(PS5-A19) The Impact of Hope and Treatment Expectancies on Changes in Anxiety and Depression in Unified Protocol

Background
The beliefs and expectations patients bring to therapy may be important predictors of change in treatment. Better expectancies about treatment have been associated with better posttreatment outcomes in a range of therapies (Constantino et al., 2018). However, patients’ beliefs about their abilities to achieve their goals in therapy have received relatively less attention. According to Snyder (2002), hope is defined as the belief that one can derive pathways to desired goals and motivate oneself via agency thinking to use those pathways. These beliefs may be especially relevant to outcomes in cognitive-behavioral therapy (CBT) because CBT involves developing pathways to solve problems while cultivating a sense of agency about one’s abilities to follow these pathways. Although hope has been associated with decreases in anxiety (Gallagher et al., 2020) and depression (Hayes et al., 2007), it is unclear if hope and treatment expectancies predict subsequent changes in anxiety and depression during CBT.
 
Methods
We conducted a secondary data analysis of a sequential multiple assignment randomized trial of the Unified Protocol (UP; Barlow et al. 2018) in which patients with mixed anxiety and depressive disorders were randomized to receive 6 or 12 sessions of the UP. Patients (N = 70; Mage = 33.74; 67% female; 74% white) reported their level of hope (Trait Hope Scale; Snyder et al., 1991) at baseline and completed measures of treatment expectancies and credibility (Credibility and Expectancy Questionnaire; Devilly & Borkovec, 2000) immediately after session 1. Before each session, patients reported on the severity of anxiety (OASIS; Norman et al., 2006) and depression (ODSIS; Bentley et al., 2014) experienced in the past week. To maximize power, we focused on the first 6 sessions of treatment since all patients completed these sessions. We tested whether baseline treatment expectancies, credibility, and hope predicted individual differences in latent slopes of anxiety and depression using two multilevel structural equation models in Mplus Version 7.0 (Muthén & Muthén, 2008-2012). We estimated the latent person-specific intercepts and slopes of anxiety and depression across sessions 1-6 at Level-1. We then used these latent intercepts as well as latent scores representing treatment expectancies, credibility, hope-agency and hope-pathways to predict latent anxiety and depression slopes in separate models. Both models used full information maximum likelihood to account for missing data.
 
Results
Both models demonstrated adequate fit to the data: RMSEAs < .12, CFIs > .80, TLIs > .76. Higher treatment expectancies at baseline significantly predicted steeper decreases in anxiety, B = -.32, SE = .15, p = .03, and depression, B = -.39, SE = .18, p = .03. Similarly, greater hope (agency thinking) at baseline significantly predicted steeper decreases in anxiety, B = -.65, SE = .28, p = .02, and depression, B = -.72, SE = .35, p = .04. By contrast, neither treatment credibility nor hope (pathways thinking) significantly predicted changes in anxiety or depression, ps > .07.
 
Discussion
These results suggest that agency motivation and expectancies of change may be particularly important to assess and bolster at the start of CBT for anxiety and depression.