2
Approaches to Measurement and Costing Methodology
Several workshop participants noted that no comprehensive framework for estimating the true economic and social burden of violence exists. Speakers offered numerous theories for the lack of such a framework, citing weaknesses in the knowledge base both in economic costing and in violence prevention, difficulty in creating a universal algorithm for diverse settings, and disagreements in types of costs to include. However, participants held mixed opinions regarding the importance of creating such a framework and the focus on relying on economic data.
The need for a robust methodology to calculate costs was best exemplified by the desire to determine the cost-effectiveness of intervention programs. Speaker Phaedra Corso from the University of Georgia outlined three major reasons for accurately measuring costs:
- To determine the true impact of violence beyond morbidity and mortality,
- To place violence in the context of and make comparisons to other public health issues, and
- To compare the cost of violence to the cost of preventing violence, and determine the cost-effectiveness of intervention programs.
Ultimately, Dr. Corso felt that the goal of accurately estimating cost was to determine the return on investment in both the public and the private sectors. Speaker Jack Shonkoff of Harvard University agreed and emphasized the point that prevention is generally cheaper than a cure, stating that “although you could always make improvements, the end point will never
be as good as it would have been had you gotten things right in the first place, and the cost of getting to a better place is going to be higher than the cost of getting it done right the first time.”
However, some speakers cautioned against overreliance on economic data because the complexity of violence does not always lend itself to quantifiable impact. Some outcomes, such as pain and suffering, cannot be reduced to calculated costs, and relying solely on numbers would potentially miss larger effects of prevention. Also, some outcomes of violence, when calculated in terms of medical costs, suggest potential financial benefits from preventing violence-related mortality. For example, speaker and Forum member Michael Phillips of the Shanghai Jiao Tong University School of Medicine pointed out that survivors of suicide attempts often have ongoing psychological illnesses that require medical attention and can pose a burden on the social welfare system through requiring subsidized health care or disability. He referenced a study in which researchers determined that $5 billion was saved in 1990 due to just over 30,000 suicides, which did not take into account the costs of psychological and physical pain and suffering, suggesting that suicide prevention should not be assessed solely in terms of monetary cost (Yang and Lester, 2007).
Still, most speakers felt that determining the economic costs of violence held some value and developing a framework for doing this would be useful in showing the importance of primary prevention. The framework would allow for comparisons across types of violence, types of impact, and types of context. These speakers also noted that such a framework would be most effective through accounting for social costs and that research into developing a methodology for calculating costs of pain, suffering, and other nontangible effects was important.
A number of factors that would be useful in thinking about a broader process for calculating costs were discussed. Participants discussed ways of categorizing costs, criteria for inclusion in cost calculations, noneconomic effects, and placement of costs in a larger context.
In terms of categorizing costs, participants began with a matrix that broke out costs by traditional economic categories and by ecological elements (see Table 2-1 and Box 2-1). For example, traditional cost categories include health services, social services, workforce, community development, and criminal justice. Ecological elements include individual, family, community, and societal. Thus, speakers were able to consider costs at various intersections of these two axes, as well as the relative ease or difficulty of obtaining such figures.
Dr. Corso further simplified the categories of costs into three: medical, nonmedical, and productivity. Medical costs include medical claims and other acute costs, as well as the costs of long-term sequelae. At various levels, these could include out-of-pocket expenses (individual), effects
TABLE 2-1 Sample of Potential Costs of Violence at Various Ecologic Levels
Individual | Family | Community | Societal | |
Economic development |
|
|
|
|
Criminal justice services |
|
|
|
|
Health services/health effects |
|
|
|
Individual | Family | Community | Societal | |
Social services |
|
|||
Perpetration of future violence |
|
|||
Mortality |
|
|||
Educational system |
|
Workforce |
|
|
|
|
Community development |
|
|
||
Household resources |
|
|||
BOX 2-1
Example of Ecological Approach to
Assessing Costs: Child Maltreatment
A child who is abused at home can experience behavioral difficulties in school. If the school fails to identify and address such issues, the response may be limited solely to punishment such as removal from class, detention, suspension, etc. The child then misses school, is labeled a “troublemaker,” and may not reach full intellectual potential due to these missed opportunities. Later repercussions can include decreased employment or financial opportunities, increased stress due to lower income, potential for perpetration of violence in future family settings, increased risk for substance abuse, and chronic health outcomes due to stress. Other children in school might also experience poorer educational quality due to the disruptions, which can have similar effects on their earning potentials later in life.
These costs have the potential to extend further, to other individuals who come in contact with the abused child or his classmates later in life. For example, the job or income stress could result in workplace violence or other instability, affecting other workers.
on indirect victims (family), cost of local care (community), and costs of health insurance (societal). Productivity costs were defined as those related to absenteeism, or absence from the workplace, which has costs to both individuals and families (loss of income) as well as to community and society (loss of profit to businesses). Dr. Corso also referenced “presenteeism,” a less easily defined or calculated cost, which includes being present in the workplace but not working to full capacity. Methods of calculating the impact of presenteeism are less straightforward than absenteeism; however, both are often calculated from the standpoint of lost wages. One possible alternate method is to consider the willingness of society to pay a certain price for prevention. Finally, she also discussed nonmedical costs, which were all those costs that did not readily fall into the other two categories. These costs are often not calculated or included in traditional calculations.
Two methods of cost reporting were also discussed by Dr. Corso—using prevalence-based and incidence-based data. Prevalence-based data have the limitation of providing an accurate picture of medical costs only, but not other types of costs, and providing a picture of acute costs only. Incidence-based data are not as robust, because they require reporting along the entire life span and much violence is unreported. However, such an approach would better parse nonmedical costs.
Speakers also discussed the difference between direct and indirect costs, with the former being much easier to define. Dr. Corso suggested that most nonmedical costs fall under indirect costs. Speaker Hugh Waters from the
RAND Corporation also mentioned that various types of indirect costs exist and specified the difference between a cost incurred in the provision of care that is not part of the care protocol itself and the cost of an indirect effect. Speakers Gary Milante from the World Bank and Theresa Betancourt from Harvard University and the François-Xavier Bagnoud Center for Health and Human Rights also named another important indirect cost— the cost of inaction. Both cautioned against waiting too long for the full picture of costs at the risk of waiting too long to act and prevent violence from escalating. Dr. Milante also pointed out that determining the cost-effectiveness of a program is important, but reducing the cost of inaction could have greater impact than ensuring that a preventive intervention is the most effective.
Along with the cost of inaction, several speakers suggested that opportunity costs be considered as well. Dr. Waters noted that certain costs will be incurred by certain sectors regardless of whether violence occurs. For example, medical and social services providers and infrastructure costs will be paid. However, when resources are spent to address the effects of violence, fewer resources are available to be allocated toward other issues. Speaker Aslihan Kes from the International Center for Research on Women pointed out that women who are victims of violence lose opportunities to complete household or income-generating activities. Dr. Corso suggested that productivity costs are often calculated in terms of opportunity costs.
The identification of noneconomic effects and the ability to place dollar values on such effects was also a major discussion at the workshop. Some participants strongly felt that such costs lie at the heart of the massive burden of violence but are currently difficult to enumerate. Finally, the need to place costs within a larger context was an important element of discussions; as Dr. Shonkoff highlighted, the issue is not simply about saving money, but also about ensuring a higher quality of life.
REFERENCES
Yang, B., and D. Lester. 2007. Recalculating the economic cost of suicide. Death Studies 31(4):351-361.