Accuracy of Covid-19 Tests with Bayes Rule
Source: https://www.cochrane.org/CD013705/INFECTN_how-accurate-are-rapid-antigen-tests-diagnosing-covid-19
Bayes Rule, a probability theory discovered by Thomas Bayes in 1763, provides us with a way to update our beliefs based on the arrival of new, relevant pieces of evidence. As we learned more about Bayes Rule during the lecture, I could not help but think about the application of Bayes Rule during the time of Covid-19 Testing.
The article discusses a significant concern through the development of Covid-19 testing, which was the accuracy and validity of the tests. The reverse transcription polymerase chain reaction test (RT-PCR) is very accurate when performed by a healthcare professional. PCR tests are accurate 95% of the time and only displayed false negatives or false positives 5% of the time. On the other hand, rapid antigen tests are only accurate 89% of the time and displayed false negatives or false positives 11% of the time. In these scenarios, this data was derived by researchers using the Bayes Rule. For example, researchers calculated the conditional probability that a patient actually had covid, given that they received a positive PCR covid test. The accuracy of these tests is crucial, as people with suspected COVID-19 need to know quickly whether they are infected so that they can self-isolate, receive treatment, and inform close contacts.
Bayes rule played an integral role during the height of the Covid-19 pandemic, as it affirmed the accuracy and reliability of Covid-19 tests. People all over the world were influenced by the accuracy rates of the two tests, as certain places only let tourists travel if they received a negative PCR test, not a negative rapid test because PCR tests have a smaller chance of producing false positives or negatives. Without the influence of Bayes Rule, the entire testing mechanism of Covid-19 would have been extremely chaotic and unreliable.