This whole testing thing is pretty bad when no one can seem to give you a definitive answer on how many of these tests we are getting right.
Timothy Sly, a professor emeritus at Ryerson University's School of Occupational and Public Health, says this is the difference between a test's sensitivity and its specificity.
"The sensitivity is to say: How many of the true positives have we really got? And therefore, the others are false. And the specificity is: How many of the true negatives were we able to find? And of course the others are false," he says.
Many factors can affect these outcomes, including the quality of the sample itself, the type of tools used to obtain and assess the sample, and at what stage of infection the sample was obtained. Patterson said that variability makes it hard to define a hard-and-fast accuracy rate.
"False negatives can occur, certainly up to 30 per cent of the time if we swab people who are asymptomatic and swab them too soon after they've been exposed to COVID-19," said Patterson.
"False positives depend a lot on the machine itself, but also how much COVID-19 is in a town or community that you're assessing or sampling. And so it's really hard to nail down a rate that applies to every machine that we use."
Testing negative for the coronavirus doesn't necessarily mean that you don't have COVID-19, the illness caused by the virus, and experts say the high rate of false negatives could mean many cases aren't being caught.
A recent study in the Annals of Internal Medicine from researchers at Johns Hopkins University in Baltimore found the effectiveness of
RT-PCR tests, the most common type of test for the coronavirus globally, varies dramatically over the course of COVID-19 infection.
In the first few days before symptoms typically show up, researchers found, the chance of getting a false negative result ranged from 100 per cent on Day 1 to 67 per cent on Day 4.
After symptoms began to show, false negative results fell to 38 per cent on Day 5 and to 20 per cent on Day 8 – but rose again everyday after that.
That means at best, COVID-19 patients were found to have a one in five chance of getting a negative test result during their infection even if they were actually positive.
"The important takeaway is that we know the false negative rate is very high," said Lauren Kucirka, a resident physician at Johns Hopkins and a lead author of the study.