Generalized Anxiety Disorder 439
Seeds, & Dozois, 2008; Mohlman, 2004), with approximately 45–54% no longer meet-
ing diagnostic criteria for GAD at posttreatment (Wetherell, Gatz, & Craske, 2003;
Stanley et al., 2003).
In a large meta- analysis involving 65 studies Mitte (2005) concluded that CBT was
a highly effective treatment for GAD (i.e., average effect size = .82 for anxiety when
compared to no treatment controls) as indicated by reductions in both primary anxiety
symptoms and depression (see also Covin et al., 2008, for similar conclusion). However,
there was no consistent evidence that CBT was significantly more effective than phar-
macotherapy, leading the author to conclude that CBT is at least equivalent in effective-
ness to pharmacotherapy. However, CBT may be better tolerated and may have more
enduring effects at least when compared to the benzodiazepines (Gould, Otto, Pollack,
& Yap, 1997; Mitte, 2005).
Unfortunately, very few treatment studies have reported follow-up periods greater
than 12 months. The one exception is an 8- to 14-year follow-up conducted on two
studies in which cognitive behavior therapy was one of the randomly assigned treatment
conditions (Durham, Chambers, MacDonald, Power, & Major, 2003). There was a
trend for the CBT groups to be more improved than the non-CBT conditions at post-
treatment but the differences were not statistically significant. At long-term follow-up
the majority of patients were symptomatic although average symptom severity was still
lower than the pretreatment level, indicating maintenance of symptom improvement.
However, there were no appreciable differences in recovery rates between the CBT and
non-CBT conditions. In a later 2- to 14-year follow-up study involving individuals who
participated in one of eight randomized clinical trials of CBT for GAD, PTSD, or panic
disorder, the authors concluded that “CBT was associated with a better long-term out-
come than non-CBT in terms of overall symptom severity but not with regards to diag-
nostic status” (Durham et al., 2005, p. iii). Overall, then, we cannot say with any degree
of certainty that cognitive therapy or CBT produces more enduring treatment gains in
GAD, although some of the findings are somewhat promising.
A number of studies have directly compared cognitive therapy and applied relax-
ation training. Generally both interventions produce equivalent treatment effects at post-
treatment and follow-up, with recovery rates in the 53–67% range (e.g., Arntz, 2003;
Borkovec & Costello, 1993; Borkovec et al., 2002; Öst & Breitholtz, 2000). Cognitive
therapy is significantly more effective than analytic psychotherapy, with approximately
twice as many CBT patients reporting clinically significant improvement at posttreat-
ment and follow-up than the psychodynamic group (Durham et al., 1999; Durham et
al., 1994). Moreover, cognitive therapy appears to produce more enduring clinically
significant change than anxiety management (e.g., Durham et al., 1999; Durham et al.,
1994).
There have been some attempts to determine if certain treatment modifications
might improve the effectiveness of CBT for GAD. Fisher (2006) concluded from his
updated review of recovery rates that the efficacy of cognitive therapy, CBT, and applied
relaxation is highly variable across studies and thus rather limited. However, initial
outcome studies of more innovative cognitive therapy approaches that focus on specific
cognitive factors in pathological worry such as intolerance of uncertainty and metacog-
nitive beliefs produced better recovery rates than more standard cognitive therapy and
CBT approaches. According to Fisher (2006), the combination of cognitive therapy plus
applied relaxation may be more effective than either treatment alone. However, indi-