Multiple drug interviews as a diagnostic technique: a clinical case study.

Diagnostic information obtained from a single drug interview is subject to conscious and unconscious distortion by the patient and to interviewer effects. Rationales for the various provocative techniques are reviewed, with particular reference to the amphetamine psychosis and to state-dependent learning. A format for a controlled series of multiple provocative interviews which "triangulates" on the same clinical problem from several theoretical perspectives is proposed, and a clinical case is presented in which this format was employed. Used in this way, drug interviews may provide useful diagnostic information.