Review 3: Community engagement for health via coalitions, collaborations and partnerships (on-line social media and social networks) – a systematic review and meta-analysis.
EPPI-Centre, Social Science Research Unit, UCL Institute of Education, University College London (UCL): London, UK.
This report describes the methods and findings of a systematic review on community engagement (CE) for health via online social media and social networks. It is the third and final review of a programme of work on the use and effectiveness of CE in interventions that target health outcomes. Social networks are one of many forms of CE. Our first two reviews suggested that the extent and particular processes of CE may be linked to effects on people’s health. The emergence of online, electronic peer-to-peer social network sites (e.g. Facebook) and online social media tools (e.g. Twitter) have increased exponentially in recent years, and existing evidence on their effectiveness is ambiguous.
We aim to evaluate the effectiveness of online social media/social networks on: the extent of CE across designs, delivery and evaluation; the types of health issues and populations that have been studied; their effectiveness in improving health and wellbeing and reducing health inequalities; and any particular features that account for heterogeneity in effect size estimates across studies.
Systematic review methods were applied to comprehensively locate and assess the available research evidence. The search strategy employed previously run searches used for Reviews 1 and 2 of this project (described elsewhere). The included studies were descriptively analysed and the findings were synthesised using three components: framework synthesis, meta-analysis and qualitative component analysis (QCA).
A total of 11 studies were included in the review, none of which was set in the UK. The community was not explicitly involved in identifying the health need for any of the 11 studies. No studies demonstrated a high level of CE, where participants were involved in the three measured elements: design, delivery and evaluation. Framework analysis indicated that peer delivery of the intervention was the predominant type of CE. Two processes of CE were reported – bidirectional communication and the use of facilitators – but none of the studies evaluated these processes. Professional facilitators were used more often in healthy eating/physical activity studies. Peer facilitators were used more often in youth-focused interventions and professional facilitators were utilised more frequently for interventions targeting older populations. Studies focusing on women only may incorporate peer or professional facilitators to aid intervention delivery. Peer or professional facilitators were used slightly more consistently in interventions targeting minority ethnic groups. Meta-analyses and meta-regression showed no evidence of beneficial effects on any outcomes. There was moderate (I 2 = 25≤50) to high (I 2 = ≥50) heterogeneity between studies for primary outcomes, suggesting the existence of potential moderators. None of the tested study characteristics explained the variation in effect sizes. The QCA demonstrated that including a facilitator in online social media/social networking interventions showed higher effect sizes for studies that focused on topics other than healthy eating and physical activity.
The results from this study suggest that CE is not utilised across the design or evaluation of health interventions, and the type of CE undertaken with intervention delivery focuses on peer interactions alone. This suggests that there is very little co-creation of knowledge or building of social capital occurring in evaluated health intervention studies using online social media/networking.
Archive Staff Only