Abstract
Patients who completed a detoxification/rehabilitation drug abuse program were compared with program dropouts on a new personality instrument, the Millon Clinical Multiaxial Inventory. No differences were found between groups on any of the 20 comparison variables, replicating previous results with the MMPI [Minnesota Multiphasic Personality Inventory]. Discriminant function predictors lose their potency within 3 yr and maybe sooner. Clinical judgments must form the basis of predicting dropouts until it can be demonstrated that personality tests can make reliable predictions on treatment dropouts.