(PS11-76) Factors Influencing Remission Following Transdiagnostic Digital CBT for Emotional Disorders

Digital Mental Health Interventions (DMHIs) decrease symptoms of mental health conditions (Lattie et al., 2019). However, certain hurdles to DMHI engagement exist, including age (Borghouts et al., 2021) as it pertains to digital literacy, education level (Wu et al., 2022) as it pertains to content comprehension, sex (Borghouts et al., 2021) and baseline symptom severity (Karyotaki et al., 2021) as they pertain to willingness to engage in DMHIs. Relationships between these factors and symptom remission have not been studied in the context of a digital, cognitive-behavioral intervention for emotional disorders. This study aims to evaluate relationships between our focal variables and remission from the clinical threshold in a sample of treatment-seeking adults. Participants (n = 120, M(age)= 34.03, 71.7% White, 73.3% Female) were from a clinical trial comparing a digital version of the Unified Protocol (iUP; Barlow et al., 2018) to its enhanced counterpart: the iUP+, additionally targeting positive affect. 

Given our research questions, we collapsed the iUP/iUP+ treatment groups. Analyses were conducted in R v. 4.3.2. Sex was coded as Female=1, Male=2, and Intersex=3. Consistent with prior research (Oh et al., 2021), age was grouped: 18-45=1 and 46-75=2. Consistent with prior research (Wu et al., 2022), education level was grouped: Some High School or Less, High School Diploma, Some College, Trade School Certification=1; Associate’s Degree, Bachelor's Degree, Master's Degree, Professional or Doctoral Degree=2. Remission status was determined based on DSM-5 criteria using the Anxiety Disorders Interview Schedule for DSM-5 (Brown & Barlow, 2014). 

We conducted binary logistic regressions to determine the effect of sex (model 1), age (model 2), level of education (model 3), or baseline clinical severity (model 4) on the likelihood that participants would be in remission at the end of treatment. For models 1-3, the chi-squared test of model fit was not statistically significant, meaning that we could not conclude that these variables had a statistically significant effect on post-treatment remission status. The chi-squared test for Model 4 was statistically significant (𝜒2 (91) = 14.21, p < .01), and the odds of being in remission at the end of treatment were 0.24 (95% CI [0.11 to 0.55]) times lower for each one unit increase in baseline severity, meaning that higher baseline severity was associated with lower likelihood of remission (b = -1.42, p < .01).

Demographic factors were not associated with remission following treatment with the iUP/iUP+, while higher baseline symptom severity was associated with lower odds of remission. This finding is consistent with literature highlighting that baseline symptom severity is linked to lower engagement with DMHI, which in turn, affects the likelihood of remission. An alternative interpretation of these findings is that greater clinical severity necessitates a more substantial degree of change to transition below the clinical threshold. Future interventions could identify individuals with high baseline clinical severity and, more explicitly, target engagement amongst these individuals.