Testing for COVID-19
What tests are there and are they accurate?
Why testing is important
Dr Tedros Ghebreyesus, the Director General of the World Health Organization in March urged all nations to “test, test, test”, and indeed, testing for COVID-19 is one of the three pillars that epidemiologists use to control an epidemic: testing, contact tracing and isolation. Testing allows us to find out how much active disease is in the community, quarantine those found to be infected, and contact trace and quarantine those who have been in close contact with the cases and are potentially infected. Importantly, it also allows us to work out whether our control programs are working.
Donald Trump is correct in saying that the more you test, the more likely you are to find cases. However, the whole object of testing is to find cases! More important is the rate of positive tests. If 5% or more tests are positive, this is a sign that the epidemic is out of control. In the USA at the moment the rate of positive tests is over 15%. This compares to 0.4% in Australia as a whole, and 0.9% in the state of Victoria where there is a current outbreak. The rate of positive tests is also affected by who you test. If you test only people with symptoms, you are likely to have a much higher rate than if you tested random samples of the population.
The RT-PCR Test
Currently, testing for an active infection for COVID-19 involves taking a swab of the nose and throat, and then sending the swab to a laboratory where an RT-PCR test is undertaken. The RT-PCR looks for the genetic material of the virus in the sample, and is very accurate. We measure accuracy two ways. The first is called sensitivity. If someone actually has COVID-19, then sensitivity is the probability that the test will come out positive. If the test is incorrect, we call this a false negative. The second thing we measure is specificity. If someone really does not have COVID-19, then specificity is the probability that the test comes out negative. If the test gets it wrong, we call this a false positive. Ideally, we would love to have high sensitivity and specificity. Unfortunately, as one goes up, the other goes down.
False positives are not a major issue, since if the test gets it wrong and finds someone positive when they don’t have the disease, the worst that can happen is that they have to go into 14 days of isolation. However, false negatives are much more worrying, because it means that someone who is actually infectious can then wander around thinking they are disease-free and spreading the infection. It is therefore very important that any test for an active COVID-19 infection is very sensitive. No test is 100% accurate. The RT-PCR test has a sensitivity of 71–98%, in other words, the percentage of false negatives range from 2–29%. The RT-PCR device is expensive, needs to be in a proper laboratory, requires highly trained technicians, and takes several hours to produce results, and two or more days to get results back.
Point of care antigen tests
Point of care (POC) antigen devices are currently being assessed to diagnose COVID-19. Antigens are substances that produce an immune response when they enter the body. Only one antigen device has so far been authorized by the Federal Drugs Administration (FDA) and is for emergency use only. These are small, portable devices that can be used in clinics or hot spots to rapidly test people. They work by detecting fragments of antigens found on or within the virus using swab samples taken from the nose and throat. Although they can produce results in minutes, they are not as sensitive as RT-PCT tests.
Antigen POC tests are very specific — if you don’t have the disease, the test gets it right. However, they are not as sensitive as the RT-PCR test, with sensitivity ranging from 50–90%. At the lowest end, one in two infected people might be told that they do not have the disease.
For most diseases, we usually like a diagnostic test to have at least 80% sensitivity. Antigen POC tests have the advantage of being cheap, portable, inexpensive, and not requiring highly trained staff to run them. This means that tests can be undertaken quickly, and anyone found positive immediately isolated and contact tracing undertaken. This is incredibly important in reducing transmission of the diseases. One or two days can mean the difference between controlling the outbreak, or it getting out of hand. However, for most diseases, we usually like a diagnostic test to have at least 80% sensitivity. Potentially, the POC tests could be made more sensitive by detecting smaller traces of antigens. This would be at the expense of reducing specificity and increasing false positives. However, as I pointed out earlier,false positives are not a major issue, and until the antigen POC tests can be made more sensitive, it is unlikely that they will be widely used.
If you enjoyed this post, read my other articles on COVID-19.
