Examining Moderators of Outcomes in a Smartphone-Delivered Interpretation Bias Intervention

HabitWorks (HW) is a personalized, transdiagnostic, Smartphone-based intervention designed to improve interpretation biases related to anxiety and depression. Using data from a randomized controlled trial evaluating the HW app’s effectiveness in comparison to a credible control condition, this secondary data analysis examined potential demographic and clinical moderators of intervention effects on post-intervention outcomes (i.e., anxiety, depression, functional impairment, and clinical global improvement). U.S. adults with anxiety and depression symptoms ( N =340; M age =33.04 years, 57.4% women, 60.6% White, 14.7% Black, 11.2% Asian, 10.9% Multiracial, 22.1% Hispanic/Latinx) were randomly assigned to either the HW condition (three weekly interpretation bias exercises and weekly symptom surveys) or the Symptom Tracking condition (ST; three weekly symptom surveys) for four weeks. Multiple linear regressions were used to test 10 potential moderators of condition (HW vs. ST) effects: endorsement rates of benign interpretations, endorsement rates of negative interpretations, depression symptoms, anxiety symptoms, age, gender, race, ethnicity, education, and sexual orientation. After controlling for baseline scores, benign interpretation endorsement moderated the effect of condition on three post-intervention outcomes (i.e., depression severity, anxiety severity, and clinical global improvement; only depression remained significant with heteroscedasticity-consistent robust standard errors), such that HW was superior to ST in participants who endorsed more benign interpretations at baseline. Response to HW appears to be similar across most clinical and demographic groups. Further studies examining the effect of dosage tailored to baseline IB in larger samples are needed to clarify moderator effects.