Understanding the Societal Value of the Health Improvements Associated with Providing Medication Assisted Treatment in a Large Urban Jail

点击翻译按钮获取中文摘要

Project Summary The economic burden of crime in the United States far exceeds that of any other westernized country and a large proportion of these costs are associated with recidivism. Clinical evidence suggests that medication assisted treatment (MAT) for jail-based populations can be effective in mitigating substance abuse. However, these programs have not been adopted on a large scale, in part due to a lack of information about the economic implications of these programs. The goal of this proposed study is to better understand the economic impact of MAT in incarcerated populations and to improve and further develop empirical techniques used to evaluate these types of programs. To do this, we will utilize data from a novel methadone maintenance treatment (MMT) program implemented in a large urban jail in the Southwest. Our analysis will build on a prior program evaluation that found the continuation of MMT among opiate dependent inmates was associated with a significantly lower probability of re-incarceration. This economic study has three aims. First, we will estimate the costs of implementing MAT in jails using the Drug Abuse Treatment Cost Analysis Program (DATCAP), a widely adopted method that has been used to calculate the economic costs of over 100 substance abuse treatment programs. Second, we will use these cost estimates in combination with detailed data from a number of New Mexico jail and court datasets to estimate the cost-effectiveness of this program. Finally, our third aim is to estimate the impact of this program from a societal perspective using a multivariate cost-of-crime approach. Notably, while this type of approach provides an efficient and inexpensive way to economically evaluate programs using administrative data, it has been utilized rarely. This is likely due to a number of methodological and statistical challenges of using monetized crime outcome variables (e.g. lack of variation, mass points at zero and skewness). As part of this aim we will address these statistical issues using a number of well-established statistical methods (e.g., logging, trimming and censoring techniques) and more advanced empirical techniques (two-stage models, Box-Cox transformations and categorical models). These improved statistical models will lead to greater and more effective use of cost-of-crime models, which in turn will increase the quality of information provided to policy makers about the benefits of substance abuse intervention programs.