Hester van Eeren

Value of information in crime prevention: An illustration | 3 35 | Introduction In order to guide policy decisions, it would be helpful to know the cost-effectiveness of interventions aimed at reducing juvenile delinquency. So far, cost-effectiveness analyses have informed an increasing number of reimbursement decisions in mental health-care (Evers, Salvador-Carulla, Halsteinli, McDaid, & MHEEN Group, 2007; Knapp et al., 2008). Accordingly, the number of cost-effectiveness analyses in the field of crime prevention is increasing (Barrett & Byford, 2012; Cary, Butler, Baruch, Hickey, & Byford, 2013; French et al., 2008; Knapp et al., 2008; McCollister et al., 2003a; McCollister et al., 2003b; McCollister, French, Prendergast, Hall, & Sacks, 2004; Romeo, Byford, & Knapp, 2005; Soeteman & Busschbach, 2008). The inputs in a cost-effectiveness analysis can be uncertain, as available information about the costs and effects of interventions is rarely perfect. As a result, the decision whether or not to reimburse an intervention is marked by uncertainty. When a decision to reimburse an intervention turns out to be incorrect, it could lead to suboptimal interventions being approved. These interventions create costs in terms of foregone benefits and resources (Briggs, Claxton, & Sculpher, 2006; Claxton, 2008; Claxton, Neumann, Araki, & Weinstein, 2001; Claxton, Sculpher, & Drummond, 2002; Oostenbrink, Al, Oppe, & Rutten-van Mölken, 2008). Further research may eliminate this uncertainty and optimize the reimbursement decision. This study aims to estimate the added value of future cost-effectiveness research. This type of analysis is referred to as a ‘value of information’ analysis and was introduced as part of statistical decision theory (Pratt, Raiffa, & Schlaifer, 1995; Raiffa, 1968). It has already been applied in other research areas, such as engineering and environmental risk analysis (Yokota & Thompson, 2004), before being introduced into health technology assessment (Briggs et al., 2006; Claxton, 1999, 2008; Claxton et al., 2001; Claxton et al., 2002; Oostenbrink et al., 2008), where the application of this analysis is now widely adopted, as well as in the field of mental health care (Mohseninejad, van Baal, van den Berg, Buskens, & Feenstra, 2013; Soeteman, Busschbach, Verheul, Hoomans, & Kim, 2011). A value of information analysis reveals the value of conducting additional research and identifies the type of research that would be most useful. Its results can inform about further research on specific parameters, and more precisely inform the decision about which intervention should be reimbursed (Myers et al., 2011). Furthermore, a value of information analysis can be used to prioritize future research, for example by highlighting the merits of certain types of research which might add to the reduction of the parameter uncertainty in cost-effectiveness analysis (Carlson et al., 2013; Oostenbrink et al., 2008; Sculpher & Claxton, 2005). The potential value of further research could then be weighed against the costs of conducting this research in order to determine whether it is worthwhile (i.e. Briggs et al., 2006; Claxton, 2008). Because a value of information analysis has not yet been applied in the field of crime prevention, we will present an example of this analysis based on an existing cost-effectiveness model in crime prevention and treatment (Schawo et al., 2012). We used two interventions aimed at reducing juvenile delinquency in the Netherlands, in adolescents aged 12-18 years. These interventions can be applied to prevent juvenile