COVID Testing

Whom to TestCopy Link!

Tool: PRIoritize_Dx, intended to help policy-makers allocate tests.
Tool: PATH COVID-19 Diagnostics Dashboard to support product selection and procurement decisions.

Tool: The Foundation for Innovative New Diagnostics (FIND), a global non-profit, conducted Independent Evaluations of test kits between April and August of 2020.

Testing Symptomatic PatientsCopy Link!

Updated Date: December 19, 2020

Prioritize testing people with symptoms suggesting acute infection (see Screening Questions and Common Symptoms).

  • The standard of care for diagnostic testing symptomatic patients is PCR-based testing, as it has the best sensitivity and specificity.
  • If testing is very limited, prioritize testing patients when it will specifically change management or isolation status/location.
  • When PCR-based testing capacity is restricted, use of the antigen test can increase testing capacity as well as offer advantages in terms of more low-cost testing with short turnaround time. The antigen test is discussed in more detail in Types of Test Section.
  • Exact testing algorithms will depend on each institution and the availability and type of testing. One potential algorithm is presented below based on whether same day testing is available or not. Same day testing has advantages.

Algorithm for Symptomatic Patients Based on Whether Same Day Testing is AvailableCopy Link!

Details for each path are discussed below:

This diagram outlines in flow-chart format content that is covered in the list below.

  1. If same-day testing is available:
  1. Fast turnaround (same-day) Nucleic Acid Amplification Test (PCR)
  1. If positive, the patient is “confirmed COVID.”
  2. If negative, categorize according to Case Definitions and clinical suspicion.
  1. If discharging home, Isolate at home and initiate contact tracing.
  2. If admitting, repeat testing in 12-24 hours.
  1. If negative on the second test:
  1. Consider alternative etiologies (influenza, malaria, other infections), and discontinue the “suspected” or “probable” Case Definition if one is found.
  2. Consider alternative testing (e.g. serology) or repeat testing (typically 72 hours from first) if clinical suspicion remains high.
  3. Consider downgrading the case to “suspected” from “probable” depending on clinical suspicion.
  1. Rapid Antigen Testing (antigen rapid diagnostic test or Ag RDT for short).
  1. If positive, admit as a “confirmed” case or isolate at home, initiate contact tracing. If in a high prevalence setting or in a symptomatic patient with likely COVID-19, confirmatory PCR testing is not needed.
  2. If negative, treat as a “suspected” or “probable” case based on Case Definitions and clinical suspicion.
  1. If discharging, isolate at home and follow-up with a call or visit, homevisit or clinic visit. Consider retesting 2 to 4 days later. Consider contact tracing if suspicion is high for COVID-19.
  2. If admitting, request PCR testing and if not available, repeat rapid antigen testing in 2-4 days (see ECDC report).
  1. If no same-day testing is available (testing is located offsite and/or turnaround times are long):
  1. Send the specimen to the facility with the fastest reliable turnaround times.
  2. Follow-up and retesting strategy:
  1. If discharging to home, isolate at home and call or arrange a visit to share results. If the test returns negative, consider retesting, especially if symptoms persist or worsen.Also, consider retesting if it is deemed important to understand whether the case is COVID-19 and doing so would lead to important contact tracing activities.
  2. If admitting, triage as “suspected” or “probable” case based on Case Definitions and clinical suspicion for disease. If the test returns negative, retest with PCR testing.
  1. If no testing at all is available:
  1. If no testing is available, use clinical judgement and risk factors to determine likelihood of COVID infection and treatment plan. Case Definitions can help. Consider other lab testing to help stratify if available, including lymphocyte count, LFTs, and C-reactive protein. Err on the side of isolation.

Testing Asymptomatic PatientsCopy Link!

Updated Date: December 19, 2020

Asymptomatic Patients with an ExposureCopy Link!

If test capacity permits, testing asymptomatic people with known exposure to COVID-19 may be helpful (ideally as a part of a contact tracing initiative).

Generally we recommend testing patients who meet criteria for an exposure as soon as possible when they become aware of the exposure and again 5-7 days after the first test. This is because initial tests are often negative early in disease. Please note that even if an initial test is negative, the patient must still quarantine until they meet criteria for release from quarantine (in selected cases, testing may be used to reduce quarantine duration). Some healthcare systems recommend against testing in these instances either because of limited testing resources, or because the concerns about false reassurance from early false negatives.

Asymptomatic Screening and Public Health SurveillanceCopy Link!

Literature Review (Not Comprehensive): Gallery View, Grid View

To understand population-level prevalence and incidence, local institutions or departments of health may perform testing (PCR or Antibody) on entire cohorts regardless of exposure or symptoms. Details on design of epidemiologic, surveillance, or infection control studies are beyond the scope of this site.

Asymptomatic people who have a high likelihood of transmission to others should they become infected may be regularly tested for COVID-19, even without a confirmed contact. This is especially important if they spend time with persons that are at risk of complications from COVID-19, the classic example is periodically testing both residents and staff of nursing homes. This is sometimes called “asymptomatic screening” or “expanded screening”.

  • Common high-risk groups that may be considered for screening:
  • Health care workers, particularly those caring for patients with COVID-19 or in high patient-flow areas
  • People living or working in congregate living settings (nursing homes, dormitories)
  • Travelers coming from high prevalence areas
  • Teachers and students
  • Other essential employees (grocery workers, sanitation workers etc).

Asymptomatic Screening Frequencies by Prevalence Indicators:

Community Spread Level





New Cases per 100,000 Persons in Last 7 days

< 10

10 to 50


> 100

Percentage of Tests that are Positive in Last 7 Days

< 5%

5% to 7.9%

8% – 10%

> 10.1 %

Frequency of Asymptomatic Screening

Focus on High-exposure People Weekly


Weekly or Twice a Week

Twice a Week or More

Impact of Vaccination on TestingCopy Link!

Neither RNA- nor protein-based vaccines should have any impact on nucleic-acid based tests (NAAT) or rapid antigen tests used to diagnose COVID-19. Some antibody tests could conceivably turn positive after vaccination in rare combinations where a vaccine and test use the same viral antigens, but this is not yet verified for this purpose. Most serologic tests look for antibodies to the nucleocapsid and not spike protein, and thus do not detect vaccine-induced andibodies. As of February 2021, vaccination status should not be routinely considered in interpreting any COVID-19 test results. (CDC)

Newly symptomatic patients who have been vaccinated should still be tested, as breakthrough cases still do occur. Exposed vaccinated people are covered here (generally not needed).

Testing Previously Infected PatientsCopy Link!

Updated Date: April 23, 2021

Patients >90 days from initial illness with new symptoms of COVID-19 can be retested on a case-by-case basis. Reinfection is rare, but not impossible. Generally, patients who are <90 days from initial illness are not tested, however with the emergence of new variants this may be changing: a patient who had fully recovered and then develops new symptoms should be considered potentially re-infected with a new variant and should be re-tested.

  • If the patient tests positive, keep in mind the possibility of residual viral RNA even 90 days after initial infection, and consider alternate causes of illness (pulmonary embolism, bacterial superinfection) and (where possible) viral sequencing.
  • Recurrence of symptoms along with reemergence of positive PCR testing (particularly in patients with weakened immune systems) can occur in the absence of true reinfection.

Types of TestsCopy Link!

As the global pandemic grows, suspect cases should be immediately isolated regardless of test status.


  • No currently available test fully rules out the diagnosis, especially if clinical suspicion is high. When possible, negative or positive results that are inconsistent with the clinical pictures should be discussed with someone who has expertise in diagnostic testing of COVID-19
  • None of the commercially available tests measures active, infectious virus. Whether a person who tests positive is infectious requires clinical judgment and knowledge of their disease course.
  • In all cases, please follow local public health authority guidelines in reporting all suspected, presumed, and confirmed cases of COVID-19.

Testing OverviewCopy Link!

Updated Date: December 19, 2020

In a pandemic, clinically suspected cases should initially be isolated regardless of test status. See Screening, Case Definitions, and Isolation. All tests have both false positives and false negatives. A high index of suspicion should be used to protect staff and other patients.

The three most clinically relevant categories of testing for COVID-19 are:

  1. Nucleic Acid Amplification Test (NAAT): While not a perfect test, NAAT is considered the gold standard. The test uses enzymes to amplify and detect the genetic material of the virus. The most familiar is RT-qPCR (reverse transcriptase - quantitative polymerase chain reaction; related versions are often referred to as PCR or RT-PCR), but there are others. These are often collectively called “molecular tests.”
  2. Antigen Rapid Diagnostic Test (RDT): Requires less time and infrastructure to perform than NAAT tests. Uses manufactured antibodies to detect SARS-CoV-2 proteins.
  3. Antibody (IgM/IgG) RDT: Has different uses and interpretation than the nucleic acid and antigen tests. Uses manufactured antigens to detect a patient’s antibodies to SARS-CoV-2. Since this depends on the body’s immune response, it takes longer to turn positive than tests that directly detect the virus, and a negative antibody test DOES NOT rule out acute infection. The antibody test is NOT used as the sole test to diagnose active or contagious disease; it is more common in epidemiology and research. The test can be used to support the diagnosis in COVID-19 in patients that present late with symptoms (at least 8 days after the onset of symptoms) or to help assess whether a symptom or sequelae is due to a post-COVID-19 infection.

In these lists and elsewhere, tests that measure nucleic acids are often referred to as molecular assays, while antigen and antibody tests are considered immunoassays or serological assays.

General test characteristics are summarized in the table below:

RT-qPCR (or other NAAT)

Antigen (Ag) RDT

Antibody (IgM/IgG) RDT


Nasopharyngeal, oropharyngeal, saliva, lower respiratory

Nasopharyngeal, oropharyngeal, saliva, lower respiratory


False Positives

Rare, except for cases of sample contamination. However, can remain positive after virus is no longer viable.

Very low

Low to moderate, most commonly due to cross-reactivity with other coronaviruses.

False Negatives

Occasional, especially early in infection.

Moderate, not as sensitive as NAAT.

Variable. Performs poorly at the onset of the symptoms.

Turnaround time/ Laboratory Requirements

Usually hours - requires a laboratory with high technical capacity.

Under 30 min – no laboratory required.

Under 30 min – no laboratory required.

Specific data on the availability and performance of commercial COVID-19 diagnostic testing options continues to change rapidly.

Tool: PATH COVID-19 Diagnostics Dashboard to support product selection and procurement decisions. Collects data from U.S. FDA, WHO, FIND, and other lists curated by private entities. Includes information on regional or national regulatory approval.
Tool: The Foundation for Innovative New Diagnostics (FIND), a global non-profit, conducted Independent Evaluations of test kits between April and August of 2020.
Tool: The Cochrane Library, a resource by the international charitable organization Cochrane, has published independent reviews of Molecular/Antigen tests (updated 2020-08-26) and Antibody Tests (Updated 2020-06-25).

Timeframe of Test PositivityCopy Link!

Updated Date: December 19, 2020

The relative time frames of exposure, symptoms, viral markers, and antibodies are illustrated in the subsequent figure. This is an illustration of average time frames and it should be noted that information is still emerging on the timeframes of Incubation and Window Period, Infectivity, and Durable Immunity.

This diagram indicates the timeline at which different tests become positive, which is also covered in the below bullet points.

  • Infectious Period: On average, the person is most infectious 2 days prior to the onset of symptoms to about 5 days after the onset. This is referred to as the pre-symptomatic and early symptomatic time frame. In general, people are no longer infectious 10 days after symptom onset (20 days in severely ill persons, see Infectivity).
  • The Antigen Test is likely to be positive at the same period that a person is most infectious from about 2 days prior to the onset of symptoms to about 2-3 days after the onset. This is because the viral load is highest in this time period.
  • The RT-PCR test (and other NAATs) is more sensitive than antigen testing, it can (but does not always) pick up cases earlier than even 3 or 4 days before the onset of symptoms or 10+ days after symptom onset. These are average time frames for when the test is most likely to yield a positive result.
  • The antibody test in some cases turns positive after the patient may no longer be infectious. For this reason, the antibody test is not typically used to diagnose active disease.

Nucleic Acid Amplification TestsCopy Link!

Updated Date: December 19, 2020

Literature Review (not comprehensive): Gallery View, Grid View

These tests work by amplifying minute amounts of viral RNA. PCR (technically, RT-qPCR) of a nasopharyngeal swab specimen is most widely used and should be considered the standard of care when available.

How it WorksCopy Link!

Reverse-transcriptase quantitative polymerase chain reaction (RT-qPCR) works by reverse transcribing the viral RNA genome to DNA, and then amplifying the DNA exponentially by repeatedly cycling the reaction. Samples with a small amount of virus require more cycles to reach a detection threshold than samples with a large amount of virus, allowing some RT-qPCR tests to quantify how much virus was present and approximate viral load.

Newer amplification technologies are being developed to make this process cheaper, faster, and less dependent on complex laboratory infrastructure. The term nucleic acid amplification test (NAAT) includes PCR, isothermal amplification, and CRISPR-based tests (Behera et al., Kilic et al). Xpert® Xpress SARS-CoV-2 is an example of an automated cartridge-based system on the same Xpert® machines used for diagnosis of tuberculosis that does not need a sophisticated laboratory setting. All have the common feature of detecting minute quantities of SARS-CoV-2 nucleic acids.

The main advantages to nucleic acid testing are:

  • The amplification steps allow detection of very small amounts of viral RNA, with higher specificity than serological assays.
  • Live virus is not required. Samples can be inactivated and made safe to handle.
  • Nucleic acid amplification is common in biology research. Government labs, universities, and highly-resourced clinical labs may be able to rapidly deploy new or modified tests before commercial kits are available. This may become relevant if a new strain or resistant mutation emerges.

The main disadvantages to nucleic acid testing are:

  • Most methods require significant infrastructure, usually including custom machines, reliable electricity, cold storage, a supply chain for reagents, and skilled personnel.
  • Test performance can depend heavily on how and when a sample is taken. For example, lower respiratory samples can be positive when upper respiratory samples are not.
  • Fragments of viral nucleic acids can persist in the body long after the virus has been killed. Persistently positive tests may not mean that a person is still infectious.
  • Variant strains of SARS-CoV-2 may affect the sensitivity of some nucleic acid-based tests, resulting in false negatives. Tests that use multiple genetic targets are less likely to be affected by the increasing diversity of SARS-CoV-2. As of February 2021, U.S. FDA advice is to consider negative results in clinical context and to consider repeating testing with a different test if clinical suspicion remains after an initial negative test.

Test PerformanceCopy Link!

Sensitivity and Specificity:

On artificial samples, NAAT sensitivity and specificity approaches 100% (Giri). NAAT tests have analytical sensitivity (the lowest viral concentration where >95% tests are positive; also called limit of detection) down to 100-5000 copies of viral RNA/ml. But published real-world estimates of sensitivities for different NAATs to diagnose COVID-19 range from ~60-95%, depending in part on what reference method is used as a comparator or “gold standard.” Specificity is excellent, and false positives are rare in the absence of contamination, though there is increasing recognition that false positives due to contamination or cross-reaction with other genetic material do occur and may have significant consequences (Surkova et al)

Interpreting reported clinical sensitivities involves a range of factors:

  1. There is significant variability in how studies define a "true positive" or "gold standard":
  1. The reference standard can be "composite," including laboratory, radiographic, and clinical data. A single center study at University of Kong Kong-Shenzhen, China, compared initial RT-qPCR in 82 patients to a a retrospective diagnosis made by combining serial RT-qPCR and chest CT findings, with a resulting sensitivity of 79% and a specificity of 100% (He et al).
  2. The reference standard is often serial NAAT. since this is often the only data available for large numbers of patients. A retrospective analysis of over 20000 patients used repeat PCR within seven days (felt too short for interim infection to play a large role) and found that only 3.5% of patients initially negative by PCR subsequently tested positive. This suggested a low false negative rate, but they pointed out that this was not a true clinical sensitivity since they lacked a final confirmatory diagnosis (Long et al).
  3. The reference standard can be other previously validated NAAT’s. FindDx reports that the tests they validated agreed 92-100% when comparing to their a reference PCR assay, demonstrating small but real variability between tests.
  1. The site of sampling might not contain virus at the time of sampling. See Sample Collection.
  2. Laboratory factors (sample storage, frequency of contamination).

These same factors should be systematically considered when a clinician suspects a false negative:

  1. Exactly what other data makes me feel like the patient has COVID-19?,
  2. Do we need to repeat the sample or collect another sample type?, and
  3. Could there have been a lab error?

Typical clinical use:

If you do not know your test’s characteristics, sensitivity of ~80% may be a reasonable approximation for nasopharyngeal swabs collected at the time of patient presentation, assuming no laboratory errors.

  • If the RT-PCR is negative but suspicion for COVID-19 remains, then ongoing isolation and re-sampling several days later should be considered.
  • In practice, test results should be interpreted based on negative and positive predictive values (NPV, PPV) rather than sensitivity and specificity, since these incorporate pretest probability.

Sample CollectionCopy Link!

Literature Review (not comprehensive): Gallery View, Grid View

Upper Respiratory Tract Specimens. Most commercial kits have been evaluated on specific upper respiratory sample types. Of these, the nasopharyngeal swab is the most common and best validated; in most situations this is the best option unless a specific manufacturer recommends otherwise or there is a clinical reason to choose an alternative site.

Sites include:

  • Nasopharyngeal Swab
  • Nasopharyngeal Wash/Aspirate
  • Oropharyngeal Swabs
  • Mid-turbinate and Anterior Nasal Swabs
  • A few manufacturers allow these to be collected by the patient at-home (unsupervised) or supervised by a provider at a safe distance resulting in less risk to the HCW.
  • Saliva
  • Potential to significantly simplify sample collection cost and complexity
  • Many new platforms being developed, but still limited comparative performance data (Wyllie et al)

Tool: U.S. CDC Collection Protocol

Tool: Video Demonstration

Lower Respiratory Tract Specimens are also sometimes used, though they often require different processing and validation due to the presence of mucus. When obtaining lower respiratory tract specimens, many sampling techniques require airborne precautions for providers (see Aerosol Generating Procedures).

Sites Include:

  • Expectorated Deep Sputum (similar to sputum collected for TB testing in patients with productive cough).
  • Bronchoalveolar Lavage
  • Endotracheal Aspirates
  • Preferred in intubated patients due to higher sensitivity, though this depends on sample quality Like any respiratory sample, high quality samples are characterized by Gram stains with many polymorphonuclear cells and few epithelial cells.

The relative performance of testing different sample types, optimal timing for sample collection relative to exposure or symptoms, and the interpretation of discordant results (for example, if the nasopharynx is negative but sputum is positive), all continue to be studied.

U.S. CDC guidelines for processing of sputum specimens for SARS-CoV-2 RT-PCR recommend the use of dithiothreitol (DTT) for liquefaction of viscous mucoid/mucopurulent material prior to nucleic acid extraction

Tool: CDC Specimen Processing

Other Specimens: Viral RNA has been documented in other body sites including stool and rarely blood, but it is not known whether this represents transmissible virus (Wang et al). Testing samples from these sites requires extra laboratory expertise for sample handling and clinical expertise for interpretation. These sites should not be tested routinely.

Antigen Rapid Diagnostic TestsCopy Link!

Updated Date: December 19, 2020
Literature Review (not comprehensive):
Gallery View, Grid View

How it WorksCopy Link!

This is a diagram showing the function of a lateral flow assay: sample on left, labeled antibody to a virus moves along the test to a "test line" (another antibody to the virus) and a control line (antibody to antibodies).

Rapid Diagnostic Tests (RDT’s) for viral antigens use premade, labeled antibodies to the virus to capture viral particles. The most common approach is the lateral flow test, where sample diffuses along a manufactured strip in a way that can be visually detected at the “test line” only if viral antigens are present. Note that a few manufacturers do make “rapid” NAAT assays. This discussion does not address those tests.

The main advantages to antigen RDT’s are:

  • Running a test involves adding the sample (and sometimes a single liquid reagent) and waiting for diffusion.
  • They generally do not require special trainings or machinery to run, and many are licensed to be run outside of a laboratory setting (e.g. CLIA waiver)
  • They are typically fast, cheap, and have a simple visual yes/no readout that does not require interpretation.

The main disadvantages to antigen testing are:

  • They are less sensitive than NAATs, since there is no amplification step. Negative tests may need confirmation with NAAT if clinical suspicion is high.
  • As with NAATs, performance can depend heavily on how and when a sample is taken.
  • Fragments of viral proteins can persist after the virus has been killed, though likely not as long as nucleic acids.
  • If future viral mutations change the antigen region targeted by a particular test, it will take longer to create new antigen RDT’s (which involves new manufacturing) than it would to modify most NAAT’s (which involves new reagents only).
  • Consult the manufacturer’s insert for specimen collection requirements; many are designed for nasopharyngeal swab only.

Test PerformanceCopy Link!

Clinical sensitivity for antigen RDT’s is highly variable. The average of sensitivity was 56% for four antigen RDT’s reviewed in an August 2020 Cochrane review, with 95% confidence interval of ~30-80%. The same review found much higher average specificities of 99.5% (95% confidence interval 98.1-99.9%) (Dinnes et al). The sensitivity may vary based on symptomatology, with some performing as poorly as 32% sensitive (Quidel EUA), and others as high as 79% (Alemany et al) in asymptomatic people. However, many of the cases that these tests miss may not be infectious, but this is still an area of active research.

Finding the real-world sensitivity of antigen RDT’s suffers from many of the same difficulties discussed in the NAAT section above, though these are usually compared with NAAT as a gold-standard. The minimum performance requirements for Ag-RDT set by the WHO are >80% sensitivity and >97% specificity compared to a NAAT reference assay (WHO)

Use of RDTs for ScreeningCopy Link!

Rapid Antigen RDTs are an alternative to NAAT as screening tests where testing capacity is limited and the proportion of test positivity is high (≥10%) (ECDC Recommendation). Positive and negative predictive values (PPV and NPV) of all in vitro diagnostic tests depend on disease prevalence in the target population and the test performance.

In a high prevalence setting CDC considers high prevalence to be when NAAT positivity over the last 14 days is greater than 5% or when there are greater than 20 new cases of COVID-19 per 100,000 persons within the last 14 days., rapid antigen tests will have a high PPV, meaning a positive result from a rapid antigen test is likely to indicate a true infection and may not require confirmation by RT-PCR. In contrast, any negative test result should be confirmed by RT-PCR immediately or with another rapid antigen test a few days later (where RT-PCR is very limited).

In a low prevalence setting CDC considers low prevalence to be when NAAT positivity over the last 14 days is less than 5% or when there are fewer than 20 new cases of COVID-19 per 100,000 persons within the last 14 days., rapid antigen tests will have a high NPV but a low PPV. Therefore, a negative antigen test most likely represents a true negative and may not require confirmation by NAAT, however false negatives are still possible and people being tested should be reminded that they should still take all precautions to prevent spread (e.g. masking, distancing, etc). In this situation, a negative test result may not require confirmation by NAAT, whereas a positive test will need confirmation by NAAT.

Scenarios where Antigen RDT Can be Used for Screening of asymptomatic individuals: (modified from WHO and ECDC). (For use of Antigen RDT for symptomatic individuals, including contacts, see testing symptomatic patients).

Scenarios for use of SARS-CoV-2 Ag-RDT

Populations Where RDT Can Be Used For Screening where NAAT testing is limited

Negative testing should NEVER exempt people from standard transmission prevention practices (masks, distancing, hand washing). See IPC.

Outbreak Response

To respond to suspected outbreaks of COVID-19 in remote settings, institutions and semi-closed communities

Outbreak Investigation

To support outbreak investigations (e.g. in closed or semi-closed groups including schools, care-homes, cruise ships, prisons, workplaces and dormitories, etc.)

Monitor Trends in Disease Incidence

To monitor trends in disease incidence in communities, and particularly among essential workers and health workers in regions of widespread community transmission

Community Transmission Screening for Congregate Settings

Where there is widespread community transmission, RDTs may be used for early detection and isolation of positive cases in health facilities, COVID-19 testing centers/sites, care homes, prisons, schools, front-line and health-care workers.

Testing of Asymptomatic Contacts/Contact Tracing

Testing of asymptomatic contacts of cases (either as part of outbreak investigations or household contacts) may be considered even if the Ag-RDT is not specifically authorized for this use. However, given the high pre-test probability in this population, a negative test often does not rule out infection (has a low negative predictive value) and where possible should be confirmed by NAAT or repeat RDT as described above. Even in the setting of a negative test, contacts should continue to remain in quarantine until they meet criteria to discontinue quarantine.

Antibody TestingCopy Link!

Updated Date: December 19, 2020
Literature Review (not comprehensive):
Gallery View, Grid View

How it WorksCopy Link!

Antibody tests measure the host adaptive immune response, rather than the presence of the virus. This test is most often performed on circulating blood (from fingerstick or blood draw).

Adaptive immune response requires several days to make antibodies that bind to the pathogen. See Antibody Response and Durable Immunity for a more in-depth discussion of antibody patterns over time. It is not yet known what impact antibodies have on the risk of transmission to others or risk of re-infection.

The main advantages to antibody testing are:

  • Respiratory samples are not required. Antibody testing can be done on blood drawn for other reasons in clinical care. There are versions to test dried blood spots.
  • IgG may last for months to years, so it is useful in epidemiology to know who has been previously infected (even if they were asymptomatic).
  • A strategy that uses both NAAT and serology may improve sensitivity for COVID-19 over using NAAT alone (Guo et al).

The main disadvantages (and why antibody testing alone is not recommended to guide clinical decision making) are:

  • Antibodies take several days for the human body to develop, so antibody testing is often negative in early infection; this is known as the “Window Period.” In fact, a positive IgG argues against acute early infection.
  • False positives can occur due in patients who have been exposed to coronaviruses other than SARS-CoV-2, including some types of the common cold.
  • IgM is often less specific than IgG, so false positives may be more common for IgM results, making the test less accurate for acute infection.
  • Although antibody RDT’s using principles of lateral flow are available, the most sensitive versions require laboratory infrastructure for techniques such as ELISA.
  • The immune system simultaneously makes many different antibodies, but test manufacturers choose a single antibody to detect. This can result in increased variability between test performance from different manufacturers.
  • For these reasons, antibody testing alone is not recommended to guide clinical decision making.
  • Prior vaccination for COVID-19 is unlikely to affect the interpretation of antibody testing in most circumstances, though this is still being studied (CDC).

Test PerformanceCopy Link!

Combined IgM/IgG testing has low sensitivity early in infection (30%) but reaches 91% by 15-21 days after onset of symptoms, in a June 2020 Cochrane Review summarizing 54 cohorts with a total of nearly 16000 patients. The same review found a high average specificity of ~98% (Deeks et al).

InterpretationCopy Link!






  • No serological evidence of infection with COVID-19.
  • Potentially in the “window period” before antibodies have developed
  • Also might be a weak, late or absent antibody response, particularly in older patients, those with poor nutritional status or immunodeficiency, and rarely in severe COVID-19 disease.



  • Potentially early infection, before IgG is detectable.
  • Also might be a false-positive IgM (cross-reaction to other coronaviruses).
  • IgM is often less specific than IgG, so false positives from other viruses may be more common in this case.



  • Likely either late or resolved infection.
  • Also might be a false-positive IgG (cross-reaction to other coronaviruses)..



  • Potentially active infection.
  • Also might be late or recovery phase of the disease, before IgM has declined.
  • Also possibly a false-positive resulting from cross-reaction with other coronaviruses.

Viral CultureCopy Link!

Updated Date: December 19, 2020

Viral culture is not generally used in clinical settings. Availability is very limited, since safe viral culture requires laboratories with advanced biosafety capabilities (typically BSL 3 in the USA). It is the most definitive test for the presence of viable virus, since both antigens and RNA can persist even after the virus is “killed.”

  • It is often used in research settings to tell which types of samples are potentially infectious.
  • It can be used to confirm that patient or pharmaceutical antibodies neutralize viral replication, since some antibodies might bind to the virus without inhibiting replication.

Viral SequencingCopy Link!

Updated Date: December 19, 2020
Literature Review (Novel Diagnostics):
Gallery View, Grid View

Full-genome viral sequencing is not generally useful in the acute clinical setting. When available, viral genomic sequencing from patient samples can be used for local outbreak tracing, assessing re-infection, and large-scale epidemiology. Sequencing may also be used in the future to look for mutations and decreased responsiveness to vaccines or therapeutics, though this will require significantly improved understanding of SARS-CoV-2 biology.

Several groups are developing technologies to reduce the hardware and infrastructure investment needed and finding innovative applications that may eventually impact front-line workers (Khatib et al).