Abstract
Social network diffusion interventions are a promising strategy for promoting health behavior change at the community level. Individuals impacted by the criminal legal system (CLS) experienced disproportionately high morbidity and mortality during the COVID-19 pandemic. As part of the National Institutes of Health Rapid Acceleration of Diagnostics-Underserved Populations (RADX-UP) initiative, we evaluated a group-level social network diffusion intervention incorporating motivational interviewing strategies to increase COVID-19 testing and vaccination among CLS-impacted individuals and their social networks across four urban cities. Between January 2023 and February 2024, participants were randomized by cohort to either a Motivational Interviewing-based intervention (MI) or a Prevention Education (PE) comparison condition. The MI intervention incorporated motivational interviewing-informed strategies to support participants in engaging and influencing members of their social networks, whereas PE participants received COVID-19 prevention education. The primary outcome was uptake of COVID-19 testing or vaccination within 90 days among previously unvaccinated primary participants, with testing and vaccination outcomes for social network members assessed based on participant report. Secondary outcomes included COVID-19 testing and vaccination knowledge among primary participants and reported resource sharing within social networks. A total of 810 primary participants were enrolled (MI: n = 403; PE: n = 407), with most identifying as Black or African American (89.8%) and 9.5% as Hispanic. At baseline, 83% of primary participants had previously been tested for COVID-19, and 70% had received at least one vaccine dose. At the 90-day follow-up, no statistically significant differences were observed between MI and PE groups in COVID-19 testing or vaccination uptake among primary participants or their network members. However, primary participants in the MI condition demonstrated significantly higher mean COVID-19 testing and vaccination knowledge scores compared with those in the PE condition. Patterns of resource sharing differed across study arms. Relative to MI, PE participants were more likely to share COVID-19 testing and vaccination resources at 30 days (OR = 0.54, 95% CI: 0.39-0.74, p < 0.01), though this difference was not sustained at 90 days (OR = 0.89, 95% CI: 0.58-1.35, p = 0.576). While the intervention did not produce significant differences in COVID-19 testing or vaccination uptake, both approaches facilitated dissemination of COVID-19-related resources and improved knowledge within CLS-impacted communities. These findings highlight both the potential and limitations of social network-based approaches for translating increased awareness into sustained preventive health behaviors.