Abstract
PURPOSE: Public health surveillance systems require high-quality data to represent the population. In the absence of comprehensive or random testing throughout the COVID-19 pandemic, we have developed a proxy method for synthetic random sampling to estimate the actual community-level viral incidence, based on viral testing of patients who are asymptomatic and present for elective procedures within a hospital system. METHODS: The approach collects routine testing data on SARS-CoV-2 exposure among outpatients and performs statistical adjustments of sample representation using multilevel regression and poststratification (MRP), a procedure that adjusts for nonrepresentativeness of the sample and yields stable small group estimates. We extend MRP to accommodate time-varying data and granular geography. RESULTS: We have developed an open-source, user-friendly MRP interface for public implementation of the Bayesian analysis workflow. We illustrate the MRP interface with an application to track community-level COVID-19 viral transmission in Michigan. We present the estimated infection rate over time for the targeted population and across demographic and geographic subpopulations. CONCLUSION: The interface provides timely, substantive insights into population health trends and serves as a valuable surveillance tool for future epidemic preparedness. Beyond monitoring COVID-19, the MRP interface can analyze a wide range of health and social science data, making it broadly applicable to diverse research areas with reproducibility and scientific rigor.