This study investigated the use of biofeedback and relaxation training on six patients with posttraumatic stress disorder (PTSD) referred routinely for biofeedback treatment in a VA medical center. Subjects received between 8 and 14 sessions of training overall, as well as concurrent individual and group therapy. Measures used to assess treatment outcome include pre- and posttreatment MMPI, State-Trait Anxiety Inventory, Beck Depression Inventory, and Multidimensional Health Locus of Control scores, as well as electromyographic and subjective measures of tension within each session. Additionally, an overall posttreatment clinical rating of change and 1- to 2-year follow-up data were obtained for each subject. Slight to marked improvements were demonstrated for each subject, as evidenced by improvements on the State Anxiety Inventory Scale and the Beck Depression Inventory, a decrease in overall MMPI scores, and lowered EMG and subjective tension ratings for all participants. Possible alternative explanations for improvement (situational demand characteristics, regression toward the mean, lack of independent subject evaluation) are described, along with other study limitations. This preliminary investigation suggests that the use of relaxation training and biofeedback may be a particularly useful component within a comprehensive treatment program for this disorder.