Bayes’ Theorem and COVID-19 Testing
Taking a look within the global pandemic that we have all been fighting against for the last two years, there’s been a lot of data that’s been computed and analyzed to help create a safe environment for us all. Bayesian probability and statistics has a very useful concept in regards of COVID-19 testing. In the article “Varsity Explains: Bayes’ Theorem and COVID-19 Testing” by Nick Scott, Scott analyzes the true efficiency and positivity rates that are displayed by the current types of COVID-19 testing. In the beginning of the article, Scott starts off by saying that just because an individual tests positive for COVID-19, doesn’t necessarily mean that that individual actually has COVID-19. Most COVID-19 tests are not fully accurate, generally speaking they are about 95% accurate. Scott then proceeds to discuss some of the principles of the Bayes’ Theorem. The Bayes theorem requires conditions given a certain event.
Scott gave an example of an individual contracting COVID-19, however he gave different scenarios. The first scenario was that the individual was in the UK during February 2021, this was one of the worst outbreaks during the pandemic, the other scenario was that the individual contracted the virus on an isolated island where there has been no reported cases. These scenarios show that context of events are essential to the computation of data, this is how the conditional probability of Bayes helps us understand the probability of something given a specific event. Bayes Theorem was a large part of a lecture during the last several weeks. This article demonstrates how the theorem can be applied and efficiently to the context of real world scenarios.
https://www.varsity.co.uk/science/21149