Further evidence of construct validity for the Agoraphobic Cognitions Questionnaire and the Body Sensations Questionnaire.

Abstract

Chambless, Caputo, Bright, and Gallagher (2) developed and provided initial evidence of reliability and construct validity for the Agoraphobic Cognitions Questionnaire and the Body Sensations Questionnaire. The authors note that, although these instruments were developed for assessment of agoraphobia, the questionnaires could be effective tools when used to assess other anxiety disorders. We further hypothesized that the two scales would be useful in assessment of anxiety disorders in general. Given recent data that suggest the ubiquity of panic across the anxiety disorders (1) and that demonstrate the relationship between catastrophic cognitions and body sensations for agoraphobics, claustrophobics, and panic-disordered patients (3), we predicted a significant correlation between scores on the questionnaires when administered to a sample of patients who presented a variety of anxiety disorders. Subjects, 26 female and 7 male outpatients (M age = 32.0, SD = 8.5) who sought treatment for anxiety from 1986-1987 at one of four mental health clinics near Portland, Oregon, were administered the Agoraphobic Cognitions Questionnaire and the Body Sensations Questionnaire. Scores were obtained Erom 30 subjects. The frequency of these 30 subjects by DSM-I11 diagnostic category, as determined by licensed psychologists and/or supervised graduate students, was as follows: 5 agoraphobia with panic attacks, 3 social phobia, 2 simple phobia, 9 panic disorder, 7 generalized anxiety disorder, 3 posttraumatic stress disorder, and 1 atypical anxiety disorder. Means and standard deviations for the Agoraphobic Cognitions Questionnaire (M = 2.08, SD = .67) and the Body Sensations Questionnaire (M = 2.52, SD = ,851 provided by our sample of 30 subjects were similar to those reported by Chambless, et al. (1984) for two samples of agoraphobics (i.e., MA,, = 2.32, SD = .07, n , = 78; MBJQ = 3.05, SD = 2 6 , n, = 53). As predicted, a significant product-rnoment correlation of .69 @< ,001) between the questionnaire scores in this sample was obtained. This value is consistent with those reported by Chambless, et al. (1984) for two samples of agoraphobics (r = .34, n, = 95; r = .67, n, = 50) but also provides evidence of construct validity for the scales when used to assess anxiety disorders in general. Our findings, however, should be reviewed in terms of limitations associated with a single, relatively small sample. Research with much larger samples is needed to evaluate the suitability of these instruments for discriminating among various anxiety-disordered clients, between agoraphobic and nonagoraphobic clients, and between anxiety-disordered and nonanxiety-disordered clients.