Researchers at UC Berkeley and UCSF evaluated the reliability and accuracy of 12 COVID-19 test kits in response to the need for accessible and dependable testing.
The project focused on assessing the quality of antibody test kits, which test for the presence and quantity of antibodies produced by the body in response to a virus, according to Patrick Hsu, a lead researcher and UC Berkeley assistant professor of bioengineering. The test, which can complement and confirm current diagnostic tests, can determine who has been infected weeks after symptoms first occur, Hsu added.
“You might feel ill and have the symptoms and then go and get a molecular test, but the sensitivity of the test could go down after a couple of weeks,” Hsu said. “Using antibody tests, we can see who had an immune response and who didn’t present symptoms despite showing the presence of these antibodies.”
In the project, researchers assessed the sensitivity and specificity of each test in order to determine the effectiveness of the kits. According to Hsu, measuring the sensitivity of a test helps determine how often the test provides a false negative, meaning that a patient who has COVID-19 tests negative. The specificity of a test shows the odds of a person falsely testing positive despite not having the disease, Hsu added.
The sensitivity of each test was measured by using the antibody kits on blood samples that were already confirmed positive for COVID-19, and specificity was assessed by using the kits to test blood samples collected from 2018 or earlier that were known to be negative for COVID-19.
“We have our website with all of our data to serve as resources that will contribute to medical research and inform government implementation,” Hsu said. “We aren’t making recommendations, but this can inform leaders and companies which tests work and guide them to develop better tests.”
One question regarding policy that could be influenced by identifying more effective tests is when governors and other political leaders should relax current shelter-in-place guidelines, according to Robert Sanders, UC Berkeley manager of science communications.
While the antibody tests can identify individuals who contracted SARS-CoV-2, the virus that causes COVID-19, despite not showing symptoms, they cannot definitively say whether someone is immune, according to Hsu. Researchers need to look into the implications of testing positive for the presence of antibodies before assuming that recovering from an infection offers protection in the future.
“People are discussing the idea of an immunity passport, meaning that people who are immune to the virus can go back to their normal lives,” Hsu said. “Right now, that is something that we cannot conclude from antibody tests alone, no matter how reliable they are.”