Economics of Violence Prevention
A free program focuses on a problem that costs US taxpayers billions.
Violence has a tremendous financial burden on society that is mostly paid for by the public sector (i.e. your tax dollars at work). The economic toll related to violence calls for the development and testing of violence prevention interventions. These can be risk-based or population-based. Traditionally, focus has been on risk-based interventions that provide important services to those who have been identified as being at high risk of violence (e.g. children with a history of aggression).
To avoid stigmatization and to ensure that no one is missed, population-based services are being developed and tested. Here we will discuss the economics of violence prevention and challenges of conducting population-based violence prevention research on a brief intervention.
In the United States, cost estimates of interpersonal violence are 3.3 percent of the gross domestic product (GDP).* It is estimated that the annual costs of child abuse alone is 1 percent of the US GDP.* Thus, a population-based intervention that results in a relatively small reduction in the rate of interpersonal violence could result in a large cost savings for the public sector.
United States gross domestic product in 2005: $12,000,000,000,000 ($12 trillion).
Estimated cost of child abuse in the US in 2005 (1 percent of GDP): $120,000,000,000 ($120 billion)
Estimated cost of interpersonal violence in US in 2005 (3.3 percent of GDP): $400,000,000,000 ($400 billion)
Potential cost savings of reducing interpersonal violence by 10 percent: $40,000,000,000 ($40 billion)
Potential cost savings of reducing interpersonal violence by 1 percent: $4,000,000,000 ($4 billion)
Potential cost savings of reducing interpersonal violence by 0.1 percent: $400,000,000
Potential cost benefit of a population-based intervention
Note: The Play Nicely program can be viewed at no cost. Any costs would be administrative.
Number of children born in the US every year: 4,000,000 (4 million).
Assume the administrative cost of delivering a violence prevention intervention to the parents of every newborn child in the US was $20 per family: $80,000,000 ($80 million = $0.08 billion).
Assuming that a $20 population-based intervention could reduce the rate of violence by 10 percnet, the potential cost benefit would be $500 saved for every $1 spent ($40 billion/$0.08 billion).
Assuming that a $20 population-based intervention could reduce the rate of violence by 1 percent, the potential cost benefit would be $50 saved for every $1 spent ($4 billion/$0.08 billion).
Assuming that a $20 population-based intervention could reduce the rate of violence by 0.1 percent, the potential cost benefit would be $5 saved for every $1 spent ($400 million/80 million).
Violence prevention research outcomes
Our efforts have focused on assessing the efficacy of a brief educational program, Play Nicely, that addresses one of the strongest risk factors of violence, persistent early childhood aggression. Although it would be best to conduct studies that test whether a violence prevention program can actually reduce rates of violence, the sample size required for such conclusive studies is prohibitively large.
Rates of violence
It would be best to implement population-based interventions only after randomized controlled trials have demonstrated a reduction in the rate of violence in an intervention group which received the intervention compared with a control group which did not. However, the sample size needed to enroll participants in a randomized control trial designed to assess a small shift in the rate of violence is prohibitively large.
Reducing the violent crime rate
The U.S. violent crime rate is 500 violent crimes per 100,000 inhabitants. Assuming that a researcher would like to detect a reduction of 10 percent in the rate of violent crime in an intervention group compared with a control group, a study would need to enroll 300,000 participants at birth and follow them through early adulthood (alpha = 0.05 and beta = 0.8). Assuming that a researcher would like to detect a reduction of 1 percent in the rate of violent crime in an intervention group compared with a control group, a study would need to enroll 30,000,000 participants at birth and follow them through early adulthood (alpha = 0.05 and beta = 0.8).
Reducing Child Maltreatment
The rate of child abuse is difficult to measure exactly because of issues related to defining what constitutes abuse and how often it is reported; many experts feel that it is around 1 percent. Assuming a child maltreatment rate of 1 percent and that one would like to test whether a brief population-based intervention could reduce the rate of child maltreatment by 10 percent, a study would need to enroll 150,000 participants (alpha = 0.05 and beta = 0.8). Subsequently, decisions to implement population-based interventions that attempt small, but important, reductions in violence by addressing childhood aggression will likely need to be based upon studies that look at outcomes other than rates of violence later in life.
Another excellent outcome to consider is conduct disorder (i.e. persistent childhood aggression) in young children because it is a risk factor for violence later in life. However, enrolling enough participants to detect a small, but important, effect on the rate of conduct disorder will still be challenging. The rate of conduct disorder is higher in boys and varies between 1 and 10 percent. For this sample size estimate, we will assume a rate of conduct disorder of 5 percent. Assuming that one would like to be able to detect a reduction of 10 percent in the rate of conduct disorder in an intervention group compared with a control group, a study would need to enroll 28,000 participants at birth and follow them through early childhood (alpha = 0.05 and beta = 0.8).
Sample size requirements limit measuring rates of violence and conduct disorder for brief interventions such as Play Nicely. Until larger studies can be conducted that measure these important outcomes, researchers will need to focus on other research questions related to childhood aggression. Some of these might include:
- Is the program embraced (i.e. high satisfaction) by those with different backgrounds?
- Can the program increase comfort level with managing aggression?
- Can the program increase knowledge of aggression management strategies?
- Can the program improve the quality of counseling provided to parents by health care professionals?
- Can the program increase parents' willingness to discuss discipline options and child behavior with health care provider or other counselor?
- Can the program affect parents' disciplining strategies such as spanking, time-out, redirecting, and promoting empathy.
- Can the program affect parents' attitudes about disciplining strategies such as spanking?
- Do parents who view the program report that it helps them manage aggression with their own children?
A population-based intervention that results in a relatively small reduction in the rate of interpersonal violence could result in a large cost savings for the public sector. Sample size issues will likely prohibit studies from being conducted that conclusively prove that a brief, population-based intervention can reduce rates of violence and waiting for such research will likely delay progress in the area of violence prevention. Decisions to implement population-based violence prevention interventions should be driven by the results of studies that examine outcomes other than rates of violence.
* The economic dimensions of interpersonal violence. The World Health Organization. Geneva. 2004.
© 2016 Monroe Carell Jr. Children's Hospital at Vanderbilt