An additional change would be to adjust the PAS to bring it more in line with the recent changes to the definition of a panic attack in the revised edition of the DSM-III. Specifically, these changes should reflect differences in the spontaneous vs. nonspontaneous onset, association with specific stimuli, and the presence of avoidance behaviors.

Another possible improvement to the present study is to modify the IA scale. While many items of this scale are clearly related to the accurate identification of emotion and visceral sensations, several of the items are strictly related to eating. This is understandable, as the IA scale has been taken from the Eating Disorders Inventory. An improvement to the study would be to eliminate these food related items, and replace them with items clearly reflecting difficulty in accurately interpreting internal sensations.

It would also be of interest to replicate the present study with a clinical sample, and compare it to an age and sex-matched non-clinical group. While an analogue sample, such as the present study, does point to certain directions for research, the clinical utility of these findings can only be established if similar results are found with a clinical sample.

This study was the first to directly examine the relationship between panic attacks and irrational beliefs. Given the limitations previously discussed, the use of a non-clinical population which experiences panic attacks might provide researchers with a readily available population for testing theories of the etiology and treatment of panic attacks and panic disorder.