Pre-procedural testing improves estimated COVID-19 prevalence and trends

Background: COVID-19 positivity rates reported to the public may provide a distorted view of community trends because they tend to be inflated by high-risk groups, such as symptomatic patients and individuals with known exposures to COVID-19. This positive bias within high-risk groups has also varied over time, depending on testing capability and indications for being tested. In contrast, throughout the pandemic, routine COVID-19 screening tests for elective procedures and operations unrelated to COVID-19 risk have been administered by medical facilities to reduce transmission to medical staffing and other patients. We propose the use of these pre-procedural COVID-19 patient datasets to reduce biases in community trends and better understand local prevalence. Methods: Using patient data from the Maui Medical Group clinic, we analyzed 12,640 COVID-19 test results from May 1, 2020 to March 15, 2021, divided into two time periods corresponding with Maui's outbreak. Results: Mean positivity rates were 0.1% for the pre-procedural group, 3.9% for the symptomatic group, 4.2% for the exposed group, and 2.0% for the total study population. Post-outbreak, the mean positivity rate of the pre-procedural group was significantly lower than the aggregate group (all other clinic groups combined). The positivity rates of both pre-procedural and aggregate groups increased over the study period, although the pre-procedural group showed a smaller rise in rate. Conclusions: Pre-procedural groups may produce different trends compared to high-risk groups and are sufficiently robust to detect small changes in positivity rates. Considered in conjunction with high-risk groups, pre-procedural marker groups used to monitor understudied, low-risk subsets of a community may improve our understanding of community COVID-19 prevalence and trends.