Über die Autorin bzw. den Autor

Graham Davey is Professor of Psychology at the University of Sussex in Brighton, UK. He has been an active researcher in areas associated with anxiety and its disorders, especially pathological worrying, phobias, and perseverative psychopathologies generally. He has published his research in many high-impact international journals including Journal of Abnormal Psychology, Behavioral & Brain Sciences, Journal of Experimental Psychology, and Behaviour Research & Therapy. He has authored or edited a total of 11 books, including Davey, G.C.L. (1997) (Ed) Phobias: A handbook of theory, research and treatment, Chichester:Wiley,andDavey, G.C.L.&Tallis, F. (1994) (Eds) Worrying: Perspectives on theory, assessment and treatment, Chichester: Wiley. He was elected President of the British Psychological Society from 2002–2003.

Adrian Wells is Professor of Clinical & Experimental Psychopathology at the University of Manchester, and Professor II in Clinical Psychology at the Norwegian University of Science and Technology, Trondheim. He is Honorary Consultant Clinical Psychologist in Manchester Mental Health Trust. He has published over 100 scientific papers and book chapters in the area of cognitive theory and therapy of emotional disorders. His books include Wells, A. (1997) Cognitive Therapy of Anxiety Disorders: A Practice Manual and Conceptual Guide, Chichester, UK: Wiley, and Wells, A. (2000) Emotional Disorders and Metacognition: Innovative Cognitive Therapy, Chichester, UK: Wiley. He is the originator of metacognitive therapy and has also contributed to the development of cognitive therapy for anxiety disorders. He is a founding fellow of the Academy of Cognitive Therapy, USA.

Von der hinteren Coverseite

Anxiety-based disorders are among the most common mental health problems experienced in the population today. Worry is a prominent feature of most anxiety-based disorders including generalized anxiety disorder, specific phobias, obsessive-compulsive disorder, panic disorder, and post-traumatic stress disorder.

Written by international experts, Worry and its Psychological Disorders offers an up-to-date and complete overview of worry in a single volume. Divided into four sections, the book explores the nature of worry, the assessment of worry, contemporary theories of chronic and pathological worry, and the most recently developed treatment methods. It includes in-depth reviews of new assessment instruments and covers treatment methods such as Cognitive Behavioural Therapy and Metacognitive Therapy. Useful case studies are also included.

This important volume provides an invaluable resource for clinical practitioners and researchers. It will also be of relevance to those studying clinical or abnormal psychology at advanced level.

Auszug. © Genehmigter Nachdruck. Alle Rechte vorbehalten.

Worry and its Psychological Disorders

John Wiley & Sons

Chapter One

Robert M. Holaway, Thomas L. Rodebaugh and Richard G. Heimberg

THE EPIDEMIOLOGY OF WORRY AND GENERALIZED ANXIETY DISORDER

Once considered synonymous with the cognitive components of anxiety (Mathews, 1990; O'Neill, 1985), worry has emerged as a more specific construct that can not only be distinguished from a larger subset of cognitive aspects of anxiety, but also studied in its own right (Davey, 1993; Davey, Hampton, Farrell & Davidson, 1992; Zebb & Beck, 1998). One of the first attempts to define worry was provided by Borkovec, Robinson, Pruzinsky, and DePree (1983, p. 10):

Worry is a chain of thoughts and images, negatively affect-laden and relatively uncontrollable; it represents an attempt to engage in mental problem-solving on an issue whose outcome is uncertain but contains the possibility of one or more negative outcomes; consequently, worry relates closely to the fear process.

More recent formulations have extended this definition of worry, describing it as an anxious apprehension for future, negative events (Barlow, 2002) that involves "a predominance of negatively valenced verbal thought activity" and minimal levels of imagery (Borkovec, Ray & Stober, 1998, p. 562). These definitions have been largely derived from participants' reports regarding what they do when they worry.

Research on the epidemiology of worry has largely evolved over the past 20 years. Much research appears to have been spurred by the adoption of worry as the essential feature of generalized anxiety disorder (GAD) in the revised, third edition of the Diagnostic and Statistical Manual of Mental Disorders (DSM-III-R; American Psychiatric Association [APA], 1987). These studies have provided valuable data regarding the prevalence, content, and functions of worry and GAD. In this chapter, we review the existing research on the prevalence and phenomenology of worry (both normal and pathological) and GAD and available data on gender, age, ethnic, and cultural differences in the manifestation and occurrence of both phenomena.

The Phenomenology of Normal and Pathological Worry

Few empirical studies have actually examined the occurrence and phenomenology of worry independent of GAD (Tallis, Davey & Capuzzo, 1994). As a result, much of our empirical understanding regarding what actually occurs when people worry, what they most often worry about, and how frequently they worry has been derived from examinations of nonanxious control groups. As noted by Ruscio (2002), these studies may not provide an accurate representation of the frequency and manifestation of normal worry because participants in these groups have been selected based on low worry scores and an absence of anxiety. In much of the empirical literature, normal worry has been regarded as "mild, transient, generally limited in scope, and experienced by the majority of individuals" (Ruscio, 2002, p. 378). However, without adequate studies of worry in normal individuals (i.e., not simply low-anxiety individuals), it is difficult to determine how much the above perception is based on specific characteristics of the available samples.

Tallis and colleagues (1994) conducted one of the few direct examinations of the phenomenology of non-pathological worry. In a mixed sample of 128 university students and working adults (aged 18-59), 38% reported worrying at least once per day; 19.4% indicated they worried once every 2-3 days; and 15.3% reported they worried about once a month. It is unclear how frequently the remaining 27.3% experienced worry. Participants were also asked how long their worry episodes typically lasted. About 24% reported that their worries were fleeting or lasted less than 1 minute, and 38% endorsed a typical duration of 1-10 minutes. The remainder endorsed longer durations of their typical worry (18%, 10-60 minutes; 11%, 1-2 hours; 9%, two hours or more). In addition, participants reported that they most often worried during the late evening or early morning hours and that their worries frequently occurred in response to impending matters, such as upcoming events or interpersonal interactions (Tallis et al., 1994). Participants' mean score on a measure of pathological worry fell in the moderate range and was significantly lower than scores typically associated with a diagnosis of GAD (see Fresco, Mennin, Heimberg & Turk, 2003; Molina & Borkovec, 1994). In terms of worry content, 17% of respondents reported they worried most often about their competence at work, followed by academic performance (11%), health issues (10%), financial circumstances (10%), and intimate relationships (9%). Finally, 83% of respondents reported that they believed worry helped them to find solutions to problems in their environment (Tallis et al., 1994). This finding is, to some extent, consistent with recent research by Szab and Lovibond (2002), in which 48% of naturally occurring worry episodes primarily reflected a problem-solving process (i.e., using worry to generate solutions to problematic situations), whereas 17% were characterized as primarily involving the anticipation of negative outcomes. Further examination revealed that more severe levels of worry were associated with reduced problem-solving success, although the causal direction of this relationship is unclear.

Studies have consistently, and perhaps not surprisingly, found that people who experience pathological worry as a part of GAD rate their worry as more pervasive and less controllable than people without pathological worry. Craske, Rapee, Jackel and Barlow (1989) examined several dimensions of worry by comparing individuals with DSM-III-RGAD to a nonanxious control group consisting of friends of clients receiving treatment for anxiety. Both groups reported similar ratings of worry duration, worry aversiveness, attempts to resist worry, anxiety associated with resisting worry, and perceived likelihood of the occurrence of worrisome outcomes. However, individuals in the nonanxious control group reported that they worried, on average, 18.2% of the day during the past month compared to 60.7% reported by the GAD group. In addition, nonanxious individuals rated their worries as more controllable, reported greater success in resisting or reducing their worries, indicated that their worries were more often associated with a specific and discernable precipitant, and perceived their worries to be more realistic than those reported by individuals with GAD. Other studies have also found differences with respect to the pervasiveness of worry, as nonanxious controls have consistently reported fewer worrisome topics than individuals with GAD (Borkovec, Shadick & Hopkins, 1991; Dugas et al., 1998; Hoyer, Becker & Roth, 2001; Roemer, Molina & Borkovec, 1997).

Studies comparing the content of worry among individuals with GAD and nonanxious controls have typically reported on the frequency of specific worry domains: 1) work and school, 2) family and interpersonal relationships, 3) financial issues, 4) illness, health, and injury, and 5) miscellaneous topics (e.g., minor matters, punctuality, home repairs). Across several investigations, roughly one-third of participants' worries, regardless of GAD status, have pertained to family and interpersonal issues (Borkovec et al., 1991; Craske et al., 1989; Roemer et al., 1997). Relationships thus seem to be a common source of worry, a conclusion further bolstered by two studies finding that over 70% of people with GAD endorsed frequent worry about either family or relationships (Dugas et al., 1998; Sanderson & Barlow, 1990). Contrary to this conclusion, Craske and colleagues (1989) found health and injury to be the most frequently reported topic of worry among individuals with GAD (30.6% of reported worries). However, most studies report health and injury worries to be rather infrequent in both GAD (Borkovec et al., 1991; Dugas et al., 1998; Roemer et al., 1997; Sanderson & Barlow, 1990) and nonanxious control groups (Craske et al., 1989; Roemer et al., 1997).

The most consistent finding regarding differences in worry content between nonanxious controls and GAD samples has pertained to worry regarding miscellaneous topics, such as car troubles or being late for appointments. Across three studies, miscellaneous worry topics reported by nonanxious control groups comprised 0%-19.7% of all reported worries, whereas miscellaneous worries among individuals with GAD comprised between 25.2%-31.3% of reported worries (Borkovec et al., 1991; Craske et al., 1989; Roemer et al., 1997). Other content differences between individuals with and without GAD have been observed with regard to worry about work and school. Two studies found nonanxious controls to report a greater proportion of worries related to work and school (30.4%-36.6%) than individuals with GAD (13.9%-22%) (Craske et al., 1989; Roemer et al., 1997), although Borkovec and colleagues (1991) found the opposite. The conclusion that people in nonanxious control groups worry more about work and school is fairly consistent with Tallis and colleagues' (1994) assessment of non-pathological worry, in which the most frequent topics of concern reported by participants pertained to academic performance and competence at work. Similar to concerns regarding work and school, worries about financial circumstances have generally been more frequently reported by individuals without GAD, as two studies have reported the proportion of total worries pertaining to finances to range from 12.5%-26.1% among nonanxious control groups and 2.8%-8.9% among GAD samples (Borkovec et al., 1991; Craske et al., 1989). However, in contrast, Roemer and colleagues (1997) found individuals with GAD to report a greater proportion of worries related to financial circumstances (10.8%) than nonanxious controls (5.6%).

The studies reviewed above have revealed several similarities and differences in the phenomenology of worry among individuals with and without GAD. Most notably, individuals with GAD spend significantly more time worrying, report more worry topics, and perceive themselves as having considerably less control over their worry than nonanxious controls. In addition, miscellaneous worry topics appear to be more prevalent among individuals with GAD than nonanxious controls. Most similarities observed between the two groups have regarded the frequency of worries pertaining to family and interpersonal relationships, with roughly a third of all reported worries relating to this topic.

Despite these general patterns, there have been many inconsistencies across studies. Several factors may account for these differences. First, with the exception of Roemer et al. (1997), sample sizes for both GAD and nonanxious control groups have been relatively small (e.g., n = 13-31), which may limit external validity. Second, the manner in which the frequency and content of worry was assessed varied by study. For example, whereas participants in the Craske et al. (1989) study monitored and recorded the nature of their worry each day for three weeks, other studies have assessed worry phenomenology using diagnostic interviews (e.g., Roemer et al., 1997). Finally, demographic differences across study samples, especially with respect to age, gender, and employment, may have influenced the frequency of specific worry topics, as these concerns seem likely to shift according to the nature of one's daily life.

Differentiating Pathological Worry from GAD

Recent research by Ruscio, Borkovec, and Ruscio (2001) has provided empirical support for a dimensional structure of worry, suggesting that normal and pathological worry represent opposite ends of a continuum, not discrete constructs. However, in most cases, investigations of normal and pathological worry have typically examined individuals with a diagnosis of GAD and have rarely examined pathological worry independent of GAD, leaving pathological worry outside the context of GAD poorly understood (Ruscio, 2002; Ruscio & Borkovec, 2004).

In an attempt to identify delimiting characteristics of pathological worry and GAD, Ruscio (2002) recently compared high worriers with and without a diagnosis of GAD. Surprisingly, only 20% of individuals who reported experiencing extreme levels of pathological worry (worry scores above the threshold commonly associated with GAD) actually met diagnostic criteria for the disorder. Follow-up analyses indicated that, across two samples, 68%-78% of people who reported high levels of worry but not GAD met only 0-1 of the four required DSM-IV criteria, with chronic/excessive worry and associated distress and impairment best differentiating individuals with GAD from high worriers without GAD (Ruscio, 2002, Study 1). Individuals with GAD also reported greater levels of depression, more frequent worry, and less control over their worry. In a follow-up study, individuals with high levels of worry but without GAD experienced all symptoms of GAD less severely than individuals with GAD, even though they reported their worry to be excessive and uncontrollable (Ruscio, 2002, Study 2).

Ruscio's (2002) findings underscore the need for future studies to distinguish GAD from pathological worry. Specifically, they suggest that examining differences between worry in normal participants and participants with GAD may not actually provide information about the differences between nonpathological and pathological worry. In a recent comparison of people with high worry who either did or did not have GAD, Ruscio and Borkovec (2004) found that negative beliefs about worry (e.g., "worry is harmful") were specific to participants with GAD. In line with Roemer and colleagues' (1997) position that worry may function as a strategy for avoidance of more emotional topics among persons with GAD, Holaway, Hambrick and Heimberg (2003) found that people with GAD reported experiencing their emotions as more intense and more confusing than people without GAD who experienced high levels of worry. Such results, although preliminary, suggest that pathological worry within the context of GAD may be subject to additional factors (e.g., different beliefs about worry, increased emotion dysregulation) that may render it significantly different from pathological worry without GAD. This caveat should be kept in mind when large-scale epidemiological studies, which concern GAD rather than worry per se, are reviewed below.

The Epidemiology of Generalized Anxiety Disorder

Since their first iteration in DSM-III (APA, 1980) to their current version in DSM-IV (APA, 1994), the diagnostic criteria for GAD have been revised repeatedly, with revisions resulting in a greater focus on the presence of excessive and uncontrollable worry, an increase in the required duration of symptoms, fewer required physical symptoms, and the added requirement that worry and associated symptoms be accompanied by significant distress or impairment. In later editions, GAD was no longer considered a residual category that could only be diagnosed in the absence of other anxiety disorders. These significant changes to the structure of GAD have hampered long-term investigations of the course of the disorder and resulted in considerable heterogeneity in studies examining prevalence rates (Kessler, Walters & Wittchen, 2004; Wittchen, Zhao, Kessler & Eaton, 1994). Nevertheless, several epidemiological surveys provide valuable information regarding the prevalence, course, and associated features of GAD.

Prevalence

Table 1.1 shows the current, 12-month, and lifetime prevalence rates for GAD in population-based surveys of adults conducted in several countries around the world. Most likely because the diagnostic criteria for GAD in DSM-III-R are more stringent than the criteria in DSM-III, prevalence rates appear to have dropped from studies employing DSM-III to those using DSM-III-R. Though lifetime prevalence rates of DSM-IV GAD among adults in the general population have yet to be reported, existing studies have found the current and 12-month prevalence rates for the disorder to be equivalent to, or perhaps slightly higher than the rates found using the DSM-III-R.

(Continues...)

Excerpted from Worry and its Psychological Disorders Copyright © 2006 by John Wiley & Sons, Ltd. Excerpted by permission.
All rights reserved. No part of this excerpt may be reproduced or reprinted without permission in writing from the publisher.
Excerpts are provided by Dial-A-Book Inc. solely for the personal use of visitors to this web site.