Information about the novel coronavirus (the virus that causes COVID-19) is constantly evolving. We will refresh our novel coronavirus content periodically based on newly published peer-reviewed findings to which we have access. For the most reliable and up-to-date information, please visit the CDC website or the WHO’s advice for the public.
Which COVID test is the most accurate?
Typically, the most accurate COVID tests are the PCR tests. But they’re not always the best choice.
Overall, the PCR test does a great job identifying people who have COVID, making it highly sensitive. And while they have near-perfect accuracy in a laboratory setting, in real life, COVID PCR tests are about 80% sensitive (Yohe, n.d.). So the PCR test may miss people who actually have COVID roughly 20% of the time.
Because no test is 100% accurate all of the time, most healthcare providers will use other information—like potential exposure, symptoms, and how high-risk someone is—when interpreting your test results. While PCR is very good at detecting the COVID virus, it’s more expensive and takes longer than antigen tests. Samples often need to be sent to laboratories with expensive PCR machines, which can take hours or even days before you get your results.
Rapid antigen tests are not as accurate as PCR testing, but they can be performed faster and are relatively inexpensive. But if the virus isn’t present in your sample, or it’s too early in the infection, the test may be negative when you actually have COVID.
If you have an inconclusive or negative antigen test and have symptoms of COVID, the Centers for Disease Control and Prevention (CDC) recommends getting a PCR test to confirm the results. No matter which test you get, you should quarantine while waiting for your results (CDC, 2021a).
Antibody testing is an accurate way to determine if you’ve been infected with the coronavirus in the past, but isn’t used for diagnosing whether you have it right now. Fortunately, most of the available antibody tests are great at picking up people with antibodies so you can be pretty confident that the results are accurate (CDC, 2021a).
Overall, most of the COVID tests that have been approved by the FDA are pretty good at finding out if you have COVID or not. PCR tests are considered the most accurate. While they were developed to catch every single case of COVID, in real life, it doesn’t hold true. There are all kinds of factors that can make these tests less accurate, including improper collection, contamination with other samples, and more.
And while antigen tests aren’t as good at catching every single case of COVID, that’s not always such a bad thing. Researchers have found that those super-accurate PCR tests remain positive after people are no longer infectious, meaning they’re not a good option in deciding whether or not you still need to quarantine (Mina, 2020).
- The accuracy of a test depends on how good it is at correctly identifying people who have the virus, and correctly identifying people who do not. While PCR tests are really accurate, they can take longer than antigen tests, and sometimes speed is essential. The accuracy of a test can also be affected by sampling techniques and other factors.
What are the different types of COVID tests?
The most commonly used COVID tests include (La Marca, 2020):
- PCR tests detect the presence of the virus’s genetic material.
- Antigen tests look for virus-specific proteins (like the spike proteins on the virus’s outer shell).
- Antibody tests, sometimes called serology tests, check for your body’s reaction to the virus and indicate a past infection.
How is the accuracy of different COVID-19 tests measured?
Testing is an integral part of managing the spread of the pandemic. People who test positive can then isolate themselves to avoid exposing others to the virus. But for a test to be useful, it needs to be accurate.
And while no test is 100% accurate, scientists can measure two important characteristics to determine test accuracy: sensitivity and specificity (Swift, 2020).
Sensitivity is a measure of how often a person who has COVID will actually test positive. Specificity indicates how well a test correctly identifies people who do not have the virus as negative (actually test positive).
Let’s use an example to illustrate how sensitivity and specificity work.
Imagine we have two people. This is Dan. He has COVID. He gets tested, and the test comes back positive. This is a true positive because it’s true that he’s positive for COVID. But what if the test is wrong and says that he doesn’t have COVID, even though he does? That is considered a false negative because it’s false to say that he’s negative.
Let’s look at our other friend, Stan.
Stan doesn’t have the coronavirus. He takes a COVID test, which comes back negative. This is a true negative, confirming that he does not have COVID. But if the test incorrectly says Stan has COVID, it’s a false positive because he is not actually positive.
A perfectly accurate test would identify all the people who have COVID as positive and all people who don’t have COVID as negative.
So what are sensitivity and specificity?
Sensitivity measures how frequently a test correctly spots people who have COVID (true positives). If a test is 98% sensitive, it means that out of 100 people who actually have COVID, 98 cases will be true positives and two people who have COVID will test negative, even though they have the virus.
Specificity measures how well a test correctly marks people who do not have COVID as negative (true negatives). If a test is 98% specific, it means that out of 100 people without COVID, it will correctly identify that 98 of those people are COVID-free. But it will incorrectly identify two of them as positive for COVID (false positives).
When it comes to using COVID tests, we want ones with high sensitivity so we can catch every case of COVID. This way, people can isolate themselves and avoid spreading the virus. But we also want to make sure that the tests have high specificity and don’t incorrectly say that person has COVID too often.
Telling people that they have COVID when they don’t can be stressful and disruptive. That’s why scientists aim for tests with the highest rate of sensitivity and specificity possible.
What to do if you test positive for COVID-19?
If you test positive for the coronavirus, you should follow the CDC guidelines and isolate at home, including from people in your own household (CDC, 2021b). Be sure to inform your healthcare provider and any close contacts you may have had over the past few days.
You can stop isolating once (CDC, 2021a):
- At least 10 days have passed since your positive test if you did not have any symptoms
- At least 10 days have passed since your symptoms first appeared, and
- You have been fever-free for at least 24 hours without using fever-reducing medication, any other COVID symptoms you have are improving.
- Centers for Disease Control and Prevention (CDC). Interim Guidance for Antigen Testing for SARS-CoV-2. (2021a, August 2). Retrieved 13 January 2021 from https://www.cdc.gov/coronavirus/2019-ncov/symptoms-testing/testing.html
- Centers for Disease Control and Prevention (CDC). Isolate if you are Sick (2021b, July 29) Retrieved 14 January 2021 from https://www.cdc.gov/coronavirus/2019-ncov/if-you-are-sick/isolation.html
- La Marca, A., Capuzzo, M., Paglia, T., Roli, L., Trenti, T., & Nelson, S. M. (2020). Testing for SARS-CoV-2 (COVID-19): a systematic review and clinical guide to molecular and serological in-vitro diagnostic assays. Reproductive biomedicine online, 41(3), 483–499. Retrieved from https://www.ncbi.nlm.nih.gov/pmc/articles/PMC7293848/
- Mina, M. J., Parker, R., & Larremore, D. B. (2020). Rethinking Covid-19 Test Sensitivity – A Strategy for Containment. The New England journal of medicine, 383(22), e120. Retrieved from https://www.nejm.org/doi/10.1056/NEJMp2025631
- Swift A., Heale R., Twycross A. (2020). What are sensitivity and specificity? Evidence-Based Nursing, 23:2-4. Retrieved from https://ebn.bmj.com/content/23/1/2
- Yohe, S., MD. (n.d.). How Good are COVID-19 (SARS-CoV-2) Diagnostic PCR Tests? CAP: College of American Pathologists. Retrieved January 9, 2021, from https://www.cap.org/member-resources/articles/how-good-are-covid-19-sars-cov-2-diagnostic-pcr-tests#:~:text=The%20analytic%20performance%20of%20PCR,specificity%20is%20near%20100%25%20also