A language-matching model to improve equity and efficiency of COVID-19 contact tracing

In this article researchers in Santa Clara, California, developed and tested a computer algorithm to predict language needs when performing contract tracing in the community. Census data and mortgage data was used to build the model, which was evaluated using a randomized control trial design, and surveys were administered to the contract tracing staff to learn about their experiences with language discordance. Results indicate that the intervention decreased the amount of time between when the case is opened and when the interview with the individual is conducted.