Background: Anxiety sensitivity (AS), the belief that bodily sensations have harmful consequences, is a reliable predictor of panic attacks in both clinical and nonclinical populations. Recently, a new measure of AS has been proposed. The AS profile (ASP) was designed to be a more comprehensive measure of AS, and to be more suitable for the measurement of different AS dimensions. Preliminary evidence (college student sample) suggests that the ASP has 4 dimensions. In the present study, the dimensional structure of the ASP was further investigated, as well as its relationship with temperament and character traits.
Methods: Exploratory and confirmatory factor analysis of ASP scores in two large samples of psychiatric outpatients and nonclinical controls (combined n = 742). Correlations and partial correlations of ASP with temperament and character.
Results: Exploratory factor analysis yielded a single AS factor. However, confirmatory factor analysis showed that the 6-dimensional structure, as Taylor and Cox had originally intended it, might be a defendable solution. However, the number of items is much too high, with many subscales consisting of semantic clusters. ASP scores were found to be weakly related to the temperament dimension harm avoidance, corroborating earlier findings that were not statistically significant because of small sample sizes.
Conclusions: The ASP may be shortened from 60 to 24 items without loss of reliability or content. Future studies using challenge paradigms and studies with general hospital patients may further investigate the usefulness of a shortened version of the ASP.
Copyright 2003 S. Karger AG, Basel