What works in interventions targeting loneliness: a systematic review of intervention characteristics

BACKGROUND: Loneliness has been linked to negative health and economic outcomes across the life course. Health effects span both physical and mental health outcomes, including negative health behaviours, lower well-being, and increased mortality. Loneliness is however preventable with effective intervention. This systematic review aims to identify what has worked in interventions for loneliness to guide the development of future interventions. METHODS: Eight electronic databases (Medline, Embase, PsycINFO, Social Policy and Practice, Social Sciences Citation Index, Epistemonikos, CINAHL, Cochrane Library) were systematically searched from inception to February 2022 using terms for intervention and loneliness to identify relevant interventions in the general population. No restrictions on age, socio-economic status, or geographic location were imposed. Studies were to measure loneliness as the primary outcome through a validated scale or single-item question. Case studies were excluded. Additional studies were identified through citation chasing. Extracted data included study and intervention characteristics, and intervention effectiveness for cross-study comparison. Critical appraisal was conducted using the Joanna Briggs Institute and Critical Appraisal Skills Programme tools before the studies were summarised in a narrative synthesis. RESULTS: Searches identified 4,734 hits, from which 22 studies were included in this review. Of these studies, 14 were effective in reducing loneliness. Additionally, five studies presented unclear findings, and three concluded no decrease in loneliness. Interventions varied between group vs. individual format, online vs. in person delivery, and regarding both intervention duration and individual session length. Furthermore, this review highlighted five key areas when considering designing an intervention for loneliness: use of between session interaction, inclusion of clear learning mechanisms, role of active participation, number of opportunities for group or facilitator interaction, and variation in teaching and learning styles. CONCLUSIONS: Group sessions seem preferred to individual formats, and interaction through active participation and group or facilitator contact appear beneficial, however studies also recognised the importance of a person-tailored approach to delivery. Studies suggest there is no ‘quick fix’ to loneliness, but that learnt practices, behaviours, and community connection should be built into one’s lifestyle to achieve sustained intervention effectiveness. Future interventions should consider longer follow-up periods, male and populations with lower educational levels. SUPPLEMENTARY INFORMATION: The online version contains supplementary material available at 10.1186/s12889-023-17097-2.