The CDC's relationship-violence prevention work

The CDC's involvement in violence as a public health problem dates to the early 1980s, when Mark Rosenberg and James Mercy began publishing on homicide as a leading cause of years of life lost. The Violence Epidemiology Branch was formally established in 1983, expanded into the Division of Injury Control in 1986, and reorganized in 1993 into the National Center for Injury Prevention and Control with a dedicated Division of Violence Prevention. The institutional history matters because each step required congressional acquiescence at a time when violence research was politically suspect, particularly when it touched firearms. The agency's caution about messaging was learned from these fights.

The National Intimate Partner and Sexual Violence Survey, launched in 2010 and continued in 2011, 2015, and beyond, gives the country its main ongoing measure of relationship violence prevalence. The survey defines forms of violence using Saltzman's definitions, allowing comparability across years. NISVS revealed that roughly one in three women and one in four men experience contact sexual violence, physical violence, or stalking by an intimate partner during their lifetimes, with severe forms concentrated among women. The numbers are higher than most policymakers expected. The survey's contribution is making the numbers public and consistent enough that arguments have to engage them.

Preventing Intimate Partner Violence Across the Lifespan organized the evidence into six strategies, each with specific approaches and an evidence rating. The package distinguished "best available evidence" interventions like Safe Dates and Dating Matters from "emerging" approaches like bystander programs and policy strategies like earned income tax credits. The grading is important: it signals to local implementers what they should expect from each choice. Programs that adopt the package's high-evidence interventions are eligible for federal funding through the Rape Prevention and Education program and other streams. The package is policy as well as science.

Dating Matters was designed by the CDC team including Phyllis Niolon and tested from 2012 to 2016 in four cities. The program combined classroom curricula for 6th through 8th graders, parent programs, communications campaigns, training for educators, and policy work. The comparison condition was the Safe Dates curriculum delivered in 8th grade only—itself an evidence-based program. The full Dating Matters condition produced reductions in physical, emotional, and sexual dating violence perpetration and victimization that persisted into high school. The effect sizes were modest (around fifteen percent relative reduction on most measures) but the design was rigorous and the persistence is unusual in the field.

Critics have asked why the effects are not larger given the intensity of the intervention. The answer is structural: dating violence in adolescence is shaped by family, neighborhood, media, and peers, not just school programs. A program operating in one of those settings can shift the school-level contribution but cannot overwhelm the others. The ecological model predicts exactly this pattern. Researchers who expect dramatic effects from school-only interventions have not absorbed the model. Researchers who write off school interventions because the effects are modest have not absorbed the scale: a fifteen percent reduction across a city, sustained, is thousands of fewer assaults.

The RPE program, authorized under the Violence Against Women Act and administered by the CDC, funds state health departments to implement primary prevention. The grants total roughly fifty million dollars per year nationally—small relative to the criminal justice spending on related problems, large relative to other CDC prevention lines. The grants require alignment with the technical package's evidence-based strategies. State-level evaluation has been inconsistent, and the GAO has flagged that the CDC's monitoring of state grantees does not always confirm that funded programs match evidence-based designs.

From 1996 to 2018, the Dickey Amendment effectively prohibited the CDC from funding research that could be construed as advocating gun control. Intimate partner homicide research, where firearm access is the single strongest predictor of lethality (Campbell's work), was constrained for over two decades. The amendment was clarified in 2018 to allow research, and small appropriations followed in 2020. Two decades of evidence that could have been produced were not. The collective cost is invisible because the counterfactual—what we would have known—is unknowable. The structural lesson is that scientific agencies are politically constrained even when the science is uncontroversial.

The CDC published Connecting the Dots in 2014 to formalize the connections among forms of violence and to argue for shared prevention strategies. The framework drew on adverse childhood experiences research (Felitti, Anda) and on the violence-cluster epidemiology developed at the agency. The practical effect was that prevention grants began rewarding programs that addressed multiple forms of violence rather than competing for siloed funding. Programs based in early childhood, school climate, and economic support could be funded under any of several violence prevention lines.

Green Dot, Bringing in the Bystander, and related programs train community members to intervene in situations that could escalate to sexual or relationship violence. The CDC has supported evaluation of these programs at high school and college levels. Results are mixed: clear effects on bystander behavior and attitudes, smaller and less consistent effects on perpetration. The agency has classified bystander approaches as "promising" rather than "best available evidence," which has frustrated advocates whose programs depend on the model. The grading is defensible. The field needs the agency to maintain the distinction.

Effective programs identified in the technical package do not always reach the populations they were designed to serve. Schools adopt whatever curriculum the district has licensed, regardless of evidence. Health departments fund whatever programs have local advocates. The CDC's role is to produce evidence and to fund implementation, not to mandate adoption, which would exceed its authority. The result is a gap between what is known to work and what is delivered. Closing the gap is a Law 5 problem that the agency cannot solve alone; it requires state legislatures, school boards, and funder behavior to align with the evidence.

The technical package's inclusion of strategies like the earned income tax credit and paid family leave as intimate partner violence prevention was unusual for a violence prevention document. The evidence is strong but mostly indirect: economic stress predicts relationship violence, and policies that reduce economic stress reduce violence at population level. The strategy implies that violence prevention extends beyond programs and into fiscal policy. It also implies that traditional DV advocacy, which has focused on shelter and legal services, has been undercapitalized in its analysis. The frame is forcing a broader conversation than the field is yet ready to have.

Federal agencies operate within messaging constraints that academic researchers do not face. The CDC will publish that firearm access is a major risk factor for intimate partner homicide; it is more cautious about recommending specific firearm policies. It will publish that comprehensive sex education reduces dating violence; it does not advocate for state curriculum mandates. The constraint is not always wrong—agency credibility depends on staying within its remit—but it produces a gap between what the science supports and what the agency says. Filling the gap is the work of researchers, advocates, and elected officials. The agency provides the evidence; what the collective does with it is not the agency's choice to make.

1. Mercy, James A., Mark L. Rosenberg, Kenneth E. Powell, Claire V. Broome, and William L. Roper. "Public Health Policy for Preventing Violence." Health Affairs 12, no. 4 (1993): 7–29.

2. Niolon, Phyllis Holditch, Megan C. Kearns, Jenny Dills, Kirsten Rambo, Shalon Irving, Tracy Armstead, and Leah Gilbert. Preventing Intimate Partner Violence Across the Lifespan: A Technical Package of Programs, Policies, and Practices. Atlanta: Centers for Disease Control and Prevention, 2017.

3. Black, Michele C., Kathleen C. Basile, Matthew J. Breiding, Sharon G. Smith, Mikel L. Walters, Melissa T. Merrick, Jieru Chen, and Mark R. Stevens. The National Intimate Partner and Sexual Violence Survey: 2010 Summary Report. Atlanta: Centers for Disease Control and Prevention, 2011.

4. Niolon, Phyllis Holditch, Allison J. Tracy, Sarah DeGue, Bruce Taylor, Andra Teten Tharp, Kathleen C. Basile, Linda Anne Valle, et al. "An RCT of Dating Matters: Effects on Teen Dating Violence and Relationship Behaviors." American Journal of Preventive Medicine 57, no. 1 (2019): 13–23.

5. Campbell, Jacquelyn C., Daniel Webster, Jane Koziol-McLain, Carolyn Block, Doris Campbell, Mary Ann Curry, Faye Gary, et al. "Risk Factors for Femicide in Abusive Relationships." American Journal of Public Health 93, no. 7 (2003): 1089–97.

6. Saltzman, Linda E., Janet L. Fanslow, Pamela M. McMahon, and Gene A. Shelley. Intimate Partner Violence Surveillance: Uniform Definitions and Recommended Data Elements. Atlanta: Centers for Disease Control and Prevention, 1999.

7. Wilkins, Natalie, Beverly Tsao, Marci Hertz, Rachel Davis, and Joanne Klevens. Connecting the Dots: An Overview of the Links Among Multiple Forms of Violence. Atlanta: Centers for Disease Control and Prevention, 2014.

8. Sumner, Steven A., James A. Mercy, Linda L. Dahlberg, Susan D. Hillis, Joanne Klevens, and Debra Houry. "Violence in the United States: Status, Challenges, and Opportunities." JAMA 314, no. 5 (2015): 478–88.

9. DeGue, Sarah, Linda Anne Valle, Melissa K. Holt, Greta M. Massetti, Jennifer L. Matjasko, and Andra Teten Tharp. "A Systematic Review of Primary Prevention Strategies for Sexual Violence Perpetration." Aggression and Violent Behavior 19, no. 4 (2014): 346–62.

10. Foshee, Vangie A., Karl E. Bauman, Susan T. Ennett, G. Fletcher Linder, Thad Benefield, and Chirayath Suchindran. "Assessing the Long-Term Effects of the Safe Dates Program and a Booster in Preventing and Reducing Adolescent Dating Violence Victimization and Perpetration." American Journal of Public Health 94, no. 4 (2004): 619–24.

11. Mercy, James A., Susan D. Hillis, Alexander Butchart, Mark A. Bellis, Catherine L. Ward, Xiangming Fang, and Michael L. Rosenberg. "Interpersonal Violence: Global Impact and Paths to Prevention." In Disease Control Priorities, 3rd ed., vol. 7. Washington, DC: World Bank, 2017.

12. Silverman, Jay G., Anita Raj, Lorelei A. Mucci, and Jeanne E. Hathaway. "Dating Violence Against Adolescent Girls and Associated Substance Use, Unhealthy Weight Control, Sexual Risk Behavior, Pregnancy, and Suicidality." JAMA 286, no. 5 (2001): 572–79.