According to the Johns Hopkins University report, the accuracy of the tests peaks at 80%. Accuracy rates appear to be contingent upon the stage of infection, with the highest percentage of false-negatives reported in the first four days of infection, when people are often asymptomatic. If testing occurs on the eighth day of infection—usually three days after symptom onset—results are more accurate.

“Most people don’t know when exactly they contracted the virus, so the fact that testing on a wrong day could produce a false-negative result is a considerable problem,” Ayah Rashwan, DMD, a New York-based pediatric dental resident who pivoted to treating COVID-19 patients in intensive care, tells Verywell. Rashwan was not involved with the Johns Hopkins Report. “This could lead to super spreaders who are rapidly spreading the virus and unknowingly getting their family members and friends sick.”

Tests Detect Different Results on Different Days 

To gauge test accuracy, researchers at Johns Hopkins assessed data from seven prior studies, which examined 1,330 nasal swab samples taken from both hospitalized and non-hospitalized patients. The researchers then compared data from RT-PCR tests to patients’ history of COVID-19 exposure and onset of symptoms.

From there, false-negative rates were evaluated on a day-by-day basis:

Other studies mirror this finding. One patient-led study found that people who got tested too late in their infection received negative results, despite exhibiting the same long-term symptoms as those who received positive results by testing earlier.

“This data implies that the timing of the testing really matters," Hannah Wei, a qualitative researcher who analyzed these patient survey responses, tells Verywell. “If you see that the symptoms are similar but the day of testing is off by even a week, it will lead to patients getting denied the treatment that they really need.”

Why RT-PCR Tests Are Not Always Accurate 

The researchers at Johns Hopkins say one reason for a high number of false-negatives results could be the variability in viral shedding, which is the rate the virus leaves your body after when it can no longer reproduce.

Faulty sample collection techniques may also be to blame, which can happen if a clinician doesn’t swab deep enough to collect the necessary samples. 

Some test kits possess faulty reagents—the chemicals used in tests to help detect SARS-CoV-2. In a press conference, Nancy Messonnier, MD, director of the National Center for Immunization and Respiratory Diseases at the Centers for Disease Control and Prevention (CDC), explained that tests may not always work as well in other labs as they do in CDC labs, leading to inaccurate or inconclusive results.

To circumvent this issue, the CDC is taking extra measures to ensure high-quality reagent manufacturing. 

How Hospitals Are Working to Prevent False-Negatives

These false-negative test results occur both in lab and hospital settings. 

Amira El Kholy, APRN, an ICU nurse practitioner who works at multiple hospitals throughout Miami, Florida, says her hospitals aren’t just relying on one test result; they’re administering three different tests on patients experiencing COVID-19 symptoms:

Abbott rapid test: Displays results in a few hoursRT-PCR test: Takes about three days to processAntibody blood test: Helps determine whether a patient has an active infection or previously had an active infection 

According to El Kholy, the Abbott rapid tests are extremely faulty, so if a patient is clearly exhibiting COVID-19 symptoms, they will be kept in isolation until RT-PCR results come back. If those results also come back negative but clinicians believe a patient has contracted the SARS-CoV-2 virus, they will perform the RT-PCR tests an average of three more times in order to obtain a positive result. 

“If a patient is positive, we know for sure to put them in isolation. If they’re negative, we still isolate them and wait to further confirm their results," El Kholy tells Verywell. “It’s important to isolate a patient if they have symptoms that are consistent with COVID-19, despite a negative test result.”

Other Diagnostic Tools

The authors of the Johns Hopkins report aren’t the only scientists emphasizing that different diagnostic measures should be used to supplement RT‐PCR tests. Researchers from Wuhan, China, for example, suggest the use of CT scans to both help diagnose COVID-19 and monitor a patient’s progress.

According to El Kholy, blood tests that check for indications of inflammation—including C-reactive protein, creatine phosphokinase, D-Dimer, and lactate dehydrogenase—would be useful and simple additions to the diagnostic process.

Other experts recommend taking samples from a patient’s lower respiratory tract, rather than relying solely on the upper respiratory tract samples typically used in RT-PCR tests.

“Despite these findings, we still encourage the use of RT‐PCR tests because they’re important for testing, tracing, and curbing the rates of COVID-19,” Rashwan says. “It’s just important to be cognizant of the limitations of them and employ other diagnostic measures.”

The information in this article is current as of the date listed, which means newer information may be available when you read this. For the most recent updates on COVID-19, visit our coronavirus news page.

There was an error. Please try again.