بررسی تاکسومتریک از موقعیت هراسی در یک نمونه بالینی و جامعه

The nosological status of agoraphobia is controversial. Agoraphobia may be a distinct diagnostic entity or a marker of avoidance severity. The current study examines the latent structure of agoraphobia through the use of taxometric analysis. The latent structure of agoraphobia was examined in two independent samples, one comprising outpatients presenting for treatment for panic disorder (PD) with or without agoraphobia (n = 365), and the other comprising community volunteers to a national mental health survey who experienced fear or avoidance of at least one prototypic agoraphobic situation (n = 640). Two taxometric procedures were carried out – maximum eigenvalue (MAXEIG) and mean above minus below a cut (MAMBAC) – using indicators derived from questionnaire measures of, and structured diagnostic interviews for, agoraphobia. Results show consistent evidence of dimensional latent structure in both samples. It is concluded that scores on measures of agoraphobia best represent an agoraphobic severity dimension.

Agoraphobia is an anxiety-related construct with a poorly understood underlying structure and a frequently evolving status within the psychiatric nomenclature (Bienvenu et al., 2006 and Schmidt et al., 2005). It is defined in the Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition Text Revision (DSM-IV TR; American Psychiatric Association, 2000) as the fear and avoidance of situations in which escape might be difficult due to concern about having a panic attack or panic symptoms. The essential nature of agoraphobia and its exact relation to panic disorder (PD) are controversial. Although PD and agoraphobia may be diagnosed independently ( American Psychiatric Association, 2000), there are strong similarities between these disorders with respect to demographic and clinical characteristics ( Andrews and Slade, 2002, Garvey and Tuason, 1984 and Goisman et al., 1995), as well as cognitions and anxiety sensitivity ( Cox, Endler, & Swinson, 1995). Agoraphobia and PD patients also respond to the same medication ( Garkani, Gorman, Liebowitz, & Klein, 1984) and behavioural treatments ( Craske, DeCola, Sachs, & Pontillo, 2003). The majority of research suggests that agoraphobic avoidance usually develops following panic attacks or panic-like sensations (Barlow et al., 1994 and Goisman et al., 1994). Accordingly, several theorists have posited that PD is the first stage of a two-stage disorder with agoraphobic fear as the resulting consequence (Garvey and Tuason, 1984, Thyer et al., 1985 and Turner et al., 1986). However, several retrospective investigations have suggested that agoraphobic symptoms precede panic attacks in some patients (Fava & Mangelli, 1999). This view is supported by findings from retrospective epidemiological surveys in the general community, which have suggested that panic symptomatology is the trigger for agoraphobic avoidance in only a minority of people (Wittchen, Reed, & Kessler, 1998). More recently, Bienvenu et al. (2006) reported results from a prospective study in which baseline agoraphobia without PD predicted first-onset panic disorder with agoraphobia (PDA). On the basis of these findings the authors argued that agoraphobia should be classified as an independent diagnosis. Despite continuing controversy surrounding the diagnosis of agoraphobia, no studies to date have examined its fundamental underlying structure. Understanding whether avoidance behaviour operates in a continuous or categorical fashion may help to provide an empirical basis for its placement within the DSM-V classification system and lead to more accurate modelling of this important construct ( Ruscio & Ruscio, 2002b). Taxometric analysis provides a set of statistical procedures that may be used to determine the latent structure of a construct ( Meehl, 1995, Schmidt et al., 2004 and Waller and Meehl, 1998). Taxometric analyses have been used to evaluate the latent structure of more than 20 psychopathological constructs, including a number of anxiety and mood-related constructs. Ruscio, Borkovec, and Ruscio (2001) used this approach to demonstrate that the latent structure of worry was dimensional within a sample of undergraduate students. Similarly, taxometric studies of posttraumatic stress disorder (Ruscio, Ruscio, & Keane, 2002) and social phobia (Kollman, Brown, Liverant, & Hofmann, 2006) supported the dimensionality of these constructs. Results of most taxometric investigations of depression have also favoured the continuum position (Beach and Amir, 2003, Beach and Amir, 2006, Ruscio and Ruscio, 2002a, Ruscio and Ruscio, 2000, Slade, 2007 and Slade and Andrews, 2005). Although there is growing evidence in support of a continuous perspective of psychopathology, there are a few notable exceptions, such as anxiety sensitivity (AS). Although AS has been conceptualized as a universal characteristic present to varying degrees in all people, taxometric findings across several populations indicate it may be more appropriate to conceptualize two forms of AS, a taxon (pathological form) and a non-taxon (normative) form (Bernstein et al., 2006 and Schmidt et al., 2006). These findings are relevant to the present study, as AS is proposed to be an important vulnerability factor for panic-related symptomatology. In addition to AS, a study by Kotov, Schmidt, Lerew, Joiner, and Ialongo (2005) investigated the underlying structure of anxiety itself, defined as a multifaceted construct including physiological arousal, subjective anxiety, threat-related cognitions, and behavioural change (Barlow, 2002). Interestingly, this study found evidence of an underlying anxiety taxon. As agoraphobia is conceptually related to AS, it may also operate in a categorical fashion, consistent with its placement in the current DSM ( American Psychiatric Association, 2000). If this is the case, individuals with panic-related pathology would be best conceptualized as being within or outside the agoraphobia taxon. On the other hand, agoraphobia is associated with role impairment and clinical and cognitive correlates of panic severity ( Craske and Barlow, 1988, Langs et al., 2000, Rapee and Murrell, 1988 and Telch et al., 1989), leading some to suggest that it is a indicator of severity of PD ( Kessler et al., 2006 and Noyes et al., 1987). If PD and PDA differ only quantitatively by the degree of avoidance exhibited by the individual, agoraphobia should be conceptualized as a latent dimension among individuals with PD ( Turner et al., 1986). The primary aim of the current investigation was to better understand the nature of agoraphobia through the use of taxometric analysis. We examined the latent structure of agoraphobia within two different types of samples, a clinical sample of individuals who were in treatment for PD or PDA at a community mental health centre and a community sample who were randomly selected to complete a national mental health survey. To our knowledge only one other study (Slade, 2007) has conducted taxometric analyses on a psychiatric construct across clinical and community samples. Whereas findings of a particular latent structure in one sample may be an artefact of sample type, consistent structural findings across multiple samples constitute powerful evidence for a particular latent structure of agoraphobia.