COVID-19 is a rapidly evolving situation. When working in different community settings, follow CDC guidance External Web Site Icon to help prevent the spread of COVID-19. Visit www.cdc.gov/coronavirus External Web Site Icon for the latest public health information.

TFFRS - Violence Prevention: Primary Prevention Interventions to Reduce Perpetration of Intimate Partner Violence and Sexual Violence Among Youth

PDF version pdf icon [PDF - 980 kB].

Return to the Intervention Summary.

Community Preventive Services Task Force Finding and Rationale Statement

Context

Intimate partner violence and sexual violence are public health problems that affect many men and women in the United States. While violence occurs across the lifespan, it often starts before the age of 18 (Smith et al. 2017), and the effects can include physical injury, substance abuse, poor mental health, and chronic physical health problems (Smith et al. 217, Coker et al. 2002). Adolescence represents a unique opportunity to promote attitudes and behaviors that could prevent intimate partner and sexual violence across the lifespan.

The public health approach to this problem focuses on preventing or reducing a person’s risk of committing violence, as preventing perpetration before it starts has the greatest potential to reduce population rates of violence (Cox et al.2010; DeGue et al., 2012; McMahon, 2000). Programs may target potential perpetrators, or bystanders—people close to a situation who can challenge violence-supportive norms by directly reducing risk (e.g., by noticing a risky social situation and intervening) or by indirectly reducing risk (e.g., by challenging hostile attitudes towards women such as offensive jokes or objectifying language). This review focused on interventions designed to reduce perpetration, though they also may have reported victimization outcomes.

Intervention Definition

Interventions for the primary prevention of intimate partner violence (IPV) and sexual violence (SV) among youth ages 12 to 24 years aim to prevent perpetration of IPV and SV and promote healthier relationships between peers and partners. Interventions included in this review provide information about IPV or SV warning signs or consequences. They may include one or more of the following strategies:

  • Teach healthy relationship skills
    • Promote social-emotional learning to enhance a core set of social and emotional skills including communication and problem-solving, empathy, and emotional regulation
    • Teach youth skills such as communication, conflict resolution, and stress management
    • Promote healthy sexuality by focusing on education that addresses sexual communication, sexual respect, and consent
  • Promote social norms that protect against violence
    • Challenge negative attitudes or beliefs that support violence (e.g., gender stereotypes)
    • Promote the role of bystanders in violence prevention
      • Promote bystander empowerment and education to teach youth how to intervene in situations involving potential IPV or SV
      • Engage men and boys as allies in prevention of IPV or SV
    • Engage parents and other caregivers in prevention
    • Use social marketing campaigns that incorporate multiple communication channels, such as mass media and social media, to promote social norms that protect against IPV or SV
  • Create protective environments
    • Improve school climate and safety
    • Modify physical and social environments of neighborhoods or communities (greening urban spaces, enacting alcohol policies)
    • Enact or enforce public policies with the potential to reduce risk for IPV or SV

Interventions may be implemented in schools (middle school, high school, or college), at home, in communities, or in a combination of settings. They may target groups at high risk for violence or the general population, both of which may include youth who have already experienced IPV or SV as a victim or perpetrator.

This systematic review did not include studies that focused exclusively on victimization risk reduction strategies such as self-defense.

Community Preventive Services Task Force Finding

The Community Preventive Services Task Force (CPSTF) recommends primary prevention interventions that aim to reduce perpetration of intimate partner violence and sexual violence among youth. This is based on sufficient evidence of effectiveness that these interventions decrease perpetration of both intimate partner violence and sexual violence.

Although this review included studies that provided educational information alone, this finding supports interventions that combine educational information about intimate partner violence and sexual violence with one or more of the following three strategies: (1) teach healthy relationship skills, (2) promote social norms that protect against violence, and (3) create protective environments.

  • Interventions that taught healthy relationship skills OR promoted social norms to protect against violence reported favorable and consistent decreases in perpetration.
  • Interventions that taught healthy relationship skills OR combined teaching healthy relationship skills with efforts to promote social norms that protect against violence, reported favorable and consistent decreases in victimization.
  • Interventions that used all three strategies in combination reported decreases in perpetration and victimization.
  • While some of the other strategy combinations led to favorable results, there were too few studies or inconsistent results across the body of evidence to draw conclusions about them.

The CPSTF also recommends interventions that promote social norms to protect against violence through bystander education and empowerment, engaging men and boys as allies in prevention, or both. This finding is based on a subset of studies that showed sufficient evidence of effectiveness in increasing bystander action in the short-term.

Rationale

Basis of Finding

The CPSTF recommendation is based on evidence from a systematic review of 28 studies (32 study arms). Twelve of the included studies were identified from two existing systematic reviews (Whitaker et al., 2013 [search period: 1993-2012] and DeGue et al., 2014 [search period: 1985 to May 2012]). The remaining studies were identified through a Community Guide search for evidence (search period: 2012 to June 2016).

Most of the included studies were randomized controlled trials (RCTs) that evaluated changes in perpetration, victimization, or bystander action using various outcome measures. Differences in outcome measurements and analyses precluded the calculation of summary effect estimates. Table 1 shows outcomes as favorable or unfavorable based on specific criteria (see Data Quality section below for more information about favorability).

Table 1: Behavioral Outcomes
Outcome Key Study Findings Summary

Perpetration
(21 studies, 24 study arms)

Favorable effects: 17 studies, 18 study arms
Statistically significant* effects:
12 studies, 13 study arms

No effect: 2 studies, 2 study arms

Unfavorable effects: 4 studies, 4 study arms
Statistically significant effects:
0 studies

Results showed favorable and consistent decreases in perpetration.

Victimization
(15 studies, 18 study arms)

Favorable effects: 11 studies, 11 study arms
Statistically significant* effects:
8 studies, 8 study arms

No effect: 3 studies, 3 study arms

Unfavorable effects: 3 studies, 3 study arms
Statistically significant effects:
0 studies

Results were inconsistent for decreases in victimization.

Bystander Action
(9 studies, 10 study arms)

Favorable effects: 6 studies, 6 study arms
Statistically significant* effects:
5 studies, 5 study arms

No effect: 1 study, 1 study arm

Unfavorable effects: 2 studies, 3 study arms
Statistically significant effects:
0 studies

Stratified by length of follow-up

≤6 month follow-up: 8 studies, 9 study arms

Favorable effects: 8 study arms
Statistically significant* effects:
7 study arms

No effect: 0 studies

Unfavorable effects: 1 study, 1 study arm
Statistically significant effects:
0 studies

>6 months follow-up: 4 studies, 5 study arms

Favorable effects: 2 studies, 2 study arms
Statistically significant* effects:
1 study, 1 study arm

No effect: 1 study, 1 study arm

Unfavorable effects: 1 study, 2 study arms
Statistically significant effects:
0 studies

Results showed favorable and consistent increases in bystander action ≤6 months after the intervention.

*Statistical significance p<0.05


Intervention Characteristics

Included studies evaluated interventions that provided educational information and used different combinations of the following intervention strategies: (1) teach healthy relationship skills, (2) promote social norms that protect against violence, and (3) create protective environments. Table 2 lists strategy combinations that were favorable and consistent along with examples of what was done. There were inconsistent results across studies or too few studies to draw conclusions about outcomes associated with other combinations of interventions.

Table 2: Summary of Strategy Combinations that had Favorable and Consistent Results
Strategy Result Example

Teach healthy relationship skills

Perpetration: 5 studies, 5 study arms
Favorable effects: 4 studies, 4 study arms
Statistically significant* effects:
2 studies, 2 study arms

Victimization: 4 studies, 4 study arms
Favorable effects: 3 studies, 3 study arms
Statistically significant* effects:
2 studies, 2 study arms

Exercises in social resilience aimed at body language, feeling, setting and respecting boundaries, intuition, making contact, standing up for oneself, and communication skills

Conflict management skills for dating

Promote social norms that protect against violence

Perpetration: 3 studies, 3 study arms
Favorable effects: 3 studies, 3 study arms
Statistically significant* effects:
3 studies, 3 study arms

Web portal modules that include interactivity, didactic activities, and episodes of a serial drama

Bystander education and empowerment

Teach healthy relationship skills
+
Promote social norms that protect against violence

Victimization: 8 studies, 8 study arms
Favorable effects: 7 studies, 7 study arms
Statistically significant* effects:
5 studies, 5 study arms

Socio-emotional learning programs to teach healthy dating skills (conflict resolution)

Interactive activities that address dating violence norms, gender stereotyping, conflict resolution

Teach healthy relationship skills
+
Promote social norms that protect against violence
+
Create protective environments

Perpetration: 2 studies, 2 study arms
Favorable effects: 2 studies, 2 study arms
Statistically significant * effects:
1 study, 1 study arm

Victimization: 2 studies, 2 study arms
Favorable effects:
2 studies, 2 study arms
Statistically significant * effects:
1 study, 1 study arm

Identification of hotspots coupled with an increase in staff presence in those areas

Social marketing strategies

School-based teen dating violence prevention curricula to enhance skills and attitudes consistent with promotion of healthy relationships and reduction of teen dating violence

*Statistical significance p<0.05


Included studies used different intervention approaches (specific programs, policies, or practices). The most commonly used approach was to teach healthy, safe dating and intimate relationship skills (14 studies, 15 study arms). The second most commonly used approach was to challenge negative attitudes or beliefs that support violence (9 studies, 9 study arms). While all of the approaches showed decreases in perpetration, it was not possible to determine which approaches were more effective because of the variety of effect estimates reported. Of the nine studies that used three approaches, eight reported decreases in perpetration and five of these were statistically significant (p<0.05). Of the ten studies that used more than three approaches, nine reported decreases in perpetration, and five of these were statistically significant (p<0.05).

Only two studies had an intervention duration longer than six months. Both reported decreases for perpetration and victimization (one had results that were statistically significant, p<0.05). Neither study reported on bystander action.

Applicability and Generalizability Issues

Intervention settings

Included studies were conducted in the United States (25 studies), Canada (2 studies), and the Netherlands (1 study). Studies were conducted in schools (5 middle, 6 high, 12 college, 1 middle and high), communities (2 studies), or homes (2 studies). Among studies that reported population density, ten were implemented in urban settings, and five were implemented in mixed settings (i.e., urban and suburban; urban and rural; or urban, suburban, and rural). One study was conducted in a rural setting, and no studies were conducted in suburban settings alone. For perpetration, studies were favorable and consistent regardless of country. All studies reporting on bystander action were from the United States and were found to be favorable and consistent when follow-up was conducted within six months of intervention completion.

Demographic characteristics

Study participants had a mean age of 16.5 years (16 studies) and were 53.0% female (21 studies). One study only included girls, and six studies only included boys. Four studies evaluated interventions implemented in low-income communities. Of the 25 studies conducted in the United States, 24 reported racial and ethnic distributions as follows: white (median 69.9%; 21 studies), black (median 16.1%; 20 studies), Hispanic (median 10.9%; 17 studies), Asian (median 6.9%; 12 studies), and American Indian (2.2%; 6 studies).

Sixteen studies stratified results by gender and reported interventions were favorable and consistent for both genders for perpetration, victimization, and bystander action. Three studies stratified by race/ethnicity and reported interventions were favorable for black students for perpetration and victimization and mixed for Hispanics for perpetration and victimization; four studies stratified by SES and were favorable for low SES populations for perpetration and victimization. None of the included studies stratified bystander outcomes by race/ethnicity or SES. Three studies reported mixed results for youth who were considered to be at high-risk for violence because they were in the foster care system or had experienced violence as a perpetrator or victim.

Ten studies that reported results for high school students showed interventions were favorable and consistent for decreasing perpetration and victimization; one of these studies also reported favorable results for bystander action. Twelve studies that reported results for college students showed interventions were favorable and consistent for decreasing perpetration, mixed for victimization, and favorable and consistent for increasing bystander action when observed within 6 months post-intervention. Five studies reported results for middle school students and showed mixed results for perpetration and victimization; no interventions measured bystander actions among this age group. One study evaluated a program that started with middle school students and followed them through high school. Results showed the program was effective at first follow-up and remained effective as the students moved into high school.

Data Quality Issues

Study designs included group (13 studies) and individual (9 studies) randomized control trials, pre-post with a comparison group (3 studies), other designs with concurrent comparison groups (2 studies), and repeated cross section with a comparison (1 study). Common limitations affecting this body of evidence were loss to follow-up and selection bias due to self-selection and convenience sampling.

Perpetration, victimization, and bystander action outcomes were based on self-reported data. Common limitations of self-reported data include recall bias and social desirability bias. Most studies attempted to address these limitations by using validated instruments that were frequently tailored to be culturally and age appropriate. The body of evidence included data from various measurement instruments and scales.

Favorability was based on consistency and magnitude of effect. Due to various outcome measures for perpetration and victimization, summary effect estimates could not be pooled quantitatively as they typically are for Community Guide systematic reviews. Because all effect estimates had to be looked at side-by-side narratively, the review team ensured a systematic synthesis process by considering each outcome’s results to be consistent if 75% or more of the studies reported a positive result. For each outcome, there was not a minimum number of studies required to determine consistency. For example, a body of evidence with two studies reporting positive findings and one reporting negative findings would be deemed inconsistent. Studies reporting multiple measures of an outcome (e.g., multiple measures of perpetration) were considered favorable if the majority of findings for that study were in the favorable direction. A change was considered favorable if it was greater than one unit of measurement.

Other Benefits and Harms

Several included studies provided information about possible benefits of these interventions. Coker et al. (2016) reported a reduction in any interpersonal violence among college students, and Taylor et al. (2010) reported a reduction in peer violence victimization and perpetration among middle school children. Another study (Wolfe et al., 2003) reported improved self-confidence among high-risk youth who were high school age. Studies from the broader literature suggest potential for improved academic outcomes using similar types of interventions (Brown et al., 2011; Durlack et al. 2011).

No specific harms were identified in the included studies, and the broader literature provided little information about potential harms of perpetration reduction interventions. One study in the broader literature suggests that bystanders may suffer psychological trauma (Witte et al. 2017). The CPSTF postulates that bystanders could inadvertently be physically harmed. However, included studies that evaluated bystander action reported favorable effects of the intervention on self-efficacy to intervene. This suggests interventions that promote bystander approaches can equip potential bystanders with knowledge and skills to appropriately intervene.

Considerations for Implementation

In the implementation of intimate partner and sexual violence prevention programs, the following issues should be considered:

  • Audience needs – Tailor programs and resources to the needs of youth with known risk factors (e.g., family history of violence, hyper masculine attitudes, aggressive peer relationships)
  • Program duration – Intervention effects diminish over time; follow-up may be needed to maintain positive changes, specifically for bystander action (e.g., reminders, intervention boosters)
  • Principles of prevention – Comprehensiveness of program , intervention dosage (intensity), training of staff, and utilization of multiple teaching methods can affect successful implementation of the program
  • Capacity to implement – Schools and communities should consider resources (e.g., money, personnel, community willingness) as they decide which intervention strategies to implement
  • Environment – School- or community-level changes may be necessary to create environments that support efforts to reduce perpetration and increase bystander action

Evidence Gaps

Several areas were identified as having limited information. Additional research and evaluation would help answer remaining questions or strengthen findings in these areas.

  • What combinations of intervention approaches are most effective?
  • What are the best measures for the outcomes of perpetration, victimization, and bystander action?
    • More consensus is needed to increase comparability across studies and the ability to synthesize evidence.
  • How effective are interventions that seek to create protective environments?
    • More interventions are needed that include components at the community and societal levels.
  • How effective are these interventions across different populations, including the following:
    • Students in rural settings?
    • Lesbian, Gay, Bisexual, Transgender (LGBT) youth?
    • Youth with intellectual and developmental disabilities?
    • High-risk youth?
  • How effective are age-appropriate interventions that target younger children (e.g., primary school-age children)? Rates of violence reported in the included studies suggest some students are already experiencing and perpetrating intimate partner violence and sexual violence in middle school. Age-appropriate interventions for elementary school students need to be developed and tested for later impact on SV and IPV outcomes.

References

Brown EC, Low S, Smith BH, Haggerty KP. Outcomes from a school randomized controlled trial of STEPS to RESPECT: a bullying prevention program. School Psychology Review 2011;40:423–43.

Coker AL, Bush HM, Fisher BS, Bush HM, Swan SC, et al. Multi-college bystander intervention evaluation for violence prevention. American Journal of Preventive Medicine 2016;50(3):295–302.

Coker AL, Davis KE, Arias I, et al. Physical and mental health effects of intimate partner violence for men and women. American Journal Preventive Medicine 2002;23(4):260–8.

Cox PJ, Ortega S, Cook-Craig PG, Conway P. Strengthening systems for the primary prevention of intimate partner violence and sexual violence: CDC's delta and empower programs. Journal of Family Social Work 2010;13(4):287–96.

DeGue S, Valle LA, Holt MK, Massetti GM, Matjasko JL, Tharp AT. A systematic review of primary prevention strategies for sexual violence perpetration. Aggression and Violent Behavior 2014;19:346–62.

DeGue S, Simon TR, Basile KC, Yee SL, Lang K, et al.Moving forward by looking back: reflecting on a decade of CDC's work in sexual violence prevention, 2000–2010. Journal of Women's Health 2012;21(12):1211–8.

Durlack JA., Weissberg RP, Dymnicki AB, Taylor RD, Schellinger KB. The impact of enhancing students’ social and emotional learning: a meta-analysis of school-based universal interventions. Child Development 2011;82(1): 405–32.

McMahon PM. The public health approach to the prevention of sexual violence. Sexual Abuse: A Journal of Research and Treatment 2000:12(1): 27–36.

Smith S, Chen J, Basile K, et al. The National Intimate Partner and Sexual Violence Survey (NISVS): 2010-2012 State Report. Atlanta, GA: National Center for Injury Prevention and Control, Centers for Disease Control and Prevention; 2017.

Taylor B, Stein N, and Burden F. The effects of gender violence/ harassment prevention programming in middle schools: a randomized experimental evaluation. Violence and Victims 2010;25(2):202–3

Whitaker DJ, Murphy CM, Eckhardt CI, Hodges AE, Cowart M. Effectiveness of primary prevention efforts for intimate partner violence. Partner Abuse 2013;4(2):174–95.

Witte TH, Casper DM, Hackman CL, Mulla MM. Bystander interventions for sexual assault and dating violence on college campuses: are we putting bystanders in harm's way? Journal of American College Health 2017;65:3:149–57.

Wolfe DA, Wekerle C, Scott K, et al. Dating violence prevention with at-risk youth: a controlled outcome evaluation. Journal of Consulting and Clinical Psychology 2003;71(2):279–91.