Generalized anxiety disorder (GAD) is a condition characterized by chronic and excessive worry and anxiety. Over the past 15 years, the investigators have developed a cognitive-behavioural treatment that leads to the remission of GAD in approximately 60% to 75% of affected individuals. Although these numbers are encouraging, there remain a considerable proportion of patients who do not fully benefit from treatment. With the goal of improving treatment efficacy, the investigators have recently carried out a series of related studies on the way individuals with GAD and high worriers process uncertain or ambiguous information from their environment. The findings show that these individuals display biases in attention for, and appraisal of, uncertain or ambiguous information. Specifically, individuals with GAD and high worriers preferentially allocate their attention to uncertainty-related stimuli and appraise ambiguous information in a threatening manner. In this study, the investigators examine the impact of these information processing biases, measured at intake, on the efficacy of cognitive-behavioural treatment for GAD. The investigators also examine the impact of residual information processing biases, measured at posttreatment, on the maintenance of treatment gains over 18 months following treatment. The main hypotheses are (1) that high levels of pretreatment biases will predict poorer outcomes immediately following therapy, and (2) that high levels of posttreatment biases will predict relapse during the 18 months following therapy. If, as expected, information processing biases predict poor short- and long-term treatment outcomes for individuals with GAD, the investigators will expand the treatment to integrate strategies that directly target these biases in order to increase its efficacy.

Background: The investigators have developed a cognitive model of generalized anxiety disorder (GAD) that has intolerance of uncertainty as its central component. Based on this model, they have devised a cognitive-behavioural treatment (CBT) for GAD that helps affected individuals recognize, accept, and deal with the uncertainty of everyday life. Although the treatment leads to positive outcomes in most patients, a significant minority of treated individuals do not attain full remission. One potential avenue for improving the treatment of GAD is to identify disorder-specific modes of cognitive processing (or information processing) and to directly target these modes of processing in therapy. Accordingly, the investigators have carried out a series of related studies examining the relationship between cognitive processing and GAD. The findings from these studies show that individuals who have clinical or subclinical GAD (1) preferentially allocate their attention to uncertainty-related stimuli, and (2) make threatening appraisals of ambiguous information. Further, the data show that the tendency to make threatening appraisals of ambiguous information mediates the relationship between intolerance of uncertainty and the symptoms of GAD. This latter finding is consistent with cognitive theory, which asserts that information processing mediates the influence of cognitive vulnerability on the expression of symptoms of emotional disorders. This study aims to extend this line of research by investigating the impact of these cognitive processing biases on treatment outcome for patients with GAD.

Goals and hypotheses: The main goal of the study is to examine the impact of biases in cognitive processing on short- and long-term outcomes of cognitive-behavioural therapy for individuals with GAD. The study's main hypotheses are that: (1) preferential allocation of attention to uncertainty-related stimuli, assessed at pretreatment, will predict poorer response to treatment; (2) the tendency to appraise ambiguous information in a threatening manner, assessed at pretreatment, will predict poorer response to treatment; (3) preferential allocation of attention to uncertainty-related stimuli, assessed at posttreatment, will predict relapse during follow-up; and (4) the tendency to appraise ambiguous information in a threatening manner, assessed at posttreatment, will predict relapse during follow-up.

Method: The final sample consists of 80 adult patients with a principal diagnosis of GAD, recruited from the Anxiety Disorders Clinic of Sacré-Cœur Hospital of Montreal. Participants are assessed at 9 measurement times: pretreatment, midtreatment, posttreatment, and 3-, 6-, 9-, 12-, 15- and 18-month follow-up. Assessments include the Anxiety Disorders Interview Schedule for DSM-IV, a cognitive processing task (Ambiguous/Unambiguous Situations Diary), and a battery of standardized self-report measures. Treatment consists of an empirically supported CBT protocol for GAD, which is administered over 16 weekly sessions using a session-by-session treatment manual developed in earlier studies. Growth curve analysis with multilevel modeling will be the main analytic strategy used to determine the relationships between cognitive processing and change in a range of outcome variables while controlling for relevant clinical and sociodemographic variables.

Implications: The study has important theoretical and clinical implications. In terms of theory, it begins to bridge the gap between the considerable knowledge of the role of cognitive processing in anxiety and the lack of knowledge of the impact of cognitive processing on treatment outcomes. Surprisingly, although the role of cognitive processing in anxiety has been intensely studied, the impact of cognitive processing on GAD psychotherapy outcomes has never been examined. The proposed study also informs clinical practice as to the importance of (1) systematically assessing cognitive processing, and (2) integrating treatment interventions that specifically target biased cognitive processing into current GAD treatment protocols.

Inclusion Criteria:

• At least 18 years of age
• Primary diagnosis of GAD
• Stability of medication in 4 to 12 weeks before study entry (4 weeks for benzodiazepines, 12 weeks for other medications)
• Willingness to keep medication status stable while participating in the study

Exclusion Criteria:

• Use of herbal products known to have CNS effects in the 2 weeks before study entry
• Evidence of suicidal intent (based on clinical judgement)
• Evidence of current substance abuse, current or past schizophrenia, bipolar disorder or organic mental disorder
• Participation in other trials
• Evidence of anxiety symptoms due to a general medical condition based on clinical judgement (e.g., clinical hyperthyroidism, hypoglycemia, anemia)