• We report a double-blind, randomized, placebo-controlled study utilizing a within-subjects design on 20 hospitalized, psychiatric patients who participated in sodium amobarbital interviews to determine if the drug has a specific effect in eliciting clinically useful information. The patients selected had difficulty communicating with their primary therapists during the postadmission, diagnostic interviews. Two raters completed a Hamilton Depression Scale, a New Haven Schizophrenic Index, and a Brief Psychiatric Rating Scale affer each interview. Although both the amobarbital and saline interviews were moderately useful in obtaining new information, we found no significant difference in the primary therapists' assessments of clinical usefulness. In addition, the drug interview did not uncover material that would aid in the differential diagnosis between depression and schizophrenia. There was, however, a significant negative correlation between the assessment of general usefulness and the time interval between admission and interviewing. We report our only exception, a case of catatonic schizophrenia, in which the patient responded specifically to the drug.