Convergent and divergent validity of the Beck Anxiety Inventory for patients with panic disorder and agoraphobia

Abstract

Psychometric properties of the Beck Anxiety Inventory (BAI) (Beck and Steer, 1990) were investigated in a sample of 82 patients suffering from panic disorder with agoraphobia. Before and after brief treatment, patients completed a battery of questionnaires and, for 2-week periods, kept a daily panic diary in which they recorded panic attacks, fear of panic, and average anxiety. The BAI demonstrated excellent internal consistency and good test-retest reliability over a 5-week interval. A partial multitrait, multimethod correlation matrix provided evidence of convergent validity with other measures of anxiety and of divergent validity vis á vis measures of depression. Factor analyses of pretest scores and residual gain scores used to address criticism that the BAI is excessively panic-centric yielded mixed results. In one analysis, the BAI was loaded with multimethod measures of panic and anxiety and, in the other, with questionnaire methods of assessing anxiety and depression. However, the BAI was clearly distinguished from measures of fear of fear, a central construct in panic disorder, and agoraphobic avoidance. Finally, the BAI proved sensitive to change with treatment, yielding effect sizes for improvement comparable to those of other anxiety measures.