Evolution of a coalition network during a whole-of-community intervention to prevent early childhood obesity

Ariella Korn, Ross Hammond, Erin Hennessy, Mark Pachucki, Aviva Must, Christina Economos

Contact: ariella.korn@tufts.edu

Background: “Whole-of-community” interventions hold promise in addressing childhood obesity—a holistic approach designed to target multiple weight-related behaviors at varying levels of influence (e.g., families, organizations, policy) across sectors and settings in a community. Several whole-of-community trials have engaged groups of community stakeholders, such as coalitions, throughout stages of intervention design and delivery. However, little is known about the structure of these coalition networks, how they change over time, and how they might influence the implementation of childhood obesity prevention strategies and information diffusion. Methods: We administered five semi-annual web-based social network surveys during the “Shape Up Under 5” pilot whole-of-community early childhood obesity prevention intervention in Somerville, Massachusetts, USA between 2015 and 2017. Survey participants included 16 multisector coalition members who met monthly with the research team during the study period, and their first-degree alters. Participants nominated up to 20 discussion partners and reported each tie’s interaction frequency and perceived influence related to involvement with early childhood obesity prevention efforts. Network characteristics and structure were assessed with visualization, descriptive analysis, and exponential random graph models. Results: The Shape Up Under 5 network included 16 coalition members, 201 first-degree alters, and 341 second-degree alters (nominated, but not sampled) across the five survey waves. Respondents nominated a median of five (IQR: 3-9) discussion partners per survey. Network members represented community-based organizations, parents, healthcare, nutrition services, local government, schools, childcare, and universities. Network size and membership varied across survey waves. We observed the largest network (256 stakeholders and 414 ties) at 6 months as intervention planning gained momentum, and the largest proportion of stakeholders communicating frequently (daily or weekly) about early childhood obesity prevention at 12 and 18 months during the “peak” intervention period. Networks were sparsely interconnected over time (1-2% of possible ties observed) and were most and least centralized at baseline and 24-month follow-up, respectively. Ties were increasingly perceived as influential over time as well as increasingly siloed within stakeholders’ community affiliation groups. Discussion & Conclusions: Understanding community stakeholders’ social networks may help optimize whole-of-community childhood obesity prevention efforts. The Shape Up Under 5 network’s extensive and evolving membership may indicate stakeholders’ access to a wide range of resources and ideas, and their ability to broadly disseminate intervention messages. The attenuation of network hierarchy may have supported more equal participation and control over intervention efforts over time. Future research is required to assess the generalizability of observed network patterns in other community coalition networks addressing childhood obesity, network influences on implementation outcomes like program adoption and sustainability, how various levels of homophily within community groups affect intervention coordination, and if and how to intervene with the stakeholder networks.

← Schedule