The primary type of testing for the novel coronavirus around the world, including Canada, produces “false-negative” results at least 20 per cent of the time, researchers from Johns Hopkins University found.
According to a study published in the Annals of Internal Medicine in May, the false-negative rate of RT-PCR testing used to detect the novel coronavirus changes depending on where a person is in the timeline of the infection cycle.
On day 1 of an infection, the test is completely ineffective at detecting the virus, while on day 8 of infection the test produces false negatives 20 per cent of the time, the study found. The rate of false-negatives then increases every day afterward.
On day five of infection, which is when symptoms of COVID-19 typically begin to appear, and when people are often encouraged to seek out testing, the study found the false-negative rate is 38 per cent.
“The probability of a false-negative result declines throughout the course of infection,” said Dr. Lauren Kucirka, a resident physician at Johns Hopkins and co-author of the study.
“If you test someone immediately after they’re infected, the false-negative rate is very high.”
Kucirka and her colleagues used test results collected in Germany, France, China, South Korea and the United States from more than 1,300 samples where patients first tested negative but were then found to be positive. Samples from Canada were not used.
The study doesn’t pinpoint exactly why the false-negative rate for COVID-19 testing is so high, but Kucirka said there are several possible reasons, including problems with how nasal samples are collected and “biological limitations” of the tests themselves that make it difficult or impossible to detect the virus depending on when the test is done and how contagious the person might be.
Kucirka says physicians and public health officials should be cautious about relying on negative test results when informing patients of their status and when developing public policies, such as when to initiate contact tracing, lifting quarantines or rules around personal protective equipment.
“These tests are useful but you have to be aware of the limitations,” she said.
“If you have someone where you really think they have a high probability of being infected, you shouldn’t just rest easy once you have the negative test result.
“Obviously, it would be nice to have a test that works better, and there are a lot of people working on developing things, but unfortunately this is the best we have right now.”
Public policy implications
As of Tuesday, Canada had completed roughly two million COVID-19 tests and identified 96,636 positive cases, meaning there’s been about 1.9 million negative tests so far.
Based on Kucirka’s findings, at least 380,000 of these tests could be false-negatives. Depending on when the tests were done during the course of a possible infection, the number of false-negatives may be even higher.
“That is scary,” said Benoit Burbeau, a virologist at the Université du Québec à Montréal when referring to the study.
Burbeau said the implications of the study on public policy decisions, especially contact tracing, are huge.
Since the onset of the pandemic, health officials and politicians have been stressing the importance of increased testing, including testing of asymptomatic people, when deciding how and when to introduce — and remove — certain mitigation measures, such as travel restrictions, physical distancing, bans on social gatherings and closing public parks and businesses.
But if the tests being used to determine who is and isn’t infected have a high rate of uncertainty, these decisions are being made based on unreliable information, Burbeau said.
“If you have 30 per cent false-negatives, then right now, at this moment, you’re in trouble,” he said.
The effectiveness of contact tracing – where public health officials contact anyone who’s been in close contact with a person confirmed as COVID-19 positive to advise them of potential risks and measures they should take – will also be significantly impacted by a high false-negative rate, Burbeau said.
“A way too high percentage of those tests that should have been positive are being missed,” he said.
This, Burbeau said, could lead people into a false sense of security, causing them to behave as if they’re not infected and not capable of spreading the virus, even when they are.
For this reason, he believes doctors and public health officials should be actively informing people who test negative that there’s a good chance they could be positive, especially if they show symptoms and have had only one test.
He also said testing needs to improve and more research needs to be done to address the possible causes of false-negatives, including inconsistencies with testing protocols around the world and inadequate training for those who collect samples.
Follow-up tests encouraged
Dr. Isaac Bogoch, an infectious disease specialist based out of Toronto General Hospital, said the Johns Hopkins study is important for doctors and public health officials because it underscores the uncertainty that’s typical of the novel coronavirus. It also shows that a single negative test, especially when someone is symptomatic, shouldn’t be used to “rule out” infection.
Bogoch is unsure whether the study’s findings should be used to make broad conclusions about the number of people who may or may not be infected based on negative test results. However, he agrees the study shows there are limitations to testing.
When a patient tests negative but is displaying symptoms, he encourages them to take a second test within a few days, he said.
Two negative tests within several days – assuming the samples taken were good quality – means there’s a high probability the person is not infected with the virus, he said. Physicians should then consider other possible causes for illness.
“It’s important not to have tunnel vision,” Bogoch said.
Things become challenging when a second test isn’t done, he said. Without this additional information, dedicating time and limited resources to things like contact tracing and implementing proper quarantine protocols is difficult.
But if a second test is done and turns out positive, then all the recommended precautions can be followed.
“It’s better late than never,” he said.
Global News asked the Public Health Agency of Canada to comment on the study and whether it is concerned by the findings, as well as what, if any, implications a high false-negative rate could have on public policy decisions, such as contact tracing. A response was not provided in time for publication.
Toronto Public Health, meanwhile, said testing is an important part of understanding how the virus circulates in a community and for preventing further spread.
Anyone who comes in contact with someone who tests positive for COVID-19 is also required to self-isolate for 14 days, regardless of whether they test negative, Toronto health officials said.
© 2020 Global News, a division of Corus Entertainment Inc.