Predictors and moderators of outcome of ICBT for loneliness with guidance or automated messages - A secondary analysis of a randomized controlled trial

Internet-based cognitive behavioral therapy (ICBT) is promising in alleviating loneliness in adults. Identifying individuals who benefit from ICBT for loneliness is pivotal to offering this intervention in a more targeted way and improving the intervention for those who do not benefit. This secondary analysis of a randomized controlled trial (RCT) aimed to identify predictors and moderators of outcome of an ICBT with guidance or automated messages for loneliness. In the RCT, 243 participants suffering from loneliness were randomly assigned to an ICBT with guidance (n = 98), automated messages (n = 97), or a waitlist-control condition (n = 48). In total, 180 participants completed the post-assessment (i.e., 10 weeks post-randomization). Outcomes were treatment outcome assessed with the UCLA-9 Loneliness Scale at post-assessment and treatment response, i.e., reliable improvement on the UCLA-9 from pre- to post. The relationship between a wide range of patient characteristics (grouped into socio-demographic, clinical, loneliness-specific, and treatment-related variables) and outcome was analyzed using multiple linear and logistic regressions. Feeling less burdened by loneliness resulted in higher odds of reliable improvement in guided ICBT compared to the waitlist-control condition. No treatment outcome or response moderators were identified for ICBT with automated messages compared to the waitlist-control group. Across active intervention groups, loneliness at baseline, age and fit between the tasks and goals of the intervention and participants' need predicted treatment outcome. Predictors of treatment response for ICBT with guidance and automated messages were not identified, and no variables differentially predicted the effects of ICBT with guidance or automated messages on the outcomes. In conclusion, individuals less burdened by their feelings of loneliness benefited more from guided ICBT. Lower baseline loneliness scores, younger age, and a better match between tasks and goals of the intervention and participants' needs also predicted a more favorable treatment outcome for both ICBT with guidance and automated messages.