Abstract
Morgan and Raffle highlight some of the controversies and misunderstandings regarding the use of behaviour therapy and cognitive therapy for anxiety disorders [1]. The paper begins with a statement that Patients with social phobia are exposed to social situations daily, but often do not experience a reduction in anxiety. ‘Such a reduction would be predicted by learning theory models’. This is misleading and incorrect. In fact, the majority of patients with anxiety disorders are exposed at some time to their feared object, situation or stimulus and do not habituate because they withdraw or escape from the situation. The essential requirement for habituation within a session or episode is that the exposure is prolonged. We know that intermittent and brief exposure leads to sensitisation (a response increase) where as prolonged exposure leads to habituation (a response decrease). In learning theory terms, sensitisation is explained by operant conditioning, that is, the anxiety response to a given stimulus is reinforced by avoidance or escape (negative reinforcement). One would therefore expect that, for an individual with social phobia, brief social exposure would sensitise or worsen the anxiety response, or at least maintain the level of anxiety [2].
Similarly, the idea of safety behaviours is not new, and, as the authors state, includes ‘general avoidance’. Although concepts involving ‘safety behaviours’ may fit both cognitive and behavioural theories, the therapy recommended is behavioural: confronting the feared stimuli or situation and eliminating the safety behaviour. The eliciting of the wide range of activities (behavioural and cognitive) which an individual uses to reduce anxiety is part of a standard behavioural assessment including the behavioural analysis, which elicits this information in detail. This is where the difference between behaviour therapy and cognitive therapy lies. In behaviour therapy all of this information is used to construct an exposure program. Cognitive restructuring is not used. To be successful, the individual needs to expose themselves to feared situations or stimuli at the same time as eliminating the range of strategies, including distraction, used to reduce or avoid anxiety. The exposure needs to be graded so that either the situation is not overwhelming or that the avoidance strategies are eliminated sequentially. Without being aware of the subtle avoidances individuals use, exposure programs would be unsuccessful. From systematic reviews of the literature, there is no evidence that adding cognitive restructuring to well-conducted exposure programs adds to the benefit of exposure alone in social phobia [3].
The paper goes on to highlight this observation in its results by showing that the USUSAL group did not improve from baseline [1, Table 3], despite 80 h of group and individual therapy. This is at odds with the statement in the discussion that ‘the USUSAL and SAFETY groups made significant improvement in both the specific measures for social phobia and the general measures of anxiety following cognitive-behavioural therapy group treatment’. This statement seems to be based on the method of analysis used. The bulk of the analysis is based on the use of effect sizes. The authors justify the overall claim for improvement in both groups by doing within-subject comparisons using averaging across both groups. The usual way to measure difference between groups across time is to use analysis of variance in which the real contribution of group membership and time could be measured in a single analysis. It appears that the authors have made substantial claims for overall improvement based on the improvement in the SAFETY group.
The implication of this is to question whether the elimination of safety behaviours alone or the interaction of this and some element(s) of the USUSAL program produced the improvement in the safety group. Some clues can be obtained from the literature that exposure alone produces improvement in social phobia and that this improvement can be obtained in individual sessions of much shorter duration than 80 h. A principal aim of the development of behavioural and cognitive therapies for anxiety disorders over the last 30 years has been to determine which are the essential elements of therapy so that therapy is delivered with the maximum effect in the least time for both clinician and patient. Exposure in its many forms (including elimination of safety behaviours) has consistently emerged as the essential ingredient of therapy. Applied individually, a significant reduction of time per patient from the 80 h of the USUSAL program would be expected.