Deciphering U.S. COVID Test Positivity

The Johns Hopkins Coronavirus Resource Center (CRC) launched the COVID-19 Testing Insights Initiative in April 2020 to help decipher the different approaches U.S. states and territories use to calculate test positivity in the absence of federal standards. Below, the CRC explains the different calculations that its experts have identified and then ranked from the most insightful method to the least insightful. No single calculation provides the full picture of testing capacity and infection surveillance, but the ideal methods use numerators and denominators from the same data categories.

The first five charts below display the test positivity each state would register if they possessed the data required for the calculations of each approach. Exact positivity calculations can be viewed by hovering over each bar in the charts. States without bars lack the necessary variables to register test positivity calculations.

The last chart labeled “Hierarchy” presents the most accurate test positivity for each U.S. jurisdiction based on which of the five approaches provides an optimal calculation, according to Johns Hopkins experts. If the best method (Approach 1) cannot be applied using a state’s available data, the CRC applies the next best formula until one aligns with a jurisdiction’s existing variables.

Absent federal standards for reporting COVID-19 testing data, a national apples-to-apples view is not possible. Instead of uniformly applying the best approach to all states, the team of data and testing experts decided it is best to present the full picture, no matter how complex, so that the public and policy makers can judge the information for themselves. States inconsistently report and update data, with many retroactively changing numbers to such an extent that large daily swings in percentages are visible in many charts. The CRC accounts for these irregularities and notes when such major shifts occur and how the site has adjusted for them.

(NOTE: Test positivity is a measure of testing capacity that provides important context about case totals and trends. But it is not a complete measure of how prevalent the virus is in communities, due to the fact that numbers of tests being performed and who is being tested have changed over time due to factors other than infection levels. In addition, many people now use at-home tests that are not reported or are opting to not test if they lack serious symptoms. Policy decisions, like openings and closings or interstate travel, should not be determined based on test positivity alone.)

• Approach 1: Positive Specimens / Total Specimens. Positive molecular (PCR) tests divided by total molecular (PCR) tests.

• Approach 2: Positive Specimens / Total Encounters. Positive molecular (PCR) tests divided by total molecular (PCR) tests within a specified timeframe.

• Approach 3: Positive People / Total People. People who test positive with molecular (PCR) tests divided by the total number of people tested with molecular (PCR) tests.
  The CRC no longer supports Approach 3 because data for “Positive People” is no longer available.

• Approach 4: Positive People / Total Encounters. The number of people who test positive with molecular (PCR) tests divided by total number of molecular (PCR) tests given within a specified timeframe.

• Approach 5: Positive People / Total Specimens. The number of people who test positive with molecular (PCR) tests divided by total number of molecular (PCR) tests.

Hierarchy Each bar represents the result of the best available testing approach from above.