Excessive worry is the hallmark feature of Generalized Anxiety Disorder [GAD]. Whilst this is primarily a cognitive phenomenon, the disorder also has physiological, affective, and behavioral aspects, such as somatic symptoms, distress, and avoidance. Cognitive psychologists believe that targeting and reducing worry is the key to recovery, due to the relationships between thoughts, actions, emotions, and bodily sensations. Researchers such as Dugas et al (1998) suggest that worry has four main components: intolerance of uncertainty, positive beliefs about worry, poor problem orientation, and cognitive avoidance.  Intolerance of uncertainty has become a prime target in CBT.

The primary aim of this study was to assess the efficacy of a cognitive-behavioural treatment for GAD that exclusively targets worry, with the assumption that it would lead to a significant reduction in symptoms both in the short and long term.

What is meant by "positive beliefs about worry?"

• Worry prevents negative outcomes and promotes positive outcomes

• Worry minimizes guilt if bad outcomes do occur. 

• Worry is motivating 

• Worry prepares you for the worst

• Worry helps you be a more analytical thinker

• Worry is a distraction from emotional topics

Participants were recruited via a newspaper article. Structured telephone interviews were used to screen out anyone who did not appear to have GAD. The remaining participants underwent screening using the Anxiety Disorders Interview Schedule for DSM-IV to ensure they reached the criteria for a diagnosis of GAD. The final sample comprised 20 females and 6 males (mean age 40), all of whom had experienced GAD symptoms for around 16 years.

The Penn State Worry Questionnaire (PSWQ) and The Beck Depression Inventory were used to measure worry and depressive symptoms, plus a number of other tests were used to measure trait anxiety and intolerance of uncertainty. The Significant Other Rating Scale (SORS) was given to people who were close to the participant such as friends and family, to gain an additional measure of the person’s symptoms, thus avoiding issues with self-reported data. Measures of the participants’ perception of their therapist (e.g. caring, involved), the credibility of the treatment, and the extent to which they expected it to be successful were also recorded.

Participants were randomly allocated to either the treatment group or the wait-list control group, who were told that they would start treatment in 16 weeks time. One-hour weekly sessions were delivered by four CBT-trained therapists. Sessions targeted the four aspects of worry identified in Dugas et al. (1998) (see above) and participants monitored their worries between sessions.

Individualized relapse prevention programs were presented at the end of the treatment program. Participants were reassessed at 6 and 12 months to monitor remission and relapse.

Post-treatment scores were significantly lower on all measures in the treatment group but not in the wait-list control group. For example, scores on the PSWQ reduced from 65.86 before treatment to 45.64 after treatment, whereas scores actually went up on the waiting list group from 59.25 to 64.58.

The researchers also analyzed the impact of the treatment on the wait-list control group delivered 4 months later, and the results were replicated; participants showed significant improvements across all measures. Improvements were maintained over time with no significant increase in symptom severity at 6 or 12 months. For example, PSWQ scores were maintained at 46.88 (at 6 months) and 46.27 (at 12 months). These findings also held up when looking at the ratings of friends and family, using the data from the SORS.

Post-treatment, 81% of the sample scored within one standard deviation of the mean PSWQ scores for the general population, although this had dropped to 73% at 6 months and 69% at 12 months. This suggests some increase in worry related symptoms over time, suggesting top-up CBT sessions may be advantageous over the long term. This said, 77% of the sample no longer met the criteria for GAD at 6 and 12 months, suggesting it was an effective treatment for the majority of the sample.

Strengths of this study include the fact that all interviews were tape-recorded and a second clinician listened to the recordings to ensure the reliability of the GAD diagnoses. When the clinicians failed to agree whether GAD was the most severe psychological disorder these people were not included in the final sample.

A limitation of the sample was that it was extremely gynocentric, meaning there were many more females than males. Although this is in keeping with gender differences in the prevalence of anxiety disorders such as GAD, it does mean that it is difficult to know whether males would benefit as much from CBT focusing on worries around uncertainty as a way of reducing overall GAD symptoms.

Another strength is that there was no attrition in either group; this is excellent as many studies suffer from participant drop-out that affects the analysis. For example, researchers may use "last observed" scores in place of missing data yet it is unclear how the participants would have fared had they remained in the study. Participants who drop out are often those for whom the therapy was making the least impact, meaning efficacy can be falsely inflated by attrition, but in this study that clearly was not the case.

A limitation of the study is that the therapists were all experienced CBT practitioners meaning the results may not generalize as successfully for therapists with less training and less confidence in their own ability to bring about lasting improvements for their clients.