Community Collaboration to Combat COVID-19

The target population will include English- and Spanish-speaking families residing in Baltimore City households (N=238,427). The sample size goal is 1,386 households (0.44% of households) or 3,000 individual household members. Households will be sampled using a multi-stage approach with 1) selection of 105 of 653 census block groups (CBGs) with probabilities proportional to the estimated number of occupied households selected from 9 strata of CBGs defined by socioeconomic status and race/ethnicity with over-sampling of CBGs with harder-to-reach populations (e.g., Latinos/x, low-income whites); 2) selection of residential addresses within each of the strata via non-linear optimization; and finally, 3) screening of individuals selected for eligibility (e.g., household occupied vs. not, English/Spanish speaking).

After completion of brief surveys, individuals will be asked to enroll other willing household members. One adult member of the household will be randomized 1:1:1 using a stratified, blocked approach with varying block sizes of 3, to one of three testing modalities: 1) fixed site standard of care testing; 2) community-based mobile van testing; or 3) self-collected home, based testing. Randomization will be stratified by geography (n=12 zones) and race/ethnicity (non-hispanic white/other, non-hispanic black and hispanic/latinx).

Arm 1 includes all three Johns Hopkins (JHMI) ambulatory outdoor testing sites across Baltimore City. Each site represents a traditional appointment-based scheduling system.

Arm 2 includes a convenient, accessible mobile van option which will be located in the center of each of 12 geographic zones.

Arm 3 includes a home-based testing kit which will be delivered by courier service.