COVID Testing FAQ, From Reinfection to Persistent Positives

Eugenia Yun

June 10, 2020

Editor's note: Find the latest COVID-19 news and guidance in Medscape's Coronavirus Resource Center.

Only months into the COVID-19 pandemic, confusion remains on the topic of testing for SARS-CoV-2, with many physicians grappling with similar types of questions.

"My patient is still positive for COVID-19 a month and a half later. Why?"

"Has my patient been reinfected, or is this just delayed recovery?"

Most of these questions revolve around what these tests are looking for, the nuances of interpreting test results, and persistent positive results in asymptomatic patients. Medscape spoke with Paul Auwaerter, MD, to get the answers to our readers' frequently asked questions about coronavirus testing. This interview has been edited for length and clarity.

What types of diagnostic and serologic/antibody tests are currently available, and how are they different? How reliable is each type of test?

The most commonly available tests are the molecular assays that use PCR technology to detect parts of the novel coronavirus RNA. Samples of the upper respiratory tract are taken using a swab that is inserted into the nostril to the nasopharynx. Some labs have also validated tests that may use swabs of the oropharynx, sputum, or even saliva. A positive result is highly accurate, indicating presence of SARS-CoV-2 RNA.

The number of false-negative results for these tests is not well known. One analysis of seven studies suggested that the rate may be approximately 20%, but it depends on when the sample is acquired relative to the timing of illness. For patients with a high suspicion of COVID-19, a negative test should not rule out the infection. There is also variability among manufacturers and in-house assays, as well as issues related to whether specimens were procured properly. 

Serological tests determine if antibodies thought to be specific to the novel coronavirus are detected. Studies suggest that antibodies are starting to be produced about 7 days after onset of symptoms, and that by day 14-28, most patients have developed detectable antibodies. What is not yet well understood is whether these antibodies offer protective immunity and whether they genuinely reflect a history of having had the infection in someone who has not had symptoms of COVID-19.

The Centers for Disease Control and Prevention uses purified spike protein of the novel coronavirus in an ELISA [enzyme-linked immunosorbent] assay. They feel this test offers a 99% specificity and 96% sensitivity. Other manufacturers may cite accuracy results, but they may not have been validated on a large number of clinical samples.

The CDC urges that only tests approved under the FDA Emergency Use Act (EUA) be used. CDC guidance suggests only testing people who have a reasonable probability of having had COVID-19 or using two different serological tests to help minimize the chance of false positives. The serological tests are most helpful in securing a late diagnosis of COVID-19 when molecular swab tests may be negative.

Does a positive COVID-19 PCR test result indicate that the individual is infectious?

Because the COVID-19 PCR tests only detect a fragment of the viral RNA, a positive test does not mean that infectious virus is present. In one small study of people who had mild or asymptomatic novel coronavirus infection, virus was not cultured from the upper airway after day 10 of illness, but detection of viral carriage by the molecular assay has been seen more than 80 days after the initial infection.

A study carried out by the Korean CDC (KCDC) examined patients who had been hospitalized for COVID-19 but subsequently tested negative by PCR testing and were discharged home. Of 285 people who later retested positive, no instances of infection were found in 790 household family members and other close contacts. This strongly suggests that when there is detection of viral RNA at this stage, people are no longer infectious. The KCDC, therefore, now no longer requires isolation in such cases.  

In general, people who have had the infection for 10 to 14 days are probably no longer infectious; however, this is not yet fully understood for all populations, such as immunosuppressed people or those who are severely ill in hospital. 

There are reports that this virus is unable to be cultured after day 8 or 9 of infection. Is it because the material is bound to neutralizing antibodies, or only fragments are left?

The adaptive immune system is thought to play a major role in viral clearance. Production of neutralizing antibodies plays a role, but they are probably not the only players. T-cell responses, such as CD8 (cytotoxic T cells) and CD4, help clear infected cells or coordinate immune responses. Robust generation of T-cell epitopes to viral proteins highly suggests that T cells play a significant role in dealing with this pathogen. The detection of viral RNA after day 10 in people who are less ill doesn't yield virus by culture, suggesting that only fragments of the RNA are found or that they are produced in cells but not yielding productive, intact virus.

Can the available tests differentiate between SARS-CoV-2 and other coronaviruses that may be similar? How would that affect test accuracy and reliability?

Most commercially available multiplex PCR panels for respiratory viruses do not detect SARS-CoV-2, except the BioFire® Respiratory 2.1 panel, which is authorized under FDA EUA. Many companies have in development new panels that incorporate the novel coronavirus, but they are not yet FDA-approved.

SARS-CoV-2 PCR tests do not pick up routine respiratory coronaviruses (HCoV-OC43, HCoV-229E, HCoV-NL63, and HCoV-HKU1) or MERS-CoV.

Commercially available antibody tests have shown significant variability. Moreover, even if a test is 96% to 98% specific, if the pretest probability of COVID-19 is < 5%, it is more likely that the result is a false-positive than a true positive — and one of the key reasons mass serologic testing is not recommended, as it may give people a false sense of security.

Causes for false-positives include cross-reactivity with other coronaviruses, generation of antibodies due to other mechanisms, or poor test performance.

An Annals of Internal Medicine study describes how false-negative rates vary across the disease course, from 100% 4 days before symptom onset to 20% 3 days after symptom onset. With such high variability, how should doctors interpret results? What if a patient misses the window during which the false-negative rate is lowest?

If there is high clinical suspicion for COVID-19, repeat testing is suggested. If the patient has evaluation later in the illness, perhaps in the second or third week, serologic testing may help yield an answer if PCR testing on respiratory specimens is negative.

Is there a difference in false-negative rates in asymptomatic (lower viral load) vs symptomatic patients (higher viral load)?

There are probably more false-negatives in asymptomatic people with this coronavirus; however, no large definitive studies have been carried out yet that look at both seroconversion and respiratory PCR sample assessments.

It should be noted that asymptomatic people may shed virus at high titers. Such serosurveys or other longitudinal population assessments are in progress. Analyses in nursing homes and cruise ships have noted the significant number of asymptomatic people who had infectious virus culturable up to 6 days before the onset of symptoms.

If the test sensitivity is poor in the presymptomatic phase of viral shedding, how helpful is such a test? For instance, if a patient is tested before a procedure or elective surgery, but it's a false-negative, that's not very helpful.

Surveillance testing by PCR has been adopted in many institutions for presurgical evaluation and in other instances. In an asymptomatic individual, a positive assay doesn't necessarily mean that a person is infectious. Also, rates of false-negative testing are likely higher in asymptomatic individuals. For this reason, some advocate for respiratory droplet precautions on all patients regardless, given this pandemic era.

Do antibody tests tell us anything other than that a person has been exposed and developed an immune response (eg, does it tell us anything about immunity, or how long that immunity would last)?

We know from experimental human challenge infections with routine coronaviruses in the 1990s that someone who recovered from infection and had demonstrable immunity could be reinfected with the coronavirus one year later. For SARS-CoV-1, on average, antibodies were found to be sufficient for 2 years, but during year 3 and later, were thought possibly low enough to reacquire infection. The duration of SARS-CoV-2 immunity is not yet known, as it is still less than 6 months from the first described cases.

When can you declare that a patient has been cured?

There are two different concepts to handle. First, is a patient infectious. The CDC changed its criteria in late April 2020 to also include a symptom-based strategy to determine whether patients can transmit the virus.

Symptomatic patients with COVID-19 should continue to take precautions until either:

  • 1) Symptom-based strategy — At least 3 days have passed since recovery, defined as resolution of fever without the use of fever-reducing medications and improvement in respiratory symptoms, and at least 10 days have passed since symptoms first appeared


  • 2) Test-based strategy — There is resolution of fever without the use of fever-reducing medications and improvement in respiratory symptoms, and negative results of an FDA Emergency Use Authorized COVID-19 molecular assay for detecting SARS-CoV-2 RNA from at least two consecutive respiratory specimens collected ≥24 hours apart (total of two negative specimens).

Patients with asymptomatic laboratory-confirmed COVID-19 should continue to take precautions until either 10 days have passed since their first positive COVID-19 diagnostic test, assuming they haven't developed symptoms since, or negative results of an FDA-authorized COVID-19 molecular assay from at least two consecutive respiratory specimens collected ≥24 hours apart (total of two negative specimens).

As mentioned earlier, there have been reports of prolonged detection of RNA without direct correlation to viral culture. Detecting viral RNA via PCR does not necessarily mean that infectious virus is present.

Regarding cure, as we are learning with multisystem inflammatory syndrome in children (MIS-C) and reports of patients with unexplained fever developing 3 to 6 weeks after initial recovery from COVID-19, there may be immunological issues causing health problems well after the host immune response has made the virus nonviable. Luckily, this seems to afflict only a tiny percentage of children and adults, but the factors leading to such problems remain poorly understood.

Editor's note: This story was updated to correct a statement on commercially available multiple PCR panels. The BioFire® Respiratory 2.1 panel includes the detection of SARS-CoV-2, and is authorized under the FDA Emergency Use Act.

Paul G. Auwaerter, MD, is a professor of medicine at the Johns Hopkins University School of Medicine and clinical director of the Division of Infectious Diseases.

For more news, follow Medscape on Facebook, Twitter, Instagram, and YouTube.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.