408 TREATMENT OF SPECIFIC ANXIETY DISORDERS
GAD score significantly higher on the measure than nonclinical controls (Dugas et al.,
1997; Dugas, Gagnon, et al., 1998; Dugas, Gosselin, & Ladouceur, 2001; Freeston
et al., 1994). Furthermore, GAD patients scored significantly higher on the IUS than
individuals with panic disorder (Dugas et al., 2005), although Holaway, Heimberg, and
Coles (2006) found that intolerance of uncertainty was equally relevant to OCD in a
nonclinical analogue study. Although there is considerable empirical evidence for the
importance of intolerance of uncertainty beliefs in the pathological worry of GAD (see
review by Koerner & Dugas, 2006), it is doubtful that the construct is unique to GAD
(e.g., OCCWG, 2003; Tolin, Abramowitz, Brigidi, & Foa, 2003).
Various studies have investigated whether positive and negative metacognitive
beliefs about worry are unique cognitive features of GAD. Particular interest has
focused on the positive beliefs about worry because these schemas may be particularly
instrumental in initiating worry as an avoidant coping response to perceived threat
(Koerner & Dugas, 2006; Sibrava & Borkovec, 2006; Wells, 2004). Wells (2006) con-
siders the negative beliefs about worry a unique feature of pathological worry in GAD
because a focus on the uncontrollable and dangerous quality of worry leads to meta-
worry, or “worrying about worrying,” a process that is unique to GAD. Wells and col-
leagues developed the 65-item Meta- Cognitions Questionnaire (MCQ) to assess beliefs
about worry and unwanted intrusive thoughts, with one subscale that assesses positive
beliefs about worry and the other subscale negative beliefs about the uncontrollability
and dangers of thoughts (Cartwright- Hatton & Wells, 1997). Scores on the positive
and negative beliefs subscales have a significant relationship with measures of worry,
obsessional symptoms, and trait anxiety, although negative beliefs have a much stronger
association with worry than positive beliefs (Cartwright- Hatton & Wells, 1997; Wells
& Cartwright- Hatton, 2004; Wells & Papageorgiou, 1998a). Moreover, GAD samples
score significantly higher than nonclinical controls and other anxiety disorder groups
on negative beliefs (i.e., MCQ Uncontrollability and Danger subscale) but not on the
positive beliefs subscale (Cartwright- Hatton & Wells, 1997; Wells & Carter, 2001).
Other measures have been developed to assess beliefs about worry. Positive beliefs
about worry as a coping strategy for dealing with difficult situations distinguished clini-
cal and analogue GAD individuals from non-GAD controls and correlated with mea-
sures of anxiety and worry (Dugas, Gagnon, et al., 1998; Freeston et al., 1994). Borkovec
and Roemer (1995) found that analogue GAD and non-GAD students believed worry
(1) motivates them, (2) is effective problem solving, (3) prepares them for bad events, (4)
helps avoid or prevent bad outcomes, and (5) superstitiously makes something bad less
likely to happen. However, one belief, that worry helps distract from emotional top-
ics, was endorsed significantly more by the GAD students. Davey, Tallis, and Capuzzo
(1996), who assessed beliefs about the positive and negative consequences of worry with
their own Consequences of Worry Scale, found that negative but not positive beliefs
correlated with measures of worry, trait anxiety, and poor problem- solving confidence.
They conclude that positive beliefs about worry may be involved in constructive, task-
oriented worrying as well as in chronic pathological worry. Finally, negative but not
positive beliefs about worry predicted GAD severity in a Spanish community sample of
older adults (Montorio, Wetherell, & Nuevo, 2006).
In summary there is considerable empirical evidence that threat, personal vulner-
ability, intolerance of uncertainty, and metacognitive beliefs about worry are activated
in GAD. However, capturing the core fear content that is distinctive to GAD has proven
more elusive for researchers. We do not have specific measures of GAD threat and vul-