Investigating SARS-CoV-2 Test Positivity Calculations Across US Jurisdictions

Abstract

Purpose: Throughout the COVID-19 pandemic, many US epidemiologists and policymakers turned to an indicator called test positivity, or the percent of tests coming back positive for SARS-CoV-2, to contextualize COVID-19 case counts with testing volume. But the nation's patchworked health data infrastructure, composed of 56 systems managed by each state and territory, complicated efforts to calculate the metric in a comparable way across US jurisdictions. We set out to map jurisdictional reporting differences in test positivity and investigate whether they interfered with its effectiveness and comparability as an indicator. Understanding these differences is important because jurisdictional test positivity informed consequential policy and individuals’ understanding of risk in their communities. Methods & Materials: We surveyed the health department websites of all US states and territories to examine how these jurisdictions were presenting test positivity on COVID-19 dashboards. When details about definitions were unavailable on jurisdictional websites, we reached out to jurisdictional public health officials for clarification. We also scored jurisdictions' presentations against best practices we identified for calculating the metric. Results: Among the 48 states and territories posting test positivity values, we observed no consensus on how to calculate the metric—jurisdictions used different units, test types, averaging techniques, and dating schemes. By looking at data for jurisdictions that posted multiple test positivity metrics, we observed that these definitional differences could result in variations from 31% to 300%. Only four states were following all ten of the best practices for reporting test positivity. Conclusion: The sheer number of ways states and territories define test positivity is alarming, given how much the indicator influenced US COVID-19 policy. Based on our survey, we believe the confidence of regulators in the precision and national comparability of test positivity is misplaced: The metric's value reflects state and territorial reporting decisions as much as actual viral prevalence. These findings underscore the need to invest in centralized public health infrastructure and create national reporting standards to improve unity of state reporting.