To the Editor: The introduction of acute stress disorder into DSM-IV has already spawned much useful research about a disorder that can be both hidden and overlooked, providing promise for earlier identification and more effective intervention. We therefore agree with Dr. Marshall et al. that there is a need for a diagnostic entity that defines acute reactions to traumatic stressors as something more than an adjustment disorder. That indeed was the motivation for our recommendation to the DSM-IV task force that acute stress disorder be included in the nosology. However, we disagree with the authors’ interpretation of the literature and recommendations for the following reasons.

First, it appears that the pattern of symptoms is different in the acute versus chronic phase of trauma response and that dissociation features more prominently early on (1–5; Harvey and Bryant, 1998). The authors confounded evidence used for the construct (predictive) validity of dissociative symptoms in acute stress disorder with their importance as symptoms per se. Dissociative symptoms are not included in acute stress disorder simply as risk factors for the development of PTSD; also, it would not make sense to include other risk factors, such as neuroticism or history of prior trauma or psychiatric illness.

Second, the authors suggested that dissociative symptoms be "an associated, but not required, feature of acute PTSD" (p. 1683). Yet their argument that dissociative symptoms are a less than perfect predictor of PTSD and therefore should be dropped from acute stress disorder is tautological. Dissociative symptoms are only a minor component of the current DSM-IV PTSD criteria, as are amnesia and numbness. It is axiomatic that predictive power is greatest when one is assessing the same symptom at baseline and follow-up. Thus, it is remarkable that dissociation in the acute phase predicts later PTSD as well as it does (Brewin et al., 1999). If anything, the problem may be that PTSD needs redefinition. In fact, some studies have shown that dissociative symptoms (1, 4, 6) and acute stress disorder (Brewin et al., 1999) are better predictors of long-term PTSD than are acute intrusion and hyperarousal symptoms themselves.

Third, Dr. Marshall et al. are troubled by the overlap between normal aspects of human experience and dissociative psychopathology but are unconcerned that a low symptom threshold for acute stress disorder could "pathologize" normal reactions.

Fourth, the authors concluded that "dissociation is not a core feature of acute PTSD" (p. 1681). However, the review on which this conclusion is based is incomplete. For instance, one of the studies the authors highlighted (7) is a retrospective study that purported to investigate the consequences of acute stress disorder and yet had no systematic evaluation of acute stress disorder. The authors failed to cite a recent prospective study that found clear and strong predictive power for dissociation: "The criteria of three or more dissociative symptoms and one or more avoidance symptoms specified in DSM-IV produce a realistic balance of sensitivity, specificity, and positive and negative predictive power" (Brewin et al., 1999, p. 364).

There is an arbitrariness in any diagnostic scheme, and acute stress disorder is no exception. The differences between the criteria for acute stress disorder and PTSD should be addressed through further research, but the current evidence supports the utility of acute stress disorder and suggests that the criteria for PTSD, as well as acute stress disorder, should include dissociation.