OBJECTIVE--The purpose of this article is to review the empirical evidence for the efficacy of a range of treatments for posttraumatic stress disorder (PTSD). Reviewed studies focused on rape victims, combat veterans, the tragically bereaved, torture victims, accident victims, of physical assault, and child abuse victims. DATA SOURCES--Peer-reviewed journals (Psych-Info, MEDLINE), book chapters (PILOTS database), active investigators, abstracts from the 1990 and 1991 International Society for Traumatic Stress Studies. STUDY SELECTION--We identified 255 English-language reports of treatment for PTSD. We restricted our focus to randomized, clinical trials that included a systematic assessment of PTSD using DSM-III or DSM-III-R criteria (N = 11). DATA EXTRACTION--Studies were assessed according to methodological strength: random assignment to the treatment of interest, and either an alternative treatment or control group; sample selection; and inclusion of statistical tests of significance. DATA SYNTHESIS--Drug studies show a modest but clinically meaningful effect on PTSD. Stronger effects were found for behavioral techniques involving direct therapeutic exposure, particularly in terms of reducing PTSD intrusive symptoms. However, severe complications have also been reported from the use of these techniques in patients suffering from other psychiatric disorders. Studies of cognitive therapy, psychodynamic therapy, and hypnosis suggest that these approaches may also hold promise. However, further research is needed before any of these approaches can be pronounced effective as lasting treatment of PTSD. CONCLUSIONS--Further studies should specifically address combined treatment approaches, optimal treatment length and timing, effects of comorbidity, and unstudied traumatized populations.