Assessing a facilitated social network intervention for health outcomes in lonely and socially isolated people: the pragmatic, cluster-randomized PALS trial

Introduction: Loneliness has received attention in recent years as an important public health issue due to its associations with poorer mental and physical health. This study aimed to assess the effectiveness and cost-effectiveness of a facilitated social network intervention to alleviate loneliness and social isolation in community settings. Methods: A pragmatic, cluster-randomized controlled trial (RCT) was conducted to compare participants receiving the intervention with a wait-list control group. This trial also included an embedded economic evaluation and took place in two cities in England, UK. The intervention was a facilitated social network tool. It connects people to local community resources to potentially increasing their social involvement. First, participants mapped out and reflected on their personal social networks. Second, they completed a questionnaire to ascertain their preferred activities, interests, and support needs linked to a local resource database. The intervention was delivered by a trained facilitator, either in person or remotely. Community-based organizations (n = 44) identified adults at risk of social isolation and/or loneliness. The control group received” usual care” from the organization that recruited them. The primary outcome was the 6-month 12-Item Short Form (SF-12) Health Survey mental health composite score. Physical health, wellbeing, loneliness, social isolation, and collective efficacy were explored as secondary outcomes. Intervention costs, healthcare resource use, quality-adjusted life years (QALYs), and net monetary benefit (NMB) were included in the economic analysis. Results: A total of 469 adults were recruited between November 2018 and November 2021, with an 8-month in recruitment due to the coronavirus disease 2019 (COVID-19) in 2020 (242 participants in the intervention group and 227 participants in the control group). The results showed no clinically meaningful treatment effect of the intervention on the primary (0.21; 95% confidence interval [CI]: −1.74–2.16; p = 0.834) or secondary outcomes when compared to the usual care control group. The economic evaluation indicated no significant difference in QALYs and did not demonstrate cost-effectiveness despite being inexpensive to deliver. Discussion: The findings indicate that the intervention was not effective under trial conditions. Future interventions aimed at reducing loneliness and social isolation would likely benefit from a multi-step approach that includes tailored psychological, relational, and social components and considers the structural availability of community assets. Clinical trial registration: https://doi.org/10.1186/ISRCTN19193075, ISRCTN19193075.