(PS2-21) Expecting Change: How Treatment Expectancy May Predict Decreases in Anxiety Sensitivity

Background: There is evidence to implicate anxiety sensitivity as a mechanism of change for behavioral and cognitive treatments of anxiety and anxiety related disorders. To aid the development of treatment algorithms, researchers are striving to identify predictors of engagement of the therapeutic target. One possible predictor for the change in anxiety sensitivity may be patient expectancy, or the degree to which a patient expects improvement in their symptoms based on the information they have about the treatment plan. Method: The present study attempts to address this question using a secondary analysis of a randomized clinical trial. This trial aimed to examine the efficacy of D-cycloserine (DCS) for enhancing the effects of cognitive behavioral therapy targeting anxiety sensitivity (AS) reduction in adult smokers (N = 53) also receiving smoking cessation treatment. Participants were randomized to a 7-week integrated treatment and received either DCS or placebo prior to sessions 3-5. Treatment credibility, as assessed by the Credibility Expectancy Questionnaire (CEQ), was gathered after the first treatment session, and AS, as assessed by the Anxiety Sensitivity Index (ASI-3), was measured at baseline and all treatment sessions. We predicted that higher CEQ scores at week 1 would be predictive of larger decreases in Anxiety Sensitivity at end of treatment (session 7) and at follow-up (6-months). A two phase piecewise growth curve model was used to test the predictive value of the CEQ on AS scores. Results: CEQ did not interact with group assignment to predict ASI-3 scores. CEQ scores did predict ASI-3 scores at the end of treatment, such that participants with high levels of CEQ (1 SD above the mean) had lower ASI-3 scores at the end of treatment than participants with low CEQ (1 SD below the mean, b=2.68, t(323)=2.88,p=004). However, during the follow-up period, participants with higher CEQ scores worsened significantly more as compared to those with lower CEQ (b=.11, t(323)=3.44, p< .001) such that, at the 6 month follow-up, ASI-3 did not vary based on session 1 CEQ scores.

Conclusion: These findings provide evidence to support the hypothesis that treatment expectancy is important to engage mechanisms of action, at least during the acute phase. Limitations with these findings include the use of a non-transdiagnostic sample, the lack of diversity in the sample (96% white), and that only the first four questions in the CEQ were included in the data analysis, allowing us to only make claims on expectancy and not credibility. Clinical and research implications will be discussed.