Abstract

Safety behaviour, which includes idiosyncratic strategies aimed at reducing anxiety and avoiding or averting perceived catastrophe, has been conceptualized in cognitive-behavioural models of anxiety disorders as having an anti-therapeutic, anxiety-maintaining function. There is substantial empirical support for its proposed negative effects in the context of exposure-based treatments for anxiety disorders. However, other hypotheses and findings suggest that incorporation of safety behaviour into such treatments might instead facilitate gains and/or enhance treatment acceptability. This research aimed to further understand the role of safety behaviour in the treatment of anxiety disorders. In Study 1, an experimental paradigm was used to examine the effect of ‘safety gear’ on belief change. Spider fearful participants were asked to evaluate the validity of a targeted negative belief about spiders during a brief session with a live tarantula while either using or not using protective gear during the session. The results demonstrated that negative beliefs declined equally and robustly in both groups after the session. Study 2 investigated treatment acceptability and preference as a function of safety behaviour use (judicious vs. discouraged) and treatment rationale (cognitive vs. extinction). Clinically anxious participants and undergraduate students provided acceptability ratings and preference ranks for four written vignettes describing a course of CBT for fear or anxiety. Treatment descriptions promoting judicious safety behaviour use received significantly higher acceptability ratings than those discouraging its use. Descriptions that presented a cognitive versus an extinction rationale were also rated as more acceptable. The most highly ranked treatment included judicious safety behaviour use under a cognitive rationale. The same pattern of results was observed in both participant groups. Study 3 involved the development and psychometric evaluation of a novel self-report measure of safety behaviour, the Safety Behaviour Inventory (SBI), using a large undergraduate student sample and a smaller sample of clinically anxious participants. Exploratory and confirmatory factor analyses and analyses of reliability and validity were conducted. Preliminary support was obtained for the discriminant and construct validity of a four-factor SBI. The combined results of these studies are discussed in terms of cognitive-behavioural theories and treatments of anxiety disorders, and future research directions are suggested.