Abstract
This report examines the correlates of early treatment termination by patients being treated for alcohol abuse or dependence. All of the study subjects were participants in a controlled experimental investigation of the differences in clinical outcome and treatment cost of extended inpatient and partial hospital treatment. A large number of clinical variables demonstrated no difference between groups, except for the greater likelihood of a DSM III Axis I diagnosis of alcohol abuse or dependence in the treatment completers. In contrast, three of four system variables differentiated completers from noncompleters. Patients treated by hospital-employed psychiatrists, in contrast to private psychiatrists, were more likely to complete the program; 4 sequential months in the middle phase of the study were associated with noncompletion of treatment; and patients who were detoxified on the preferentially designated unit were more likely to receive some specialized alcohol treatment. The authors interpret these data to indicate that treatment adherence is influenced by factors affecting the cohesiveness of the treatment delivery system. The failure of patient characteristic variables to predict treatment adherence may well be a result of their comparative insignificance in relation to system variables.