If Rapid Test Is Negative, 15 Percent Are Wrong. If It's Positive, You Are Infected. Period.

The data on rapid test results are becoming more conclusive. Rapid tests can be wrong 15% of the time - IF they say the person who took the test is negative. 

But if the test shows a positive result it is almost always correct. That is, 98% of the time. Those who test positive should assume they are infected and act accordingly. JL

Joe Schwartcz, PhD reports in the McGill University Dept. of Science and Society:

“Sensitivity” is the ability to detect infection, and “specificity” is the ability to detect the absence of infection. If the sensitivity is 85%, which it is for the rapid tests, then out of every hundred infected people tested, 15 will get a false negative. False negatives are relatively common, but false positives are not. If the rapid test is negative,15% of the time there will be a false negative. If a test comes back positive, it is quite certain that the person is harboring a significant viral load and is infectious. Rapid tests identify 98% of infections.

To test or not to test, that is the question, to steal a line from Shakespeare. Actually, when to test is a more important query because there is no doubt that appropriate testing of the population can cut down infections by identifying people who may be a threat to others and advising them to self-isolate. The gold standard for detecting the presence of the virus in secretions is the so-called polymerase chain reaction (PCR) test which identifies genetic material from the virus. It does this very well, capable of picking up infection just a couple of days after exposure to the virus. It is in fact so sensitive that it can give a positive result even after a person is no longer infectious since remnants of genetic material from the virus can still be floating around.  

PCR is expensive, requires samples to be sent to a lab, and delivery of results may take days. That makes the technique impractical for screening asymptomatic individuals who may have come into contact with an infected person. Such individuals may have acquired the virus, and can still transmit it even though they themselves exhibit no symptoms. Enter the “rapid COVID test” which can be easily administered and yields results in fifteen minutes. If positive, a person is immediately alerted of the need to self-isolate for ten days and to inform contacts of possible infection. The overall goal of such screening is to detect asymptomatic or presymptomatic infection and reduce transmission of the virus. 

These tests, also known as “lateral flow tests,” of which there are a number, do not detect genetic material but rather identify proteins that are encoded by the virus’ genes. These proteins, termed antigens, bind to antibodies that have been designed to recognize them and the antigen-antibody reaction produces a coloured line on a test strip much like a pregnancy test.  

There are two terms that are important to understand when it comes to discussing a test’s accuracy. “Sensitivity” is the ability to detect infection by the virus, and “specificity” is the ability to detect the absence of infection. For example, if the sensitivity is around 85%, which is roughly what it is for the rapid tests, then out of every hundred infected people tested, 15 will get a false negative. If the specificity is in the neighbourhood of 99%, which it is with most tests, then out of 100 positive tests, only one will be false. This means that false negatives are relatively common, and false positives are not. In other words, if a test comes back as positive, it is quite certain that the person is harbouring a significant viral load and is infectious. 

The next item to understand is that for the test to be able to give a reliable result, the virus will have had to multiply sufficiently to produce enough of the target protein. This takes at least 2-5 days, so there is no point doing a test if less time has passed since contact with a possibly infected person. For example, a negative test taken immediately after coming off an airplane is meaningless in terms of deciding whether you will be putting people you meet at risk. Although at this time nasal secretions may not contain enough virus for a positive test, there may still be enough virus being emitted to infect others.  

Let’s now take a look at some possible scenarios. Imagine that you have been to a gathering and are concerned that someone there may have been shedding virus, but you feel fine. At least two days have passed, and you do a rapid test. If it comes back positive, you should immediately try to isolate because false positives are rare. A PCR test now can absolutely confirm infection. If the rapid test is negative, you are not out of the woods because 15% of the time there will be a false negative. The test should then be repeated a day or two later, and if that is also negative, the chance of infection is small. Studies have shown that sequential negative tests approach PCR in sensitivity. When groups of students, employees, or athletes are tested every few days, the rapid tests can identify 98% of infections. 

Next, what if you have been in a situation in which you have had contact with others and are now experiencing minor symptoms like coughing, sneezing, sore throat or fever. You are not sure whether this is just a cold or the beginning of COVID. It is best to assume that it is COVID until it can be ruled out. Once symptoms are present, the sensitivity of the rapid test is greatly increased because the virus is now present in greater quantity. If the test is positive, it is very likely that this is not just a cold and you should immediately self-isolate and advise people with whom you may have been in contact of the situation. Again, PCR can confirm. If the rapid test is negative, repeat it in two days, and if it is still negative, then COVID is unlikely. 

What if you want to visit an older family member or attend a small gathering? For a test to be meaningful, it should be done just before the event. A two-day old test is unreliable because rapid tests reflect the viral situation only at the time the sample is taken and viral loads can change significantly day-to-day. If the test is positive, then there is no question. You cancel the visit and self-isolate and then confirm that you are indeed infected by doing a PCR test. If the test is negative, it suggests a low viral load at that point in time and the risk of being infectious is small. However, it is not zero because you may have been in contact with an infected person in the past couple of days and may have acquired the virus, but it has not had enough time to replicate in order to yield a positive test. If this is the case, you can still pass on some virus, especially if the infection was due to the omicron variant. This variant replicates so quickly that you may be negative when you test just before a gathering, but were you to repeat the test within a few hours, it might well be positive. Also, remember that there is a 15% chance that you are looking at a false negative in the first place, so there is still some risk associated with that possibility. Keep in mind, though, that life is a continuous series of risk-benefit evaluations and we take some risks all the time. A gathering where everyone has tested negative presents a very small risk. 

At this point, early evidence indicates that the rapid tests work for the delta and omicron variants as well as for the original SARS-CoV-2 since the test is not for the famous spike protein the virus uses to invade a cell, but rather for the nucleocapsid protein which is the framework around which the virus’ RNA is wound. This is less susceptible to mutations. With any of the rapid tests, the chance of false negatives is reduced by repeating the test a couple of days apart.  

Ideally, unless of course there has been no chance of contact with anyone, people would to take a rapid test every few days. This would identify infected individuals who would then isolate and prevent the spread of the virus. Realistically, this is not practical in terms of hassle or cost. On the other hand, contracting COVID is a real hassle and the cost can be very high. Of course, the most effective way to reduce the spread of the virus is to limit person-to-person contact and to get as many people vaccinated as possible. Finally, remember that a negative test means that you are less likely to infect others, but says nothing about your risk of being infected.