AAMC Guide to COVID-19 Tests and Testing
Published October 4, 2020
Months after the onset of the COVID-19 pandemic, we lack a solid national testing strategy with specific targets for the number and types of tests that should be conducted each day. In the AAMC’s Recommendations for COVID-19 Testing, the AAMC’s Research and Action Institute addresses the most common questions about the different types of tests for COVID-19 and when each type should be used. This testing guide was authored by AAMC’s Heather Pierce, JD, MPH, Senior Director, Science Policy and Regulatory Counsel, Ross McKinney, Jr, MD, Chief Scientific Officer, and Atul Grover, MD, PhD, Executive Director, AAMC’s Research and Action Institute.
- What are COVID-19 tests?
- What are the different COVID-19 test categories?
- What do the sensitivity and specificity of tests mean?
- How are COVID-19 tests used?
- Is testing alone sufficient to control the spread of the virus?
- Which types of tests are used for which purposes?
- When do surveillance testing plans need additional testing capacity?
- What are pooled tests?
- Does the FDA approve tests?
What are COVID-19 tests?
The phrase “COVID test” has been used as shorthand to describe all types of testing related to COVID-19, the disease that results from infection with the SARS-CoV-2 virus. Tests to determine if a person has COVID-19 are designed to detect the presence of the virus.
In general, there are two broad categories of tests related to COVID-19: those that detect elements of the virus in order to identify an active infection (tests for current infection) and those that detect antibodies in the blood that indicate the person has already had an infection and an immune response (tests for prior infection). Tests that detect viral material are useful to identify who is currently infected and thus potentially contagious.
What are the different COVID-19 test categories?
Tests for current infection
- Reverse-transcription polymerase chain reaction tests, or RT-PCR or PCR tests, extract small pieces of the virus’s genes (RNA) from a biological sample like saliva or material from a nasopharyngeal swab and then make many copies of those genes (called “amplifying” the virus’s RNA). If the virus is present, this amplification will make it detectable.
Widely considered the “gold standard” for COVID-19 testing, RT-PCR tests are highly sensitive and the most commonly used type of diagnostic test in the United States. The complexity of these tests means they can only be run in sophisticated labs with proven success in clinical testing and in minimizing contamination during the PCR processing. The multistep process requires specialized equipment and chemicals known as reagents, which has made RT-PCR tests vulnerable to interruptions in the supply chain for testing supplies. They are also the most expensive tests to run.
- Antigen tests detect proteins that make up the SARS-CoV-2 virus. While antigen tests are less sensitive than RT-PCR tests and therefore may miss more infected people, the ease of manufacturing, low cost, and relative simplicity of running the tests give antigen detection advantages in increasing nationwide testing capacity.
Antigen tests are currently authorized with nasopharyngeal or nasal swabs and only for use in individuals who are suspected to have COVID-19 and whose symptoms have begun in the previous several days. Testing in asymptomatic individuals using antigen tests has been proposed but requires additional study.
Unlike an RT-PCR test in which viral material is increased to make it easier to detect, the antigen test can only detect the virus if the protein is in high enough quantity in the original biological sample. Because of this limitation, samples from individuals who are infected and contagious but do not have as much of the virus in their noses may not result in a positive test.
Tests for prior infection
- Antibody tests detect the presence of an immune response in the blood of a person who was previously infected with SARS-CoV-2. Antibodies begin to appear in the blood during infection but cannot be detected until the amount of virus starts to wane. Although these tests are important in better understanding the immune response to infection and whether previous infection will prevent or reduce the impact of a second infection, antibody tests are not useful indicators of current infection. The virus appears most contagious in the early stages of infection, generally well before antibodies against the virus can be detectable in blood.
What do the sensitivity and specificity of tests mean?
Tests for SARS-CoV-2 are assessed on their sensitivity and specificity. A test with high sensitivity is more likely to detect the virus when it is present. When tested with a high sensitivity test, a person infected with SARS-CoV-2 will be very likely to have a positive result. High sensitivity tests have few false negatives. A test with high specificity indicates that when a person is not infected, they are unlikely to have a positive test result. High specificity tests have fewer false positives, especially in a population with a large number of uninfected people.
In general, the types of SARS-CoV-2 tests for current infection differ in sensitivity. RT-PCR tests are more sensitive than antigen tests, which means that RT-PCR tests will identify more individuals who have the virus than antigen tests. Differences in published sensitivity rates may also reflect how a test is used in practice, including the sample collection process (e.g., nasopharyngeal swab, nasal swab, saliva).
How are COVID-19 tests used?
The RT-PCR, antigen, and other tests that are being developed to detect the presence of the virus can be used for diagnostic, screening, or surveillance purposes. Diagnostic tests are focused on the experience, needs, and history of a single individual and are often used for those with COVID-19 symptoms.
Diagnostic tests may also be used to detect the virus in people without symptoms (asymptomatic individuals), typically after exposure to an infected individual. When applied to a population without regard to exposure history, the tests are considered screening or surveillance tests, which are essential tools in controlling the virus’s spread. The Food and Drug Administration (FDA) has issued standards for when each of these terms should be used and the acceptable levels of sensitivity and specificity for each.
- Diagnostic tests are intended to determine with a high level of confidence whether an individual is currently infected with the virus. These tests could be used for individuals who are suspected of having COVID-19 because of their symptoms, asymptomatic individuals who have been exposed to a confirmed case of COVID-19, or individuals in particularly high-risk groups who have participated in large gatherings. Diagnostic tests must demonstrate high sensitivity and high specificity, because the intent is to use those results to make treatment decisions or quarantine recommendations for individuals. Diagnostic tests may be ordered by a health care provider or obtained directly by an individual. Diagnostic tests must be run in a lab that has been certified by the Centers for Medicare & Medicaid Services to be able to run high-complexity tests.
- Screening tests are given to asymptomatic individuals for the purpose of making decisions based on that person’s test results. Examples of common uses of screening tests include testing everyone in a nursing home, testing students upon their arrival on a college campus, or requiring a negative test before admitting someone back to an office. In screening tests, all individuals are often told of their results and those individuals who test positive for the virus are typically asked to take additional steps to protect their health and those around them through quarantine or other actions.
- Surveillance testing refers to broad, typically nonidentified testing of populations to inform public health actions. Examples of surveillance testing include assaying wastewater or surfaces to detect presence of the virus or testing a large number of people and looking at aggregate results to determine the prevalence of the virus in a community. In true surveillance testing, there is usually no intention to return individual results to those tested.
Is testing alone sufficient to control the spread of the virus?
No. Testing is an essential resource to understand where the virus is and where it is spreading. Testing can only let individuals know if they have already been infected and are infectious so that they can change their behavior to protect other people around them. Any COVID-19 test will only tell you where the virus has been. There is a period of time between when a person is exposed to the virus and when there is enough of the virus in the body to detect in a test (sometimes called sufficient viral load). During that time, a person may have no symptoms and would have a negative COVID-19 test, but they would still be infected and potentially contagious to others.
To protect individuals from contracting COVID-19, testing is insufficient. Infection prevention measures require avoiding contact with those who are infectious, maintaining a distance from other potentially infected individuals, wearing face coverings, and washing hands.
Which types of tests are used for which purposes?
The type of test being administered does not necessarily indicate how the test is being used. For example, RT-PCR tests are often used for diagnostic testing because of their high reliability, but they take longer to run and require more resources. While the cost and complexity of running an RT-PCR test may make it a poor choice for broad surveillance testing, it may well be a reasonable approach for a targeted screening program.
The speed and low cost of rapid antigen testing may make it a useful diagnostic tool in health care facilities. In each of these examples, the test is designed to detect the presence of the virus and thus an active infection. Recognizing that a more sensitive test will provide more accurate information about an infection in one person, it is reasonable to limit the use of more expensive and complex tests for diagnostic testing, where an accurate result is needed to drive health care decisions.
As a practical matter, the terms screening and surveillance are often used interchangeably, as in “a surveillance testing program where each student on campus is tested once a week on a randomly assigned weekday.” This plan has elements of screening, where all students will be told about their results and those who test positive may be moved to quarantine housing and only attend classes remotely. Whether this describes screening or surveillance is more significant than a linguistic choice. The categorization of a test as screening versus surveillance may have implications for the type of test used, the level of regulation imposed on its use, where the test may be run, and how the test is paid for or reimbursed.
When do surveillance testing plans need additional testing capacity?
In most cases, surveillance testing is used to understand where the virus is present, not to help individuals make choices about their health or activities. However, there are ethical considerations in the administration of a surveillance testing effort where individuals are tested using less sensitive tests. Those tests may provide an institution with a good sense of the infection rate when administered to thousands or tens of thousands of people, but they could give individuals who receive negative results a false sense of security, allowing the virus to spread through social interactions when there are inevitable “false negatives.”
Surveillance testing plans need to account for the fact that any time people are tested, whether through a nasal swab or saliva sample, they expect to get results back and be able to take action based on what they learn.
In virtually all cases, when surveillance tests result in presumptive positive cases, those individuals will be identified and referred to a clinical lab for a confirmatory diagnostic test. However, individuals who do not receive a positive test result may engage in riskier behaviors on the assumption that they can rely on the test results, even when they are told that the test will return negative results in some cases where the person is infected.
What are pooled tests?
Pooled tests refer to the collection of biological samples from more than one person to evaluate using a single test. This method is most efficient and can conserve testing supplies when the rates of positive tests are relatively low. For example, a negative result for a single test on a pool of samples from five people would mean that none of those five people are currently infected.
Typically, a positive result from a pooled test will require additional confirmatory tests for each person in the pool to determine which ones are infected. Pooled samples may be used in different test types and for different purposes, but with COVID-19, the concept is generally applied to RT-PCR testing or other forms of gene detection. The fact that a test is from a pooled sample does not necessarily mean that the test is being used only for diagnostic purposes.
Does the FDA approve tests?
Commercial tests that identify SARS-CoV-2 are overseen by the FDA. However, in a public health emergency, the agency uses a process called Emergency Use Authorization (EUA) to rapidly move tests into use. The standard for releasing a new test through an EUA is lower than the standard for FDA approval. The FDA approval process can take months and requires the submission of studies to validate the test’s performance. There are no FDA-approved tests for SARS-CoV-2, but many have been granted an EUA for use. Therefore, disclaimers that a commercial test “has not been approved by the FDA” does not indicate that there has been no oversight by or accountability to the FDA. Commercial tests for SARS-CoV-2 must have an EUA or FDA approval to be marketed.
Laboratory developed tests, which are made in a single lab for use in that lab, are not required to have an EUA or approval from the FDA, but many such tests have gone through the EUA process.