Here's a group of 200 students. We'll test them all for Covid-19.

Before we do that, we need to know their pre-test risk of infection. On campus, 11% of the population has coronavirus. We'll use this as their baseline risk, for now.

Choose a different pre-test risk of infection. Use these presets, which are rough estimates, or enter your own.

Assume all 200 students have a mild cough

Assume all students have serious Covid-19 symptoms

Pre-test risk of infection:
%
K

Infected

d

Not infected

Now, let's simulate a diagnostic test. On average, this test correctly identifies % of people who K are infected (sensitivity) and it correctly identifies % of people who T aren't infected (specificity).

Try a type of test currently in use:

PCR or molecular tests are the most accurate. In general, they’ll have a sensitivity around 97% and a specificity of 100%.

Rapid tests are generally less accurate than PCR tests. An antigen test might have a sensitivity of 70% and a specificity of 99.5%.

Rapid molecular tests have an average sensitivity of 95% and an average specificity of 99%.

K

Infected

d
m

Not infected

people tested positive.

people tested negative.

That means that if you took the test and got a positive result, there’s a % chance it’s correct. If you got a negative result—told you didn’t have the virus—there’s a % chance your result is accurate.

Positive results

Our test correctly identified C with Covid-19, but said K had the virus when they actually didn’t.

Negative results

Our test correctly identified P without Covid-19 but F with the virus were incorrectly told they were uninfected.