While many studies have demonstrated the short-term effectiveness of substance abuse treatment, a major concern of policy-makers and clinicians has to do with the long-term impacts of treatment. This continuation application examines determinants of 7 year trajectories of treatment readmission, substance use, and medical services utilization and cost. Of particular interest is the stability of 6 and 12 month treatment outcomes. The original study randomized 1204 individuals entering an HMO s substance abuse treatment program to two treatment intensities, day hospital and traditional outpatient. We followed those randomized, as well as those who were unwilling or unable to be randomized, thus having a comprehensive sample of the treatment program. We use our findings from the original study to hypothesize clinically and policy relevant predictors of long- term outcomes. We use growth curve models and other analytic approaches to examine the factors influencing 5 and 7 year trajectories of treatment readmission, outcome, utilization and cost. We study the effects of an index treatment and cumulation of treatment and other hypothesized influences on long-term outcomes. We address such research questions as: do the effects of one treatment episode decay over time if not reinforced by changes in social networks and ongoing participation in self-help groups? How do these factors impact long-term utilization and cost? We study explanatory factors or: 1) individual characteristics, such as demographic and severity characteristics, 2) treatment characteristics, such as treatment intensity, length of stay, type of services received, and readmissions; and 3) extra-treatment factors, such as change in social networks and participation in NA/AA, measured over the following 7 years. We have the opportunity to study long-term treatment effects in the large, heterogeneous treatment population of an HMO. The HMO has on-going computerized data to document the dates, type of substance abuse treatment, and length of stay of treatment episodes, as well as all medical services received and costs. We have also collected self-report data on substance abuse treatment utilization and medical services utilization not covered by the health plan. Our multiple data points are important in examining drinking patterns and service interventions, and their 7 year coverage allows for sufficient length of time to examine long-term effects of index treatment episodes and cumulative effects of treatment, other medical services and changes.
Showing the most recent 10 out of 18 publications
