My Antibodies Test
In mid-February, after spending four days in New York City, I developed a respiratory viral infection that featured certain symptoms associated with Covid-19. Yesterday I had an antibodies test that is supposed to be 99% accurate. I expect to learn the results today. You can’t evaluate the results of these tests without some statistical analysis. If the test shows a false positive only 1% of the time, that doesn’t mean that, if you test positive, there’s only a 1% chance it’s a false positive. Your chance of a false positive is affected by how many people who take the test are positive, and how many are not. For example, if out of 1,000 people who take the test, only 10% are positive, out of the 900 people who aren’t positive, 1% –– 9 people –– will test false positive. So, if you are test taker #1001, you’re roughly nine times more likely to be a false positive than a true positive.
I think I have that right.
When I get my test results today, I’m going to try to find out what percent of the people who have taken the test in my cohort, if there is such a thing as my cohort, tested positive. I’ll be surprised if I get a straight answer to this question. Even if I do, I’m not sure it will help me figure out what the odds are that I really do have antibodies even if my test came out positive.