Panic Disorder 323
and rapid overbreathing, disorganized breathing patterns, and frequent sighing (see
Meuret et al., 2005; Meuret, Wilhelm, Ritz, & Roth, 2003; Salkovskis et al., 1986).
Individuals are trained in slow, deep abdominal breathing to eliminate hypocapnia and
its uncomfortable physical sensations thereby reducing the anxious state. Table 8.9
presents a typical diaphragmatic breathing retraining protocol.
Current Status of Breathing Retraining
There is currently considerable debate over the role of breathing retraining in CBT for
panic disorder. There are three reasons why cognitive- behavioral therapists are now
questioning the use of breathing retraining. The first is a very practical, clinical concern.
Like using other forms of relaxation, a person with panic disorder might use controlled
breathing as a safety response or coping strategy to escape from an anxious state (Antony
& McCabe, 2004; Salkovskis et al., 1996; White & Barlow, 2002). This, of course, would
undermine the effectiveness of cognitive therapy by reinforcing a fear of the anxiety and
the client’s faulty evaluation of the dangerousness of the physical sensations. If there is
any evidence that the client is using controlled breathing because of a fear of anxiety and
its symptoms, then the coping response should be discontinued immediately.
Second, the rationale for offering breathing retraining in panic disorder has been
called into question with evidence that hyperventilation and hypocapnia are often not
present even in panic attacks that occur in the natural setting (see review by Meuret et
al., 2005; Taylor, 2000). And third, the therapeutic effectiveness of breathing retrain-
ing has been questioned (e.g., Salkovskis, Clark, & Hackman, 1991; D. M. Clark et
al., 1999). Schmidt and colleagues conducted a dismantling study that compared the
effectiveness of 12 sessions of group- administered CBT plus breathing retraining, CBT
without breathing retraining, and a wait list condition at posttreatment and 12-month
follow-up (Schmidt, Woolaway- Bickel, et al., 2000). At posttreatment both active treat-
ments were significantly improved over the wait list condition but there was no statisti-
cally significant difference between the CBT and CBT + breathing retraining conditions.
At 12-month follow-up 57% of the CBT group met recovery criteria compared with
37% for the CBT + breathing retraining group. The authors concluded that the addi-
tion of diaphragmatic breathing does not add any therapeutic benefits to CBT for panic
beyond the standard treatment components of education, cognitive restructuring, and
exposure. They recommended that respiratory- control techniques be used only as a
behavioral experiment to provide corrective information for the catastrophic misinter-
pretation of bodily sensations and that therapists refrain from using them as an anxiety
management technique. Based on these findings we conclude that breathing retraining
should be considered optional in cognitive therapy for panic.
Clinician Guideline 8.17
Breathing retraining should be limited to individuals who clearly hyperventilate during a
panic attack. In most cases breathing retraining will not be necessary. If it is included in
the treatment protocol, careful monitoring is needed to ensure it does not become a safety-
seeking response.