Given the enormous social cost of drug abuse and comparatively small resources for treatment and other medical expenditures, appropriate, timely and effective care could potentially have a large economic impact. Although limitations exist, evidence is growing that drug treatment is often effective, particularly when the treatment strategy is appropriately matched to the user's needs. Yet very few studies to date have examined the cost-effectiveness (CE) of drug treatment. The proposed research will develop methods of conducting CE studies of drug treatment including application of a stochastic simulation model. Most of the data used in the simulation model will be collected at two study sites where we propose to conduct CE analyses of randomized trials of various approaches to aftercare treatment following residential or inpatient treatment. The proposed research has four specific objectives: (1) Develop practical methods for cost analysis that include treatment program, program component, and episode of care costs; (2) Develop multivariate methods for CE analysis that demonstrate the impact of program and client characteristics on treatment costs and effectiveness; (3) Perform a comprehensive CE analysis comparing five approaches to aftercare of varying intensities, including a relapse prevention model; and (4) Develop and test a stochastic simulation model to predict drug use, treatment costs, and other social costs under alternative treatment policies over an addiction career (the periods of active abuse and abstinence over an individual's lifetime of drug abuse). These methods, when utilized by policymakers and program directors, will assist them to determine the most cost effective programs for specific client populations, thus facilitating more appropriate allocation of existing resources. The proposed research will also evaluate whether relapse prevention aftercare is more effective and cost-effective compared to several less costly alternatives. Finally, the proposed research will allow us, through stochastic simulation, to model addiction careers, thus capturing the long-term costs and benefits of substance abuse treatment.
