In vivo exposure has been one of the strongest and most consistently demonstrated treatments for agoraphobic avoidance. In fact, it has often been demonstrated to be superior to placebo interventions as well as other credible psychological treatments – a none too easy achievement in psychological research. Furthermore, when anti-exposure instructions are included in comparison therapies, the strength of exposure becomes even more evident. Barlow’s (1988) conclusion that, following exposure alone, around 75% of agoraphobics (excluding dropouts) will evidence some clinical benefit is representative of most similar reviews. Yet, despite the strength of in vivo exposure in the treatment of agoraphobia, it is noteworthy that there remains little room for complacency. Around one-quarter of patients do not experience any clinical improvement during treatment, not all are completely symptom free at follow-up, and not all those who benefit from treatment maintain their gains. The research literature provides some details about the ingredients of a successful exposure package.
First, it appears that the greater the exposure exercises resemble the real situations avoided by individuals, the better is the outcome. For instance, in vivo exposure is usually superior to imaginal exposure. Second, the more frequently the person confronts their feared situation and the greater the duration of exposure sessions, the higher is the proportion of treatment completers who reach a high end-state functioning. Finally, it is probably the case that exposure that continues until anxiety has subsided is preferable to shorter exposures. For instance, Rayment and Richards (1998) found that allowing panic symptoms to develop and pass predicted less avoidance of phobic situations. However, while among specific phobics continuing exposure until anxiety decreased was superior to “escaping” when anxiety had risen, among agoraphobics there are some data suggesting that both approaches are equally effective.
One additional factor that may enhance exposure programs is an effort to assure that exposure brings about cognitive changes. Since certain cognitive variables predict the degree of avoidance, exposure should also aim to modify the cognitions predictive of the avoidance. That is to say, it is important to assure that exposure treatments include instruction regarding the true causes of panic attacks and be combined with training in anxiety-management strategies. That said, it is apparent that the role that the cognitive variables identified in cognitive theories play in treatment is not completely clear. For example Soechting et al. (1998) compared the efficacy of a cognitive and a behavioral rationale for exposure treatment among people with panic disorder and agoraphobia, and there was no overall superior efficacy of the cognitive rationale. However, Hoffart (1998) did find that agoraphobics given cognitive therapy showed superior outcomes to those who received guided mastery and that a path analytic strategy implicated the variables identified by cognitive models. In particular, it appears that changes in the situational fear of agoraphobics is particularly mediated by changes in self-efficacy, as opposed to changes in catastrophic beliefs and perceived thought control.
A further way in which the efficacy of exposure-based programs has been improved is with the involvement of spouses or partners in treatment. While a complete discussion of all the findings regarding marital and family interventions in agoraphobia is beyond the scope of the present book, a summary statement is in order. Barlow et al. (1984) found that among agoraphobics with apparently well-adjusted relationships, the inclusion of the partner in treatment added little to improvement rates. In contrast, among individuals with poorly adjusted relationships, the inclusion of a partner enhanced treatment outcome, overriding any effects of the poor interactions. Therefore it is useful advice to include partners in exposure-based programs, particularly where there is concern about the negative impact the relationship may have on treatment. Behavioral therapy for couples may be a useful adjunct, but concurrent therapy raises a number of other issues, including the difficulties of combining couple-focused interventions and agoraphobic treatment. Thus our practice is typically to sequence the treatment for agoraphobia and the couples therapy, deciding on a case-by-case basis the order for particular clients.
Finally, one way that in vivo exposure programs have been improved is by the addition of more and different exposure. Fava et al. (1997) reported data that suggested that additional exposure was beneficial for patients who were not initially responding to exposure. More novel treatments, e.g., “interoceptive exposure” have been used to target panic attacks directly. People prone to panic attacks are requested to engage in a series of exercises that produce sensations similar to those that occur during a panic attack. For instance, individuals who fear that tachycardia may signal a heart attack are requested to run on the spot. The exercise produces a rapid heartbeat, which can then be included in the person’s graded exposure hierarchy. We have demonstrated that among agoraphobics the exercises tend to improve overall treatment outcome and Craske et al. (1997) found that the addition of interoceptive exposure reduced panic frequency posttreatment and at a 6-month follow-up. The Craske et al. study is noteworthy because it also found treatment specificity. That is to say, the interoceptive exposure seemed to exert particular effects upon measures of panic frequency, suggesting that this is a treatment to be stressed with patients who have frequent panic attacks.
In summary, in vivo exposure is the treatment of choice for agoraphobic avoidance. Alone, exposure is a powerful treatment but it is not able to assure a high end-state function in all patients. Therefore, additional treatments need to be considered and combined with exposure. In fact Murphy et al. (1998) found that, while exposure was an important component in treatment, anxiety management added to the overall improvement. Specifically, they argued that once a critical threshold of exposure practice has been achieved, anxiety-management techniques became increasingly important. Therefore, another possible addition to treatment could involve strategies to control the emergence of panic attacks. One such intervention is breathing retraining.
Hyperventilation can produce symptoms similar to those experienced during a panic attack and hyperventilation often occurs during a panic attack. Therefore, it is reasonable to suppose that teaching individuals appropriate breathing strategies aimed to control hyperventilation will alleviate panic symptoms and assist long-term symptom management. An analysis of the treatment literature is complicated by the variety of breathing techniques being taught. For present purposes, attention will be limited to studies that compared breathing retraining with exposure-based programs. Although breathing retraining alone can reduce the frequency of panic attacks, given that exposure-based programs have a demonstrated efficacy, the clinically relevant question is “Can breathing retraining enhance the efficacy of standard exposure treatments?”. Three studies meet these criteria. One of the studies failed to demonstrate a decrease in panic frequency following breathing retraining and cognitive restructuring. This result is surprising, since such a treatment should decrease the frequency of panic attacks. Given the study’s failure to demonstrate a treatment effect, it is difficult to interpret the comparisons with exposure treatments. The remaining two studies failed to find any posttreatment differences between exposure-based programs with and without breathing retraining, leading to the conclusion that, while breathing retraining may reduce panic attacks, it does not improve exposure-based programs in the short term. However, one study examined the patients at a 6-month follow-up point and found that the addition of breathing retraining to exposure enhanced treatment outcome. To summarize, when used alone, breathing retraining can reduce symptoms, but, more importantly, when added to a comprehensive cognitive behavioral exposure-based program, it can improve long-term outcome, perhaps by consolidating the progress made during treatment and enhancing the stability of treatment gains.
Another strategy that has been used in the treatment of panic disorder and agoraphobia is relaxation. When the different forms of relaxation are combined, relaxation alone has been estimated to bring about a clinically significant improvement in around 47% of patients with panic disorder and agoraphobia. However, in the context of panic disorder, applied relaxation has been found to be superior to progressive muscle relaxation. Applied relaxation is a rapid form of relaxation that enables individuals to elicit the relaxation response quickly when needed. While effective, Arntz and van den Hout (1996) found that applied relaxation was less effective than cognitive therapy, particularly on measures of panic frequency. There have been some suggestions that progressive muscle relaxation may detract from a cognitive behavioral therapy program for agoraphobia. However, the design of the study involved the introduction of relaxation as one of the first components into a cognitive behavioral therapy program. If it is assumed that most individuals with a chronic anxiety disorder will have been taught relaxation skills at some point, the higher dropout rates observed in this condition could have occurred because the patients became disillusioned with receiving a treatment they had already attempted. While further research is needed to address this question, it maybe speculated that it may be more helpful to include relaxation later, rather than earlier, in a comprehensive cognitive behavioral therapy program.
Cognitive restructuring and combined cognitive behavioral therapy packages
One of the most exciting developments in recent times has been the addition of cognitive techniques to exposure programs. Most often, exposure is targeted at the avoidance behavior, while cognitive interventions are focused at decatastrophiz-ing the interpretations of the panic symptoms. Typically, cognitive approaches are not used alone (although they are somewhat efficacious when used as the sole treatment intervention), but are combined with exposure (both to external and internal triggers of panic), relaxation, and breathing retraining. When such combinations are used, the improvement is considerable. While it is not always clear that the addition of cognitive therapy enhances the effects of exposure, the combined packages are very effective. Barlow et al. (1989), Klosko et al. (1990), and Beck et al. (1992) similarly found that up to 90% of panic-disordered patients were panic free following their combined panic control treatment. Furthermore, it is becoming clear that the effects of these treatment packages have a broader impact on the clients, bringing about an overall improvement in quality of life.
In addition, it is becoming increasingly clear that these treatment packages are effective not only in clinical research centers. The effects transport well to self-help delivery format and can be maintained when face-to-face treatment is reduced in duration but compensated for by a computer program that incorporates the basic principles of cognitive behavioral treatment. The treatment packages can generalize favorably to adolescent samples and to settings that are representative of community treatment, even when the patients are more severe than those typically seen in research trials. When used in a community mental health center, 87% of panic-disordered patients were panic free after 15 sessions of cognitive behavioral therapy; a figure that compares well with that obtained in clinical research settings. The patients also showed reductions in anticipatory anxiety, agoraphobic avoidance, generalized anxiety, and depression; a pattern of data that once again would be expected on the basis of research conducted in clinical research settings.
Manualized cognitive behavioral therapy-based treatment packages have also been criticized as being inappropriate for clients with comorbid disorders. Marchand et al. (1998) examined the rates of change in clients suffering from panic disorder and agoraphobia with and without a comorbid personality disorder and found that both groups responded to the treatment for the anxiety disorder; however, clients with a personality disorder responded more slowly. In addition, Hoffart and Hedley (1997) noted that dependent personality traits appear particularly detrimental to treatment progress, but that cognitive behavioral therapy for panic disorder and agoraphobia did reduce symptoms of personality disorder (especially avoidant and dependent traits). Thus it seems more reasonable to conclude that clients with comorbid personality disorders may require longer, rather than inherently different, treatment than those without personality disorders.