Efficacy of a Cognitive—Behavioral Treatment for Generalized Anxiety Disorder Evaluation in a Controlled Clinical Trial Robert Ladouceur École de Psychologie Université Laval Michel J. Dugas Department of Psychology Corcordia University Mark H. Freeston Centre de Recherche Fernad-Seguin Eliane Léger École de Psychologie Université Laval Fabien Gagnon Center Hospitalier Université Laval Nicole Thibodeau Center Hospitalier Université Laval ABSTRACT Recent advances in the understanding of worry have led to the development of treatments for generalized anxiety disorder (GAD). The present study tested a GAD treatment that targeted intolerance of uncertainty, erroneous beliefs about worry, poor problem orientation, and cognitive avoidance. Twenty-six primary GAD patients were randomly allocated to a treatment condition ( n = 14) or a delayed treatment control condition ( n = 12). Self-report, clinician, and significant other ratings assessed GAD and associated symptoms. The results show that the treatment led to statistically and clinically significant change at posttest and that gains were maintained at 6- and 12-month follow-ups. Furthermore, 20 of 26 participants (77%) no longer met GAD diagnostic criteria following treatment. With regard to the treatment's underlying model, the results show that intolerance of uncertainty significantly decreased over treatment and that gains were maintained at both follow-ups. Although nonspecific factors were not significant predictors of treatment outcome, their role in the treatment of GAD requires further investigation. Since the inclusion of generalized anxiety disorder (GAD) in the Diagnostic and Statistical Manual of Mental Disorders (3rd ed. [DSM—III] ; American Psychiatric Association [APA], 1980 ), its diagnostic criteria have undergone sweeping changes. In 1987, the main diagnostic criteria for GAD in the DSM—III—R ( APA, 1987 ) ceased to be a series of somatic symptoms and became the presence of excessive worry. Seven years later, following a period of intense research on worry, the DSM—IV ( APA, 1994 ) retained excessive worry as the hallmark of GAD, and this has led to improved identification and, in some cases, improved treatment outcomes (see Dugas & Ladouceur, 1998 , for a review). Borkovec and his collaborators have underscored that anxiety involves a process of interacting subsystems (e.g., cognitive, physiological, affective, and behavioral; see Borkovec & Costello, 1993 ; Borkovec & Newman, 1999 ). In other words, changes in one subsystem may lead to changes in the others. Thus, if people decrease their level of worry, which is first and foremost a cognitive phenomenon, this should lead to changes in their physiological responding, their subjective level of Journal of Consulting and Clinical Psychology © 2000 by the American Psychological Association December 2000 Vol. 68, No. 6, 957-964 For personal use only--not for distribution. Page 1 of 149/5/2001http://spider.apa.org/ftdocs/ccp/2000/december/ccp686957.htmlaffect, and their worry-related behaviors. Given the recent advances in our understanding of worry, it may be possible to design a treatment intervention for GAD that exclusively focuses on worry to lead to changes in all interacting subsystems. For instance, if people with GAD learn to control their worry, they may find that their somatic symptoms decrease accordingly (i.e., restlessness, frequent fatigue, difficulty concentrating, irritability, muscle tension, and sleep disturbance). By intensively focussing on worry, a key feature of the cognitive subsystem, rather than attempting to target many interacting subsystems, a cognitive—behavioral intervention may lead to clinically significant change in the cognitive as well as other subsystems. The most obvious implication of this reasoning is that a potentially effective treatment for GAD could draw on recent applied research to target worry in a highly specific fashion. The second implication is that relaxation training, which represents one of the most common treatment components for GAD (see Barlow, Rapee, & Brown, 1992 ; Borkovec & Costello, 1993 ), would not be applied. The present study tests such a treatment package. The main goal of the proposed treatment is of course to decrease the tendency to worry and eliminate GAD. To do so, the intervention targets the four components of our GAD worry model: intolerance of uncertainty, erroneous beliefs about worry, poor problem orientation, and cognitive avoidance (see Dugas, Gagnon, Ladouceur, & Freeston, 1998 , for a detailed description of the model). First, the treatment stresses the importance of dealing with uncertainty in everyday life. Applied research has shown that GAD patients have a lower threshold for uncertainty than do other anxiety disorder patients and nonclinical individuals ( Ladouceur et al., 1999 ). Other research teams have also shown that nonclinical high worriers have difficulty dealing with uncertainty in problem-solving situations ( Metzger, Miller, Cohen, Sofka, & Borkovec, 1990 ; Tallis, Eysenck, & Mathews, 1991 ). Finally, experimental manipulations of intolerance of uncertainty in nonclinical participants lead to changes in their level of worry, with increased intolerance of uncertainty leading to greater worry ( Ladouceur, Gosselin, & Dugas, in press ). Second, the treatment attempts to correct erroneous beliefs about worry. Compared with moderate worriers, nonclinical high worriers believe that worry is useful because it helps prevent negative outcomes from occurring and minimizes the negative effects (e.g., feelings of guilt or shame) if these outcomes should occur ( Freeston, Rhéaume, Letarte, Dugas, & Ladouceur, 1994 ). Nonclinical high worriers also believe that worry is more useful because it motivates them, prepares them for the worst, and distracts them from emotional topics ( Borkovec & Roemer, 1995 ). Another independent study has shown that individuals with high levels of worry believe that worrying helps analytical thinking ( Davey, Tallis, & Capuzzo, 1996 ). Finally, compared with nonclinical study participants, GAD patients report that worrying is more effective to avoid negative outcomes and promote positive ones ( Ladouceur et al., 1999 ). The treatment also targets poor problem orientation. A growing body of evidence shows that high levels of worry are related to poor problem orientation (i.e., the person's cognitive set when faced with a
View Full Document
Unlocking...