Key Findings from A Systematic Review of Primary Prevention Strategies for Sexual Violence Perpetration, by Sarah DeGue et al. National Sexual Violence Resource Center ©2014 Preventionists and their organizations, communities, and funders are increasingly seeking evidence-based strategies to prevent sexual violence before it starts. However, identifying effective strategies has been a challenge with little comprehensive information available about “what works” to prevent sexual violence. To address this gap, DeGue et al. (2014) systematically reviewed 30 years of evaluation research to describe the current state of the field in sexual violence primary prevention and to identify the strategies with the best available research evidence. Methods The authors searched for published journal articles, book chapters, and reports from government agencies or other institutions. They also tried to gather unpublished studies. To be included in the systematic review, studies needed to: • examine the effects of a primary prevention strategy for sexual violence perpetration in an outcome evaluation, • measure outcomes related to sexual violence, including risk factors and behaviors, • include males in the study, • include a comparison group and/or use a pre- and post-test research design, • be available in English, and • be published or completed between January 1985 and May 2012. A total of 140 separate studies were analyzed. The authors coded each study for key elements of the intervention, types of outcomes measured, effects on those outcomes, and evaluation design. Evaluations were considered “rigorous” if they used an experimental design, such as a randomized controlled trial (RCT), or certain quasi-experimental designs (Shuttleworth, 2008, 2009), and if follow-up evaluation data was collected at least once beyond an immediate post-test. Based on the quality of the evaluation design and the outcomes assessed, interventions were categorized as effective, not effective, potentially harmful, in need of more research, or as having insufficient evidence for preventing sexual violence. Results Key Findings The authors looked at patterns among the studies’ results to identify characteristics of prevention strategies and evaluation designs that may affect their outcomes. • One in four programs had no effect or were potentially harmful. Overall, 27.9% of the interventions had the desired (or “positive”) effects on all outcomes that were measured. 21.4% had no significant (or “null”) effects. 41.4% had mixed (a combination of positive and null) effects. Nine studies (6.4%) found at least one harmful (or “negative”) effect, such as an increase in sexually violent behavior or other risk factors. • Evaluation design matters. Studies that collected data beyond an immediate post-test, for example, were less likely to find consistently positive effects. Similarly, less than 5% of studies that measured sexually violent behavior found consistently positive effects. These findings indicate that positive effects found immediately after the intervention or on outcomes such as knowledge and attitudes may not hold up over time or prevent violent behavior. • One-hour educational programs in college aren’t enough. 60% of the studies evaluated one-session programs with college students. None of these have shown lasting effects on sexual violence risk factors or behavior. • Longer interventions may work better. Interventions with consistently positive effects were about two to three times longer, on average, than programs with mixed, negative, or null effects. • More and better evaluation research is needed. Due to limitations in their evaluation methods, 77.1% of the studies were categorized as having insufficient evidence to draw conclusions about program effectiveness for preventing sexually violent behavior. Findings about “what works” (and what may not) to prevent sexual violence perpetration, based on the best available research evidence, are summarized below. The paper includes a table with further details about the interventions and results (see Table 3, page 353). It is important to note that sexual violence prevention programming and evaluation continue to evolve. New strategies will be developed and evaluated over time. Future research will identify additional effective strategies. And further evaluation will reveal whether promising programs are effective across different populations, age groups, settings, and sexually violent behaviors. What works (so far)? Three interventions reduced sexually violent behavior in a rigorous evaluation, suggesting that they may be effective as part of a comprehensive approach to preventing sexual violence perpetration. As with any prevention strategy, it is important for preventionists to assess these strategies to make sure they are a good fit for their own communities and settings. Further, when adapting these strategies for different populations and contexts, it is important to consider how changes might influence a strategy’s effectiveness. • Safe Dates (Hazelden, 2014) is a dating violence prevention program for middle- and high school students. It includes a 10-session curriculum, a play, and a poster contest. Students were significantly less likely to self-report sexual violence victimization or perpetration in dating contexts four years after participating in the program. • Shifting Boundaries (National Institute of Justice, n.d.) building-level intervention is part of a dating violence prevention program for middle schools that involves environmental changes, a poster campaign, “hotspot” mapping, and school staff monitoring. There is also a classroom curriculum. The building-level intervention (but not the curriculum) reduced self-reported perpetration and victimization of sexual harassment and sexual violence among peers, and decreased self-reported sexual violence victimization (but not perpetration) by a dating partner. • The Violence Against Women Act of 1994 (VAWA) increased penalties for sexual and other gender-based violence and provided funding for research, education, prevention, and victim service programs. VAWA-funded criminal justice activities were associated with reduced rates of rape and aggravated assault reported to police. What (probably) doesn’t work, or might be harmful? Five interventions that were rigorously evaluated for their effects on sexually violent behavior showed no significant effects on those outcomes. This suggests that these interventions are not effective in reducing sexual violence perpetration. Some interventions in this category showed positive effects on other outcomes and/or positive effects on behavior using non-rigorous methods. Three interventions - two early versions of the Shifting Boundaries classroom curriculum and a set of sexual assault awareness videos - were associated with an increase in reports of sexual violence. Additional research could clarify whether this reflects an actual increase in sexually violent behavior or if reporting increased because participants’ awareness increased. Without additional information, preventionists may want to select other strategies. What else might work to prevent sexual violence? Ten interventions had positive or mixed effects on risk factors for sexual violence or related outcomes in a rigorous evaluation, but effects on sexually violent behavior were not measured. The authors indicate that while these findings are promising, more research is needed to know whether reducing these risk factors will ultimately reduce sexually violent behavior. They highlight and recommend more rigorous evaluation of two particularly promising programs that have substantial potential to prevent sexual violence perpetration: • Coaching Boys Into Men (Futures Without Violence, 2014), a social norms program for high school coaches and athletes, reduced dating violence perpetration. However, since the measure combined both physical and sexual violence, the specific outcomes related to sexual violence are unclear. • Bringing in the Bystander (Prevention Innovations, n.d.), a bystander intervention program, showed positive effects on factors related to sexual violence, including participants’ beliefs that they are able and likely to intervene as engaged bystanders. What does this mean for preventionists? This paper can help preventionists identify strategies to include in a comprehensive approach to prevention, strengthen and evaluate their existing prevention efforts, and advocate with funders, policymakers, researchers, and community partners. Prevention Identifying promising practices. Existing research can help preventionists select or develop prevention strategies based on the best available evidence. Yet critical gaps remain and the current list of effective programs does not address the needs of all populations and communities. Preventionists may need to draw from interventions that “work” and those that “might work,” incorporating the strengths of multiple promising strategies while ensuring that prevention activities are responsive to the unique strengths and needs of their own communities. Mobilizing the principles of prevention. Preventionists can use the principles of prevention (Nation et al., 2003) as guideposts when considering existing strategies or developing new ones. There is substantial research evidence that effective prevention strategies are comprehensive, appropriately-timed, culturally relevant, and theory-driven; they utilize varied teaching methods and well-trained staff; they foster positive relationships; and they are long enough (or have “sufficient dose”) to create meaningful and lasting change. Changing communities and society. To date, most sexual violence prevention programming has focused on influencing the individual and relationship levels of the social ecological model (Centers for Disease Control and Prevention, 2013). Little is known about effective approaches for impacting risk at the community and societal levels. Preventionists can lead the way in developing new interventions that address risk factors for violence across all these levels, to make sure our communities and larger societies promote and sustain equitable, safe, and violence-free environments. Evaluation Strengthening evaluation. Whether preventionists conduct their own evaluations, recruit help from professional evaluators, or team up with researchers, this paper highlights evaluation methods that can increase confidence that our prevention efforts are achieving our goals (and help us make adjustments when needed). While rigorous evaluation designs may not always be feasible in the field, preventionists should work towards using the strongest evaluation design possible to determine whether a strategy is actually working. Collecting longer-term follow-up data reveals whether a program creates lasting change. And measuring sexually violent behavior can help determine more directly whether a strategy “works” for preventing sexual violence. Advocacy Investing in prevention. Short, one-session programs are common, in part, because they are relatively inexpensive to implement and because many schools and other settings allow very little time for sexual violence prevention activities. Preventionists can draw on solid research evidence to advocate for the funding, resources, and community support that are needed to implement longer, multi-level, multi-strategy approaches that can truly change behavior and prevent sexual violence. Investing in evaluation. This paper emphasizes the need for more and better research on sexual violence primary prevention strategies. However, evaluating sexual violence prevention can be complex and costly, and the need for rigor competes with other priorities. This paper can support preventionists in advocating for the funding, training, time, and partnerships needed to identify cost-effective prevention strategies. Building partnerships. Only about one in four prevention strategies in this paper were implemented by professionals with expertise in sexual violence and its prevention. This may suggest a lack of researcher-practitioner partnerships in the field of sexual violence prevention. It may also be a sign that preventionists’ own evaluation findings about “what works” have not been widely shared through publication. Collaboration among researchers and preventionists is critical to bridge the gaps in what we know about effective prevention and to identify a range of strategies to help us build communities and a world free from sexual violence. References Centers for Disease Control and Prevention. (2013). The Social-Ecological Model: A framework for prevention. Retrieved from http://www.cdc.gov/violenceprevention/overview/social-ecologicalmodel.html. DeGue, S., Valle, L. A., Holt, M. K., Massetti, G. M., Matjasko, J. L., & Tharp, A. T. (2014). A systematic review of primary prevention strategies for sexual violence prevention. Aggression and Violent Behavior, 19, 346-362. doi:10.1016/j.avb.2014.05.004 Futures Without Violence. (2014). Coaching Boys Into Men coaches leadership program. Retrieved from http://www.futureswithoutviolence.org/section/our_work/men_and_boys/_coaching_leadership/ Hazelden. (2014). Safe Dates: Prevention program for dating abuse. Retrieved from http://www.hazelden.org/web/go/safedates Nation, M., Crusto, C., Wandersman, A., Kumpfer, K. L., Seybolt, D., Morrissey-Kane, E., & Davino, K. (2003). What works in prevention: Principles of effective prevention programs. American Psychologist, 58, 449-456. doi:10.1037/0003-066X.58.6-7.449 National Institute of Justice. (n.d.). Program profile: Shifting Boundaries. Retrieved from Crime Solutions: http://www.crimesolutions.gov/ProgramDetails.aspx?ID=226 Prevention Innovations. (n.d.). Bringing in the Bystander. Retrieved from the University of New Hampshire: http://cola.unh.edu/prevention-innovations/bystander Shuttleworth, M. (2008). Quasi-experimental design. Retrieved from Explorable.com: https://explorable.com/quasi-experimental-design. Shuttleworth, M. (2009). Randomized controlled trials. Retrieved from Explorable.com: https://explorable.com/randomized-controlled-trials Violence Against Women Act, 42 U.S.C. Chapter 136, Subchapter III (1994). Retrieved from the Government Printing Office: http://www.gpo.gov/fdsys/pkg/BILLS-103hr3355enr/pdf/BILLS-103hr3355enr.pdf