follow-up. Neither SIT nor PE was superior to supportive counseling or a waiting list in terms of other measures of fear, anxiety, or depression. It was, therefore, hypothesized that combining such approaches might enhance treatment benefit by providing ways to manage stress and anxiety while confronting the feared memories and cues (Foa et al., 1991). However, the evidence to support this combination has been generally mixed. For instance, Foa et al. (1999) compared PE, SIT, and a combination of PE and SIT with a wait-list control, and found that all three therapies significantly reduced symptoms of PTSD and depression relative to the control condition. PE appeared to improve general anxiety better than SIT or SIT-PE, whereas the results for PTSD and depression were similar for the three groups across a 1-year follow-up. Moreover, larger effect sizes for the reduction in PTSD severity and depression were reported for the PE treatment. The combined treatment, however, was delivered in the same number and length of sessions, meaning that clients in this condition received less imaginal exposure than the PE group and the SIT group (Foa et al., 1999). The authors conjectured that this provides a possible explanation as to why the combined approach did not outperform the single procedures treatments.
Cognitive processing therapy consists of two integrated components: cognitive therapy and exposure in the form of writing and reading about the traumatic event (Resick et al., 1993; Resick & Schnicke, 1992). Resick et al. (1992) argued that an approach that elicits memories of the events, and directly confronts conflicts and maladaptive beliefs might be more effective in alleviating PSTD symptoms than prolonged exposure alone. Prolonged exposure activates the memory structure, but does not provide direct corrective information regarding misattributes, which are common among rape victims (Resick & Schnicke, 1992). To test the effectives of CPT, the authors examined CPT in a group format in the treatment of rape-induced PTSD (Resick & Schnicke, 1992). The findings indicated that CPT was effective in improving symptoms in a large majority of participants in comparison to the naturally occurring wait-list. Furthermore, CPT resulted in statistically significant improvement in both PTSD and depressive symptomatology compared to the wait-list when implemented in a 12-session group format.
Resick et al. (2002) conducted a controlled trial, comparing CPT with PE and a minimal attention wait-list condition among victims of rape with regard to symptoms of PTSD and depression. The authors made great efforts not to introduce “causal cognitive therapy” in the PE protocol or to conduct prolonged imaginal or behavioural exposures in the CPT protocol (Resick et al. 2002). The authors’ findings indicated that CPT and PE were highly successful in treating PTSD in this sample of chronically distressed rape victims compared to the waiting list conditions. Both therapies were also found to be effective in treating depressive symptoms; however, CPT was found to be superior to PE in remediating guilt cognitions.
Eye-movement desensitization and reprocessing (EMDR) therapy, has received considerable attention amongst researchers; however, the findings have been mixed and have been a lack of comparison groups to assess its efficacy (Devilly & Spence, 1999; Ironson, Freund, Strauss & Greenwald, 2002; Lee et al., 2002; Taylor et al., 2003). Most studies have been methodologically weak with small sample sizes, and very few studies have been tested with the victims of sexual assault. Although several researchers state that the theoretical foundation has not been well developed (Rothbaum, Astin & Morsteller, 2005), EMDR shares similar traits to CBT exposure therapies (Rothbaum, 1997). For instance, during treatment, the client is to think of past/present traumatic experiences while concurrently focusing on a stimulus, such as auditory tones, tactile stimulation, and/ or visual cues (Rothbaum, 1997, 2005). As previously stated, it has been suggested that PTSD is due to an inability to adequately process the trauma and EMDR may be useful in this reprocessing (Foa, Steketee, & Rothbaum, 1989; Rothbaum, 1997, 2005). Rothbaum et al. (2005) reported that PE and EMDR had equally led to clinically and statistically significant improvements in rape victims with PSTD symptomatology immediately following treatment and at six-month follow-up compared to the wait-list control condition. Due to the non-random assignment utilized in this study, pretreatment differences on certain measures were present and possibly influenced findings.
In summary, some researchers have provided evidence that exposure therapy in combination with stress inoculation training or cognitive therapy yields the most positive results (Hembree & Foa, 2003), others have provided evidence that inoculation does not necessarily enhance other cognitive methods and provided alone, they are equally effective (Harvey, Bryant & Tarrier, 2003; Tarrier et al., 1999a,b). It is important to note that exposure methods tend to have higher drop-out rates and as indicated above those who drop out have been found to have higher rates of trauma symptoms. This may be due to the fact that in PE treatment, trauma reduction occurs through the process of first reactivating the fear memory and secondarily providing new information that is incompatible with the fear structure in order form a new memory (Resick & Schnicke, 1993; Rothbaum, Meadows, Resick & Foy, 2000). It is possible, that those with higher levels of symptoms are unable to tolerate the treatment and therefore discontinue. From a treatment standpoint, exposure methods are more selective in the criteria for inclusion and it is suggested that this model of treatment should be used only when a sound therapeutic alliance has been formed and a thorough assessment has been completed (Calhoun & Atkeson, 1991). Individuals in this type of treatment are assessed as having the capacity to tolerate high anxiety arousal, have no active suicidal ideation, no comorbid substance abuse, and, most importantly, no current life crises (Foy et al., 2000). Thus, if they are equally effective, CBT methods without exposure may yield a lower risk of iatrogenic effects. From a research perspective, these findings support the necessity to systematically review outcomes of distress post interventions for rape and sexual assault and they highlight the specific importance of evaluating the impact of selection bias and attrition on these results.
Summary
In short, there are few treatment options for sexually assaulted individuals that have been rigorously tested, and therefore, no certainty as to which intervention is superior in addressing PTSD symptoms and post-traumatic depression. It is important to note that the presented research has been conducted primarily with middle class white Anglo-Saxon women in the USA; it is unknown how these therapeutic approaches work for individuals from other cultures. Similarly, it is not known how these treatment options respond to the needs of victims of varying social classes, sexual orientations, psychological functioning and levels of ability/disability. It is not clear as well at this time about the possible variations by type of sexual assault (e.g., single vs. repeated assaults, known vs. unknown perpetrators, assaults under “normal” living conditions vs. those that happen during other traumatic events such as war). Thus, the therapeutic approaches reviewed to date may not be as effective for all types of adult sexual assault survivors. Further research is needed to understand which forms of treatment are indeed effective, and for whom.
Although the literature suggests there may be effective treatments for trauma and PTSD in general, there remains a substantial gap in the empirical evidence related to the effectiveness of various modalities to treat rape and other forms of sexual assault. By systematically reviewing the current state of interventions to reduce distress post rape and/or sexual assault, this review will contribute to a research agenda that develops evidence-based guidelines for the treatment of distress.
2. Objectives of the Review
There is no clear consensus among therapists and researchers on the best way to treat victims of sexual assault. In addition, most research has tested behavioural, cognitive or cognitive behavioral therapies. Therefore, the objective of this systematic review will be to complete an exhaustive and comprehensive search of controlled and clinical trials of psychotherapies for adult victims of sexual assault, and to synthesize the results of these studies to assess treatment effects on outcomes related to distress and trauma. Specifically, we will: Estimate absolute and relative effects of the combined group of psychotherapies on distress and trauma symptoms among adult victims of sexual assault Estimate absolute and relative effects of psychodynamic and cognitive-behavioral separately on distress and trauma symptoms among adult victims of sexual assault Assess variations in effects by type of treatment, duration of treatment, type of assault, and circumstances of assault
3. Methodology
Criteria for considering studies for this review
Types of studies
This review will include experimental and parallel cohort quasi-experimental evaluations of psychological intervention programs to reduce distress in adult victims of sexual assault. Studies will be eligible for the review if they 1) used random assignment to create treatment and comparison or control groups or 2) used parallel cohort designs in which groups were assessed at the same points in time. Single-group designs and single-subject designs will be excluded. The studies will vary in the method of constructing comparison/control groups and will also vary in their use of statistical controls to reduce the threat of selection bias. Only studies using parallel cohort design in constructing a comparison/control group will be included in the review. All studies will be included regardless of date. Electronic databases will be searched and key serials will be hand-searched.
Types of participants
Study participants include adults, 18 years of age and older, who have been victims of sexual assault (including rape) as adults. Males and females are included. Sexual assault is defined to encompass rape; attempted rape; forced oral sex, anal sex, penetration with objects, touching of intimate parts; and other types of threats or coercion in which unwanted sexual contact is attempted or occurs between the victim and offender. Rape refers to forced or attempted sexual intercourse with a male or female, by any offender.
Types of intervention
Relevant interventions are psychological or psychosocial in nature and will be compared with “other treatment”, “no treatment” and “treatment as usual”. Although we expect to find more controlled trials of cognitive behavioural therapy, a comprehensive search will be conducted to uncover all controlled investigations of effects of interventions to treat sexual assault victims. These interventions will include: behavioural techniques such as flooding, systematic desensitization, eye movement desensitization and reprocessing [reprogramming] (EMDR); cognitive behavioural therapy; cognitive therapy; relaxation; rational-emotive therapy; group therapy; hypnosis; family/couple therapy; existential therapy; humanistic approaches; and psychodynamic therapy. We will also identify “other” categories of interventions to ensure we remain flexible in the retrieval process of this review.
Since the main objective of this review is to compare psychosocial interventions to reduce distress, interventions that are exclusively based on pharmacology will be excluded. Studies that compare psychosocial treatments to pharmacological treatment and studies that combine psychosocial treatments with pharmacological treatments will be considered.
Types of outcome measures
The primary outcome is a decrease in trauma symptoms (PTSD). Secondary outcomes are: deceased levels of depression, anxiety, and guilt; and increases in positive functioning (e.g., social support, peer relations, locus of control, self-efficacy). In addition, analyses of outcomes will include any adverse negative affect of treatment (increased distress following treatment).
The primary outcome will be assessed in terms of the independent rating of severity of traumatic stress symptoms using a standardized measure such as the clinician administered PTSD Symptom Scale (Blake 1995) and severity of self-reported traumatic stress symptoms using a standardized measure such as the Impact of Event Scale (Horowitz 1979).
Secondary outcome measures will include: 1) severity of depressive symptoms using scales such as the Beck Depression Inventory (Beck 1961); 2) severity of anxiety symptoms using scales such as the Spielberger State Trait Anxiety Inventory (Spielberger 1973).
Search strategy for identification of relevant studies
Both published and unpublished work will be considered eligible for the review. A Trial Search Coordinator will be responsible for coordinating this activity. Currently we have the capacity to review English articles; we would appreciate any help from the Campbell Collaboration in extending the search to other languages.
Bibliographic Databases:
Cochrane Central Register of Controlled Trials (CENTRAL) Cochrane Database of Systematic Reviews MEDLINE EMBASE EMBASE Classic All EBM Reviews PsycINFO ASSIA (applied social sciences) ERIC Social Sciences Abstracts Social Services Abstracts Social Sciences Citation Index Criminal Justice Abstracts Violence and Abuse Abstracts Social Work Abstracts Dissertation Abstracts International (DAI) CINAHL Gender Studies Database Contemporary Women's Issues
Rape/ Sex Offenses/ (sex$ adj2 (abus$ or offens$ or attack$ or viol$ or assault$ or victim$ or surviv$ or unwanted or unlawful or forc$ or coerc$)).tw. (rape or raped or rapist or raping).tw. or/1-4 anxiety/ exp anxiety disorders/ anxi$.tw. ((post trauma$ or posttrauma$ or post-trauma$) adj (stress or neuros#s)).tw. ptsd.tw. exp mood disorders/ Depression/ depress$.tw. traum$.tw. distress$.tw. or/6-15 exp Psychotherapy/ (psychotherap$ or psychoeducat$ or psychodynam$ or psychoanaly$ or psychosocial or psycho-social).tw. ((behavio$ or cognit$ or general or social or supporti$ or interpersonal or group or individual or brief or psycho$ or dialectic$ or mindful$ or exposure or hypno$ or wilderness or emotion-focus$ or emotion focus$ or solution-focus$ or solution focus$ or narrative$) adj2 (counsel$ or support$ or intervention$ or program$ or treatment$)).tw. relaxation.tw. (eye movement$ or emdr).tw. desensitize$.tw. cbt$.tw. dbt$.tw. therap$.tw or/17-25 26 and 16 and 5
((DE=(Rape)) or(DE=(Sex Offenses)) or(TI=(sex* within 2 (abus* or offens* or attack* or viol* or assault* or victim* or surviv* or unwanted or unlawful or forc* or coerc*))) or(AB=(sex* within 2 (abus* or offens* or attack* or viol* or assault* or victim* or surviv* or unwanted or unlawful or forced or coerc*)))or(TI=(rape or raped or rapist or raping)) or(AB=(rape or raped or rapist or raping)))
and((DE=(Anxiety)) or(DE=(Anxiety Disorders)) or(TI=(anxi*)) or(AB=(anxi*)) or(TI=((post trauma* or posttrauma* or post-trauma*) within 1 (stress or neuroses or neurosis))) or(AB=((post trauma* or posttrauma* or post-trauma*) within 1 (stress or neuroses or neurosis))) or(TI=(ptsd)) or(AB=(ptsd)) or(DE=(Affective Disorders)) or(DE=(Depression)) or(TI=(depress*)) or(AB=(depress*)) or(TI=(traum*)) or(AB=(traum*)) or(TI=(distress*)) or(AB=(distress*)))
and((DE=(Psychotherapy)) or(TI=(psychotherap* or psychoeducat* or psychodynam* or psychoanaly* or psychosocial or psycho-social)) or(AB=(psychotherap* or psychoeducat* or psychodynam* or psychoanaly* or psychosocial or psycho-social)) or(TI=(behavio* or cognit* or general or social or supporti* or interpersonal or group or individual or brief or psycho* or dialectic* or mindful* or exposure or hypno* or wilderness or emotion-focus* or emotion focus* or solution-focus* or solution focus* or narrative*) within 2 (counsel* or support* or intervention* or program* or treatment*)) or(AB=(behavio* or cognit* or general or social or supporti* or interpersonal or group or individual or brief or psycho* or dialectic* or mindful* or exposure or hypno* or wilderness or emotion-focus* or emotion focus* or solution-focus* or solution focus* or narrative*) within 2 (counsel* or support* or intervention* or program* or treatment*)) or(TI=(relaxation)) or(AB=(relaxation)) or(TI=(eye movement* or emdr)) or(AB=(eye movement* or emdr)) or(TI=(desensitiz*)) or(AB=(desensitiz*)) or(TI=(cbt*)) or(AB=(cbt*)) or(TI=(dbt*)) or(AB=(dbt*)) or(TI=(therap*)) or(AB=(therap*)))
The SIRC will allow for replication of the search strategy because each search will be recorded. Furthermore, the search strategy will be saved and “copied and pasted” into the review to avoid editing errors.
In addition to CENTRAL and other bibliographic databases noted above, the following sources will also be searched for relevant studies:
Reference lists
Reviewers will check the reference lists of all relevant articles that are obtained, including those from previously published reviews. Potentially relevant articles that are identified will be retrieved and assessed for possible inclusion in the review.
Personal communication
Face-to-face discussions at meetings, emails, requests on list-serves, and formal letters of request for information from authors, presenters and experts will be solicited to assist the review team to locate relevant studies. A list of the inclusion criteria for the review, along with a sample of relevant articles, will be sent to these key informants along with the request for studies. The list of experts to be contact will include principle investigators of eligible studies, program developers, and authors of previous reviews of relevant literature such as Edna Foa, Barbara Rothbaum, Ann Burgess and Patricia Resick
Handsearching journals
Journals relevant to the sexual assault / rape / sexual violence will be hand-searched by researchers who will be trained to conduct handsearching to uncover relevant studies not found by electronic database searches. Specifically, we will hand-search Journal of Traumatic Stress; Journal of Interpersonal Violence; Victims and Offenders; Trauma Abuse and Violence; Violence Against Women; American Journal of Psychiatry; and British Journal of Psychiatry
Grey Literature
Special attention will be made to search and collect relevant studies captured in the grey literature. Specifically, the review will include the following strategies to locate articles: 1) Conference Proceedings; 2) Research Reports; 3) Government Reports; 4) Book Chapters; 5) Dissertations; 6) Policy Documents; 7) Personal Networks; 8) Research Organizations’ Web Sites; and 9) National Rape Crisis umbrella organizations. Grey Literature web-based sites will be searched to uncover this unpublished literature, such as Grey. Net (http://www.greynet.org/index.html) and GrayLit Network (http://graylit.osti.gov/).
Description of methods used in primary research.
Selection of studies
Titles and abstracts of studies yielded by the searches will be screened by two independent reviewers to determine their eligibility for inclusion in the review. To facilitate this screening, we will use TrialStat's Systematic Review System (SRS) version 3.0, which is an online tool designed to facilitate and accelerate the execution of systematic reviews. This tool, which is being used by several Evidence-based Practice Centres in Canada and the United States, enables rapid generation of forms, virtual collaborative communities, automated management of the progression of studies, and real-time data monitoring and reporting [http://www.trialstat.com/srs.htm]. TrialStat! has the advantage of being web-based, where each member of the review team has a secure login and password to enter the program. This allows reviewers to access the program on the world wide web, thus facilitating the ongoing collaboration between C2 Canadian and Nordic partners.
The screening of the studies will be carried out by a three-stage procedure. The screening points will be established in ‘level’ format whereby each level consists of increasing scrutiny of the studies based on the inclusion and exclusion criteria of the review.
1) Initial Screening (level 1)
The first stage will consist of an initial screening to quickly determine whether a study might be appropriate for the review based on the study's title and abstract. If there is not enough information in the title and abstract to make such decisions, then full text articles will be retrieved. Any disagreements will be resolved by retrieving the full text. The purpose of this initial screen will be to include all possible relevant studies related to the objectives of the systematic review and the inclusion and exclusion criteria. Level one screening will consist of the following questions: Does the population consist of adults who have experienced rape and or sexual assault as adults? Yes / No Is there an intervention related to rape and or sexual assault experienced by the population included in the study? Yes / No Is there a parallel cohort (comparison or control group)? Are the outcomes related to distress? Yes / No
2) Strict Screening (level 2)
The second stage will consist of a strict screening process where two reviewers will independently review full copies of articles to determine whether studies should remain in the review based on the inclusion and exclusion criteria. Specific reasons for exclusion at this stage will be documented for each study. Any disagreements will be resolved by a third reviewer. Level two screening will consist of the following questions: Did the study evaluate the intervention administered to adults, 18 years of age or older? Yes / No Did the incident(s) of rape or sexual assault occur when the population was 18 years of age or older? Yes / No Did the evaluation use an experimental or parallel cohort research design? Yes / No Did the study utilize a comparison group which included a no treatment, minimal treatment, alternative treatment, or usual services? Yes / No
3) Data extraction and management (level 3)
The third stage consists of a data extraction form to record data from the articles that have made it past the two previous screenings. The study details will be extracted, using a data extraction sheet, by two independent reviewers (see Appendix 2). Differences between coders will be identified and resolved to ensure consistent extraction and management of the data and to establish inter-rater reliability. Any discrepancies will subsequently be resolved by referral back to the source of the material and conflicts will be resolved by a third reviewer based on the original source. If necessary, we will seek additional information from the original investigators.
Details to be extracted will include: Study: information regarding the author(s); year of publication; source; country; and language Characteristics of Setting and Participants: eligibility criteria for participants; explanation of recruitment procedures, setting (country, location, clinical/non clinical); demographic features of the sample Sampling: sample sizes for treatment and control; whether power analysis was used to determine sample size; allocation to the treatment and control; explanation of method used to generate the allocation Research Design: nature of research design Intervention Data: nature of interventions (for treatment and comparison/control groups); aim of intervention; length of intervention, whether manuals were used, whether fidelity checks were included, information on possible contamination reported Outcome Data: primary and secondary outcomes, measures used, information on reliability/validity of measures Results: attrition at post intervention and follow-up; number excluded from the analysis; length of follow-up; statistical methods; type of data effect size is based on; data needed for effect size calculations.
Quality assessment
Two reviewers will independently assign each selected study to quality categories. The assessment of the methodological quality of each study will be based on criteria established in the Cochrane Collaboration Handbook (Higgins, 2005) and will focus on the generation of allocation sequence, degree of allocation concealment, intention to treat, blinding of assessors, attrition, and reporting of the data (see data extraction form).
Criteria for determination of independent findings
The determination of independent findings will be completed using the following procedures: First, studies may have included more than one measure of the distress (i.e., trauma symptoms, post traumatic stress disorder, depression, stress, anxiety); therefore, to ensure statistical independence of study findings, each measure of distress will be analyzed separately. In addition, when different scales and measurement procedures are used to measure sub-classifications of distress such as PTSD, the results of each scale and measurement procedure will be analyzed separately before exploring the possibility of combining the effect sizes as a standardized mean difference (SMD).
Multiple independent effect sizes within a study will only be included only if 1) effect sizes are reported separately by gender; 2) separate effect sizes are available across different study sites within a study.
Multiple outcomes for dependent or overlapping samples (i.e., multiple treatments compared against one control group) will be coded separately. We will randomly select only one effect size for inclusion in the meta-analysis.
For studies that include multiple follow-ups, these will be divided into separate intervals (i.e. effects within 6 months, 6-12, 12-24, more than 24 months) and we will do separate meta-analyses for each separate interval.
All available reports and data will be used in this review. Authors will be requested to provide all reports and the internet will be searched for additional reports (i.e. google). Any discrepancies will be investigated with authors.
Missing data
In cases where data (e.g., subgroup means and standard deviations, valid Ns) are missing, we will contact the author(s) of the primary studies and try to obtain missing information. Data on excluded subgroups (e.g., program drop-outs) will be sought as well.
Data synthesis
Data synthesis will be conducted using RevMan 4.2.8 and with Comprehensive Meta-analysis 2.0.
For binary outcome data, effect sizes will be calculated as odds ratios (OR) with 95% confidence intervals. Continuous data will be converted into standardized mean differences (SMDs) and presented with 95% confidence intervals. When necessary, we will use formulas suggested by Lipsey and Wilson (2001) to convert correlation coefficients, F ratios, t-values, and chi-square values into SMDs. Hedges’ g will be used to correct for small sample bias. If there are binary and continuous measures of the same outcomes (within or across studies), we will use procedures recommended by Sanchez-Meca and colleagues to convert odds ratios (OR) to d. We will also inspect the data to ensure there is no clear evidence of skewness in the distribution.
We will use unadjusted means whenever possible, unless there is evidence of baseline differences or selection bias. If primary studies only report adjusted results (i.e. regression coefficients or analysis of covariance), we will obtain advice from the C2 methods group on options for handling these effect sizes.
Meta-analysis
The random effects model will be used for pooling results. We assume that there will be unexplained sources of heterogeneity across studies; hence assumptions of the fixed effect model (that all studies provide estimates of a single population effect size) are untenable. Results for randomized experiments and quasi-experimental designs will be pooled and reported separately. As well, results for conceptually-distinct outcomes will be reported separately. If a study reports two separate measures of the same outcome, an average of two or more measures of the same outcome will be utilized. An average score for each group across measures will be computed, pooled sd for each group across measures will be computed, and average scores between groups will be computed. Results of each outcome measure will be reported in a table, but average scores will be reported in the Forest plots.
As indicated in this protocol, two main types of treatment that will be reviewed are psychodynamic psychotherapies and cognitive behavioural approaches. The research literature on the use of psychodynamic and psychotherapy for the treatment of sexual assault and rape is relatively sparse and thus pending a review of the method of treatment utilized, these approaches will be combined in the analysis. Cognitive-behavioural approaches to treatment post-rape and sexual assault have been more widely researched. There is some evidence that a distinction exists between those approaches that incorporate exposure elements and those that do not. That is, those which incorporate exposure have greater exclusion criteria and higher drop out rates when compared to those using CBT without exposure. Thus, CBT models will be divided on this criteria.
“Sub-group analyses of the effects of the different types of intervention (CBT, psychodynamic) will be conducted with a sufficient number of studies (2 or more) per category. If a study reports more than one modality as part of the intervention, it will be categorized as the most predominant intervention type. In the case of overlapping intervention types (i.e. CBT plus psychodynamic), further sub-group analyses will be conducted. For example, we will look at doing subgroup analyses with the following types of intervention: 1) CBT, 2) psychodynamic, 3) CBT plus psychodynamic.”
We will keep analyses of absolute effects and relative effects separate. Analyses of absolute effects include studies that use no-treatment or wait-list controls. Analyses of relative effects include studies that compare two different treatments. Results will never be pooled across these two types of studies.
Assessment of heterogeneity
Statistical heterogeneity in the outcome measures will be assessed using the Q-statistic and the associated p-value for each analysis and the I2 statistic (Higgins, 2002). The I2 statistic will determine the percentage of variability that is due to heterogeneity where a value greater than 50% suggests moderate heterogeneity.
Moderator analysis
To the extent possible, methodological and clinical heterogeneity among studies will be explored in terms of variations associated with overall study design (experimental and quasi-experimental designs), participant recruitment (i.e. medical clinics, shelter homes, counselling agencies), baseline characteristics (i.e. age, gender, culture, education, SES), intervention types (psychodynamic, CBT), duration of intervention, setting, country, comparison condition, duration of follow-up, and outcome measures (i.e., different measures of a single outcome). If possible, we will also examine effects of interventions with different subpopulations (by gender, types of victimization, length of victimization). Moderator analysis will be performed using the ANOVA analog (for categorical moderators) and/or meta-regression (for continuous moderators).
If reported in sufficient studies (i.e. two or more) moderator analyses will be conducted to examine the impact of the following variables as potential moderators on the effectiveness of the intervention: victim sexual orientation, single vs. multiple attackers, additional vulnerabilities (i.e. disability, elderly, health risks), prior history of victimization, combat/military context, and victim demographics (i.e. age, ethnicity, socioeconomic status).
Sensitivity Analysis
Sensitivity analysis will be performed to assess the robustness of conclusions to quality of data and approaches to analysis. Sensitivity analysis will be performed by reanalysis, excluding studies with poor quality indicators (e.g., high attrition, differential attrition, lack of intent-to-treat analysis, lack of controls for baseline differences). Publication and small sample bias will be assessed with graphical inspection of funnel plots, and “trim and fill” methods that estimate treatment effect by adjusting for the number and outcomes of missing studies (Sutton, Duval, Tweedie, Abrams, & Jones, 2000)
Treatment of qualitative research
Qualitative research will not be included in this systematic review.
4. Timelines
5. Plans for Updating the Review
Cheryl Regehr, Ramona Alaggia and Michael Saini will be responsible for updating this review. We anticipate that this review will be updated as we accumulate new evidence and/or at least every two years.
Footnotes
6. Acknowledgements
We would like to thank Aron Shlonsky and Faye Mishna for their input on this protocol. We would also like to thank April Mazzuca, MSW student for her contributions.
7. Statement Concerning Conflict of Interest
Although the authors of the review have contributed to research on the topic, they have no vested interest in the treatments that are the subject of this review or in the outcome of the review. The authors also do not have any incentives to represent findings in a biased manner.
Appendix A
Appendix B
1
Adapted from Mitchell, Wilson & MacKenzie (2005)