The symptom overlap between posttraumatic stress disorder (PTSD) and other pharmacotherapy-responsive disorders suggests that pharmacotherapy might be effective. Nevertheless, of the eight published placebo-controlled trials investigating the pharmacotherapy of PTSD, only four found statistically significant efficacy for the treatment being studied. This literature possesses a number of methodologic limitations, including the fact that most studies have been conducted with war veterans, who may constitute a more treatment-refractory population. Several open trials and one controlled trial with selective serotonin reuptake inhibitors have reported improvement in some or all core PTSD symptoms (reexperiencing, avoidance, numbing, and hyper-arousal). The authors hypothesized that paroxetine might be effective in PTSD, based on findings of its particular efficacy for anxiety and agitation in studies of depressed patients. The study presented here summarizes a 12-week, open-label trial of paroxetine among patients with noncombat-related, chronic PTSD. Outcome was assessed by an independent evaluator, the treating physician, and the patient, with the use of established rating scales for depression, anxiety, general symptoms, and PTSD core symptoms. A repeated-measures analysis of variance revealed highly significant improvement in all three symptom clusters, as well as in associated anxiety, depressive, and dissociative symptoms, with 11 of 17 (65%) patients rated as much or very much improved. The mean reduction in PTSD symptom scores was 48%. Exploratory analyses revealed that cumulative childhood trauma was negatively correlated with pharmacotherapy response (r = -0.52, p = 0.03). There was also significant variation in the time course of response across symptom clusters, which is suggestive of multiple mechanisms of response. Because paroxetine seems a highly promising treatment for all three symptom clusters of PTSD, a placebo-controlled clinical trial is warranted. (J Clin Psychopharmacol 1998;18:10-18)