Using the Toronto Alexithymia Scale, the authors evaluated prevalence and significance of alexithymia in 93 ambulatory cocaine abusers treated with psychotherapy (cognitive-behavioral relapse prevention) and pharmacotherapy (desipramine), alone and in combination. Thirty-nine percent of the sample scored in the alexithymic range; alexithymic and nonalexithymic subjects did not differ significantly on demographic variables, severity or pattern of cocaine use, substance-related problems, or treatment history; alexithymic subjects reported significantly more current distress, but did not differ from nonalexithymic subjects in rates of Axis I disorders or clinician-rated evaluations; although there was no difference in overall treatment retention or cocaine outcomes, alexithymic subjects had better outcomes when treated with clinical management over cognitive-behavioral relapse prevention, a finding consistent with the theory that suggests alexithymic patients have difficulty in treatments requiring identification of affects and cognitions.