Efficacy of an Internet-based self-help intervention with human guidance or automated messages to alleviate loneliness: a three-armed randomized controlled trial

Loneliness is a prevalent and stigmatized phenomenon associated with adverse (mental) health outcomes. However, evidence-based interventions to alleviate loneliness are scarce. This randomized controlled trial (ClinicalTrials.gov-ID: NCT04655196) evaluated the efficacy of an internet-based cognitive behavioral self-help intervention (ICBT) to reduce loneliness by comparing two intervention groups with guidance or automated messages against a waitlist control group. Adults (N=243) suffering from loneliness were recruited from the general public and then randomly assigned (2:2:1) to a 10-week ICBT with human guidance (GU) or automated messages (AM) or to a waitlist control group (WL). Loneliness, assessed with the UCLA-9, was the primary outcome. Outcomes were assessed at baseline and 10 weeks (post) and analyzed using mixed-effects models. The pooled intervention conditions resulted in lower loneliness scores at post-assessment than the WL (Cohen’s d = 0.57, 95% CI [0.25; 0.89]) and reduced depressive symptoms, social anxiety, social avoidance behavior, and rejection sensitivity (d = 0.32–0.52). The GU group had lower loneliness scores at post-assessment than the AM group (d = 0.42, 95% CI [0.13; 0.70]). ICBT effectively alleviated loneliness, and guidance increased the reduction in loneliness compared to automated messages. Alleviating loneliness with ICBT further seems to reduce the overall burden of psychopathological symptoms.