Examination of differential anxiety sensitivities in panic disorder:: A test of anxiety sensitivity subdomains predicting fearful responding to a 35% CO2 challenge

Cognitive conceptualizations of panic disorder suggest that panic is produced and maintained by threatening beliefs associated with autonomic arousal. The present study tested the discriminant validity of the anxiety sensitivity model of panic by assessing the differential predictions of particular anxiety sensitivity domains. A factor analysis of the Body Sensations Questionnaire indicated four nested anxiety sensitivity factors assessing fears of cardiopulmonary, dissociation, numbness, and gastrointestinal sensations. The symptoms assessed by each factor possess varying levels of correspondence to the sensations typically produced during a 35% CO2 inhalation (i.e., Cardiopulmonary Fears/High Correspondence, Dissociation Fears/Moderate Correspondence, Numbness Fears/Moderate Correspondence, Gastrointestinal Fears/Low Correspondence). It was hypothesized that anxiety sensitivity to the high-correspondence sensations, compared to anxiety sensitivity to moderate- and low-correspondence sensations would predict greater fearful responding to a 35% CO2 challenge. Fifty-six participants meeting DSM-IV criteria for panic disorder completed a single vital capacity 35% CO2 challenge. Consistent with prediction, Cardiopulmonary Fears was the only index that predicted provocation-induced anxiety and symptoms. These findings suggest that specific anxiety sensitivities can provide a more powerful explanatory model for predicting emotional responding in panic disorder.