Drs. Gorman, Shear, McIntyre, and Zarin Reply

To the Editor: We are pleased to be able to respond to the four thoughtful letters regarding the “Practice Guideline for the Treatment of Patients With Panic Disorder.” These letters reflect the reality that American Psychiatric Association practice guidelines are very important for the field and warrant very careful attention and feedback from our profession. We appreciate all the responses—agreement and disagreement—that we have received. Such continued feedback improves the guidelines and the developmental process. In the course of preparing the Guideline for panic disorder, we solicited the input of more than 650 individuals and organizations and were pleased to receive many comments, all of which were carefully considered. In addition, we had several lengthy conferences with clinicians representing the psychoanalytic point of view and made several modifications in the Guideline to incorporate their suggestions. First, it is important to note that this is a practice guideline for panic disorder, not anxiety disorders in general. Given that, we acknowledge that no treatment guideline ever incorporates the ideas of all clinicians or investigators, but we struggled to strike a reasonable balance. We were indeed guided by the principle that rigorous clinical studies are the standard by which the effectiveness of all medical interventions should be judged. In addition, we attempted to incorporate suggestions based on clinical wisdom, as long as they were clearly identified as lacking backing from rigorous studies.

Dr. Fintzy asks why we did not indulge in etiologic speculation. Although it took much forbearance, since many of the Guideline committee members have spent years laboring to elucidate the “cause(s)” of panic disorder, treatment guidelines are not the place to make causal judgments. We are grateful to Dr. Fintzy for his comment about benzodiazepines; although many concerns have been raised about the use of benzodiazepines for the treatment of anxiety disorders, they are clearly effective for many patients.

We are sorry that the two reviews mentioned by Dr. van Balkom and Dr. van Dyck were not available to us in time to be included in our considerations. Although meta-analyses are always subject to technical objections, these reviews do suggest that combination treatment may offer patients with panic disorder an advantage over monotherapy, although not all studies substantiate this finding. As we suggest in the Guideline, this is an important area for further investigation. We thank Drs. van Balkom and van Dyck for their kind comments about the Guideline in general.

We admit that we are disappointed by the reactions to the Guideline of our colleagues from the American Psychoanalytic Association. As we note above, this Guideline reflects a careful review of existing scientific treatment studies and a long and often arduous process in which we tried to identify areas of clinical consensus; especially in those areas where there are major gaps in the research literature. We did not find a body of literature indicating that a scientific evaluation of the treatment of panic disorder with psychoanalytic therapies has been undertaken. We believe it is a universally acknowledged principle of medicine that interventions for illness are best supported by rigorous scientific study, as has clearly been the case for medication and cognitive behavioral therapies for panic disorder. Nevertheless, the Guideline does not in any way suggest that psychodynamic treatments are ineffective, and at many points it suggests that clinicians have found it to be efficacious. One of the co-chairs of the Guideline committee is a faculty member of the Columbia University Center for Psychoanalytic Training and Research, and both co-chairs are supporters of psychoanalytic research and treatment. We hope that by the time new practice guidelines are formulated, studies evaluating the effectiveness of psychoanalytic therapies for panic disorder will have been published.

There are six points that can be addressed in the letter of Drs. Busch, Milrod, and Gabbard. 1) We agree that more long-term studies of panic disorder are urgently needed and make that suggestion on page 26. There is, however, already a substantial literature on this problem (e.g., 1–3) in addition to the paper by Drs. Milrod and Busch that they cite. Exactly how the lack of long-term studies bears on the issue of acute treatment effects, however, is not clear to us. 2) It should be noted that the ratings of the recommendations is a statement of clinical confidence and, as these authors noted, is not always correlated with the ratings of evidence in the literature. 3) We strongly disagree that the “tone” or any other aspect of the Guideline in any way implies that psychoanalytic treatments “lack...therapeutic efficacy.” We agree, however, that there is no way to compensate for the lack of rigorous scientific treatment studies. 4) It seems a quibble to object to our comment that venlafaxine may be effective. It is a member of a class of medications that have been shown to be effective for panic disorder, and therefore we feel it is reasonable to suggest that it “may be effective” as well. This does not seem to be the same as saying that psychoanalytic psychotherapies are effective for panic disorder on the basis of case reports, given that they do not belong to a class of interventions that has documented efficacy. 5) We certainly hope that Drs. Busch, Milrod, and Gabbard do not believe that a treatment guideline for panic disorder is the place to rehearse the debate about the validity of DSM-IV categories. Indeed, the Guideline is written under the assumption that panic disorder is an illness that the psychiatrist has already diagnosed using the criteria of DSM-IV. 6) It is well-known that necessary exclusion criteria in clinical trials rarify the sample of patients enrolled. This clearly imposes the risk of lack of representation of community samples. Again, however, it is unclear what bearing this has on the recommendations given in the Guideline. Treatment guidelines must obviously offer recommendations based on the available scientific literature and acknowledge, as the panic Guideline does at many points, that individual clinicians must tailor treatment to individual patients. Is there evidence that psychoanalytic therapies are as effective as, or superior to, medication or cognitive behavioral therapies in the “community”? We are hopeful that the American Psychiatric Association’s Practice Research Network will contribute to the growth of evidence from the community to help with this and many other areas where rigorous studies are not available or are inadequate.

The “Practice Guideline for the Treatment of Patients With Panic Disorder” recommends that there is substantial scientific evidence for the effectiveness of medication and of cognitive behavioral psychotherapy. We believe that this conclusion is obvious and, given the current state of scientific investigation, unassailable. We think it is very important to note that the Guideline clearly states that the published evidence indicates that a form of psychotherapy, namely, cognitive behavioral therapy, is effective, and as effective as medication, for the treatment of patients with panic disorder. In addition, the Guideline addresses the use of other interventions, including psychodynamic psychotherapy, for patients with panic disorder.