Face Masks to Reduce SARS-CoV-2 Infections (COVID-19) in Bangladesh

Detailed Description: This study intends to answer the following research questions: Can mask distribution and mask promotion at homes, mosques, markets and other public areas successfully change community mask-wearing norms to increase mask-wearing, even years into a pandemic? Can community mask-wearing reduce SARS-CoV-2 infections (both symptomatic and asymptomatic)? Is increased prevalence of community-level mask-wearing associated with decreased physical distancing? To answer these questions, the investigators will conduct a cluster randomized trial which randomizes the proposed intervention at the village level in Bangladesh. Control communities will receive no intervention and treatment communities will receive an extensive intervention combining several elements. Study staff will distribute free face masks through door-to-door visits and at markets and mosques. In some communities groups of people will be encouraged to wear masks by appealing to their shared values and in other communities individuals will be encouraged to wear masks by discussing the biological value of masks (cross-randomized among intervention communities). Mask promoters in the community and at markets in treated villages will remind people about the importance of wearing masks and distribute additional masks. Markets will have signs that masks need to be worn to enter the market. Religious leaders will encourage mask use at religious services. The project will enroll 1200 communities, stratified in geographically proximate groups of 10; half will be randomized to receive the intervention. Communities will be surveilled through direct observation at 0, 2, and 4 weeks to assess mask-wearing. The investigators will assess symptomatic and asymptomatic SARS-CoV-2 by conducting household visits every 6-10 days to ask about symptoms of COVID-19 and collecting nasal swabs to test by PCR for the presence of SARS-CoV-2. End line is planned for 4 weeks after the intervention begins, but may be shifted according to seroprevalence trends.