باورهای ناکارآمد در اختلال پانیک: پرسشنامه اعتقاد هراس

The Panic Belief Inventory (PBI) was developed to assess beliefs that increase the likelihood of catastrophic reactions to physical and emotional experiences in panic disorder. In the first stage of scale development, 197 panic disorder patients completed the PBI and standard self-report inventories of psychiatric symptomatology. An exploratory factor analysis yielded a 4-factor solution from which a 35-item instrument with 4 scales was constructed. The shortened measure and its scales had good internal consistency and convergent validity and moderate discriminant validity. Subsequently, 22 panic disorder patients who received cognitive therapy completed the PBI and other self-report inventories of dysfunctional cognitions at intake, 4 weeks, 8 weeks, termination, and several follow-up intervals. Results indicated that the PBI decreased significantly across treatment, with the largest decline occurring between intake and 4 weeks into treatment. The PBI correlated more strongly with dysfunctional cognitions associated with anxiety than dysfunctional cognitions associated with depression. These results provide preliminary evidence that the PBI has adequate psychometric characteristics, is useful to assess change in dysfunctional beliefs during treatment, and has the potential to advance cognitive theories of panic.

According to cognitive theories of panic disorder (e.g., Beck & Emery, 1985; Clark (1986) and Clark (1988)), panic attacks result from catastrophic ideation and a fear of physical and psychological disaster. Individuals with panic disorder estimate that physiological and emotional experiences are more dangerous than they actually are and that they signal impending doom. Cognitive models suggest that individuals with panic disorder experience negative automatic thoughts, such as “I am going to have a heart attack,” in the context of an anxious state or when encountered by ambiguous bodily sensations. Moreover, they indicate that these individuals are characterized by the tendency to filter information in a biased manner, such that they are hypervigilant for and quickly detect changes in their physiological activity and emotional stability. Thus, distorted cognition is central to understanding the pathology of panic disorder in these models. A number of self-report inventories have been developed to assess the cognitive content associated with panic, most of which can be grouped into two broad categories. The first group of measures assesses the degree to which individuals with panic disorder experience negative automatic thoughts during panic attacks, including the Agoraphobic Cognitions Questionnaire ( Chambless, Caputo, Bright, & Gallagher, 1984), the Agoraphobic Self-Statements Questionnaire ( van Hout, Emmelkamp, & Scholing, 1994), the Panic Attack Cognitions Questionnaire ( Clum, Broyles, Borden, & Watkins, 1990), and the panic consequences scale of the Panic Appraisal Inventory ( Telch, Brouillard, Telch, Agras, & Taylor, 1989). In these measures, individuals with panic disorder are presented with a series of negative automatic thoughts (e.g., “I am going to throw up”, “This will never end”), and they rate the frequency and/or intensity with which each thought is experienced during acute anxiety. Research examining the psychometric properties of these measures has found that they discriminate between individuals with panic disorder and individuals with other anxiety disorders and nonanxious individuals and are sensitive to gains made in treatment. Thus, the development of these measures has advanced theory by providing empirical evidence that individuals with panic disorder endorse these cognitions during times of acute anxiety. A second group of measures assesses the degree to which individuals with panic disorder experience fear and anxiety during uncomfortable physiological and psychological sensations. For example, the Body Sensations Questionnaire ( Chambless et al., 1984) requires individuals to endorse the degree to which they are frightened or worried by sensations that occur in the context of being in a nervous or feared situation (e.g., “Heart palpitations”), and the Agoraphobic Cognitions Scale ( Hoffart, Friis, & Martinsen, 1992) requires individuals to rate the extent to which they fear various situations (e.g., “Fear of illness”, “Fear of making a scene”). Noting that these measures assess a cognitive aspect of fear but not necessarily catastrophic cognitions as specified in Beck and Emery's (1985) model, Khawaja and Oei (1992) developed the Catastrophic Cognition Questionnaire. Individuals who complete this inventory are instructed to rate the extent to which they believe conditions, such as being irritable or feeling dizzy, are personally dangerous. Although the Catastrophic Cognitions Questionnaire was developed from a well-defined theoretical framework and had the potential to be an especially relevant measure of cognition associated with panic disorder, only two of its five factors differentiated panic patients from non-patients, and no factors differentiated panic patients from other anxiety patients ( Khawaja, Oei, & Baglioni, 1994). A construct related to panic-relevant cognitions is anxiety sensitivity, defined as the “tendency to respond fearfully to anxiety symptoms” which “is based on beliefs that these symptoms have undesirable consequences” (McNally, 1989, p. 193). Although McNally (1989) raised the possibility that the tendency to make catastrophic misinterpretations is indeed anxiety sensitivity, he later argued that the anxiety sensitivity construct is broader than catastrophic misinterpretations because panic symptoms can be aversive even when individuals do not attribute them to a disaster (McNally, 2002). The anxiety sensitivity construct characterizes individual differences in the vulnerability to develop panic attacks, as evidenced by the vast literature that has accumulated over the past two decades (see Taylor, 1999, for comprehensive reviews). The most widely used measure to assess anxiety sensitivity is the Anxiety Sensitivity Index (ASI; Peterson & Reiss, 1992), in which individuals rate the extent to which 16 statements are true (e.g., “It scares me when my heart beats rapidly”). However, its items are limited to emotional responses to bodily and psychological anxiety sensations, and beliefs about these symptoms are inferred rather than measured directly. Thus, although cognition is a central part of the anxiety sensitivity construct ( Reiss, 1991), it is unclear whether the items included in the ASI are measuring these beliefs or affective reactions to them (see McNally, 1999, for a discussion of this issue). In general, self-report inventories of cognitive content have yielded data to support the theory that individuals with panic disorder make catastrophic interpretations of innocuous physiological and emotional experiences and that this cognitive style plays a large role in maintaining the disorder. However, nearly all studies examining catastrophic reactions assess cognition in the context of feeling anxious or otherwise actually experiencing the sensation at that moment (e.g., Chambless et al., 1984). Results from these studies shed little light on the cognitive processes that maintain pathology over time and that occur in between acute attacks. Some theoretical (e.g., Clark, 1988) and empirical (Richards, Austin, & Alvarenga, 2001) works raise the possibility that this cognitive style is a trait and that individuals with panic disorder are vulnerable to this cognitive style even before their symptoms reach a diagnostic level, but no empirical work beyond that in the anxiety sensitivity literature has been designed to examine directly biased cognition outside of episodes of acute state anxiety. In fact, Cox (1995) noted that the major weakness of cognitive models is in “identifying a measurable trait or individual difference that predisposes some individuals to make catastrophic misinterpretations (p. 364).” Although Beck and Emery's (1985) discussion of panic disorder focuses on catastrophic reactions to bodily and emotional sensations when they occur, their more general cognitive model of anxiety speaks to this issue. Specifically, they suggest that anxious individuals are characterized by a relatively enduring set of schemas that prepares them to assign cognitive resources toward threat in their environment. In addition, specific rules or beliefs are embodied in these schemas and provide the context for the manner in which ambiguous information is interpreted. These beliefs center around the concept of vulnerability and the estimation of the inability to cope with danger. According to Beck and Emery's model, catastrophic reactions to physical and emotional experiences might be the most visible indicators of biased cognition in panic disorder, but they reflect underlying beliefs that form the core of the pathology. Moreover, there is a convergence between Beck and Emery's cognitive model and Reiss’ (1991) anxiety sensitivity model, as both assume that beliefs about danger make individuals vulnerable to experience anxiety symptoms. In order to provide more complete empirical validation of cognitive models of panic, it is important to document that individuals with panic disorder are characterized by maladaptive beliefs in the absence of acute state anxiety in addition to anxiety sensitivity and catastrophic reactions to bodily and emotional sensations. To date, there is a paucity of studies that have examined these panic-relevant beliefs. A self-report inventory, the Panic Belief Inventory (PBI), was developed to assess the degree to which panic patients endorse maladaptive beliefs that, theoretically, should be related to this pathology. Although Brown, Beck, Newman, Beck, and Tran (1997) demonstrated that this inventory (which was initially termed the Panic Belief Questionnaire) varied with changes in psychiatric symptoms, the purpose of their study was to evaluate the efficacy of cognitive therapy for panic disorder, and they did not systematically evaluate other psychometric properties. Thus, the purpose of the present study was to examine the reliability, validity, factor structure, and clinical utility of the PBI in two samples—197 individuals with panic disorder who completed this inventory in the context of a standard battery of inventories given at intake at University of Pennsylvania's Center for Cognitive Therapy (CCT), and a portion of Brown et al.'s (1997) sample who completed the revised PBI at several intervals during treatment and follow-up.