In recent years, a number of studies have investigated the prediction of posttraumatic stress disorder (PTSD) through the presence of acute stress disorder (ASD). The predictive power of ASD on PTSD was examined in a population of 148 female rape victims who visited a center for rape victims shortly after the rape or attempted rape. The PTSD diagnosis based solely on the three core symptom clusters was best identified by a subclinical ASD diagnosis based on all ASD criteria except dissociation. However, a full PTSD diagnosis including the A 2 and F criteria was best identified by classifying victims according to a full ASD diagnosis. Regardless of whether cases were classified according to full PTSD status or according to meeting the criteria for the three PTSD core symptom clusters, the classification was correct only in approximately two thirds of the cases. A regression analysis based on ASD severity and sexual problems following the rape accounted for only 28% of the PTSD severity variance. In conclusion, the ASD diagnosis is not an optimal method for identifying those most at risk for PTSD. It remains to be seen whether a better way can be found. (Contains 3 tables and 1 note.)