Our perspective on..

Corona rapid test interpretation

What does my Corona rapid test result mean? The correct interpretation of test results depends not only on how good the test is, but also on how common the disease is. In this article we explain how to interpret Corona rapid tests correctly.

No Corona despite a positive rapid test? The spread does it!

The infection figures are falling, the restrictions are being lifted step by step. In this situation, the widespread use of rapid tests in particular should ensure that an orderly return to normality is possible. Thus, everyone can test themselves for Corona immediately before visiting their parents or friends, before shopping or going to a restaurant or concert. Since these tests are also widely performed by lay people, it is particularly important that they are able to understand and correctly interpret the test results. Under certain circumstances, it is likely that one is not infected despite a positive rapid test – and vice versa. How can this be? Simply Rational, in collaboration with the Robert Koch Institute, has produced a guide to interpreting rapid test results. It is not only the accuracy (the so-called specificity and sensitivity of the test) that is important, but also, and above all, the spread of the virus in a particular population or population group.

The Paul Ehrlich Institute has defined minimum requirements for the sensitivity and specificity of rapid tests. To be approved, a rapid test must have a sensitivity of at least 80% and a specificity of at least 97%. What do these values mean? Sensitivity is the ability of a test to detect people with the disease: a test with 80% sensitivity would therefore correctly identify 80 out of 100 people with SARS-CoV-2 infection, i.e. it would show a “positive” result. 20 out of 100 would falsely receive a negative result. Specificity, in turn, indicates how good the test is at detecting those who are not ill. In a test with a 97% specificity, 97 out of 100 non-infected people would also get a correct “negative” result. However, 3 out of 100 would get a false positive result.

Often overlooked, however, is a third, extremely decisive influencing factor: prevalence, i.e. the spread of the virus in a population or group. For in most cases it is not the case that there is an equal number of infected and non-infected people in a population. In most cases, those who are not infected outnumber those who are. To understand the relationship between sensitivity, specificity and prevalence, we can use so-called natural frequency trees (Fig. 1 and 2). Let us imagine a hypothetical population of 1,000,000 people. With a prevalence of 0.01%, for example, which corresponds to 100 infected people out of 1,000,000 (Fig. 1), a sensitivity of 80% and a specificity of 97%, 29,997 people will test positive, but only 80 of them will actually be infected (Fig. 1). This is because there are so many more uninfected people than infected people. If almost 100% of the population are not ill and (at 97% specificity) 3% of them get a false-positive test result, then this obviously exceeds the proportion of ill people, namely 0.01%, many times over (300 times over, to be precise). The fact that the rapid test also only correctly identifies 80% of those with the disease makes this situation even worse, so that in this situation only 0.3% of those with a positive test result actually have corona.

Figure 1

Let’s look at another situation where infection is rampant and 40% of a population or group is infected (see Figure 2). Now the infected and uninfected are almost in balance and now only 5.3% of all those who tested positive get a false positive result. The vast majority, 94.7% got a positive test result because they are really infected with Corona. So in this situation, if you get a positive rapid test result, 94.7% of you can assume that you are actually infected. Therefore, the first important finding about (rapid) tests is that their significance depends strongly on the spread of the virus. If the prevalence increases, the significance of positive rapid test results increases and that of negative ones decreases. If the prevalence falls, the opposite is true. For this reason, area-wide rapid tests are only partially effective when the virus is not widespread, as they can produce a large number of false-positive tests.

Figure 2

Therefore, a positive rapid test result should always be confirmed with a PCR test. In the current situation in Germany, however, one can be quite sure with a negative rapid test result that one is not infected (over 99%).

Have we aroused your interest? Find out more details here about when (rapid) tests can be an effective tool in the fight against Corona and when it is better not to use them.

Our perspective on..