Abstract
Loneliness is a recognized public health risk factor associated with increased morbidity and mortality. However, the effectiveness of interventions targeting loneliness remains unclear-particularly in relation to baseline severity. This systematic review and meta-analysis assessed intervention effectiveness and the influence of baseline severity and intervention characteristics. A total of 25 studies were included, of which 16 randomized controlled trials (RCTs; k = 21) were meta-analyzed. Interventions produced a moderate pooled effect at post-intervention (Hedge's g = 0.65, 95% CI [0.05, 1.26], p = 0.037), though with high heterogeneity. Sensitivity analyses confirmed a moderate effect (g = 0.55, 95% CI [0.22, 0.88], p = 0.003). Higher baseline loneliness predicted greater intervention effects (b = 0.04, 95% CI [0.02, 0.07], Z = 3.36, p < 0.001), with cognitive-behavioral therapy (CBT) showing the largest effect size (g = 0.73). No significant effects were observed at follow-up. These findings underscore the need for dual strategies: targeted psychological interventions (e.g., CBT) for individuals with severe loneliness, and universal, context-based approaches for the broader population. This aligns with Geoffrey Rose's distinction between individual-level treatment and population-level prevention and highlights the urgency of embedding loneliness interventions into public health frameworks and policy agendas focused on promoting social connectedness and equity.