This pattern of results was also replicated in the self-selected sample when we used "longest abstinence" from alcohol and drugs as the outcome criteria.
The self-selected sample was generally similar to the randomized sample regarding the prevalence of psychiatric symptoms by psychiatric severity group, as well as on the effects of DH by psychiatric severity group.
In the self-selected sample (not shown), OP visit patterns and treatment costs were similar to those of the randomized OP group.
Importantly, analysis by psychiatric status among the self-selected sample again found effects only among the mid-level of psychiatric severity, where DH subjects had almost three times the likelihood of total abstention as those in OP.
8 times the number of services) and the self-selected sample (6.
However, for subjects with mid-level psychiatric severity in both the randomized and self-selected samples, the DH program produced higher rates of abstention and was more cost-effective.
This study is the first to examine treatment intensity within a managed care population, to examine costs relative to effectiveness, to use sample sizes adequate for the examination of effects among patient subgroups, and to examine randomized and self-selected samples.
For both the randomized and self-selected samples, we employed an intent-to-treat design such that all subjects who were followed were included in the analyses, regardless of whether they returned after the intake interview to start treatment or the number of sessions they attended.
We found no significant differences between DH and OP in out-of-plan utilization rates reported for the randomized and self-selected samples.