There are several instances of COVID-19 tests yielding contradictory results. This article hopes to explain how and why this may occur. Throughout this article, I will refer to terms used in epidemiological screening for diseases. I have explained those terms in a previous article.

Available tests:

First, we will briefly look at the types of test available. Broadly, there are three types of tests that may be performed to detect COVID-19 infection: Reverse Transcription Polymerase Chain Reaction (RT-PCR), Antigen tests, and Serological tests.

RT-PCR

This test looks for viral RNA in samples. The advantage of this method is that if a very small amount of viral RNA is present, copies of the RNA can be made till the amount of material is sufficient for testing. This process is called amplification (Nucleic Acid Amplification to be precise).

A positive RT-PCR test indicates that viral RNA is present in the sample. It does not indicate that the virus is viable/ live.

The test requires a nasal or throat swab to be taken, and looks for viral RNA in the sample. The virus is present in the nose during the first week of disease, after which it is present in the lungs. Nasal samples obtained after the first week may fail to detect the virus, although it is present in the body (false negative).

Antigen Test

This type of test looks for specific proteins on the surface of the virus (surface proteins). Coronaviruses are several similar looking viruses. In order to be certain that one is dealing with SARS-CoV-2 (the specific coronavirus that causes COVID-19), antigen tests must detect surface proteins that are unique to SARS-CoV-2 (usually from the surface spikes (spike proteins)). The challenge is finding a unique spike protein. The test itself requires a nasal swab to be obtained from the patient/ suspect to look for antigen.

The main problem is that there may be insufficient antigen material in nasal swabs to be detectable- particularly in asymptomatic persons with no nasal discharge. Since there is no amplification, this may result in false negative results.

The sensitivity of antigen tests against similar viruses ranges from 34% to 80%.

Serological (Antibody) Tests

These tests do not look for the virus, but evidence of the immune system’s response to the virus (antibodies).

There are two types of antibodies tests detect: IgM antibodies, and IgG antibodies. In order to establish the utility of antibody tests, the tests need to undergo rigorous validation procedures. Further, the interpretation of test results depends on our understanding of the body’s immune response to SARS-CoV-2. At present, neither are there validated serological tests, nor is our understanding of the immune response to SARS-CoV-2 complete.

IgM antibodies

These antibodies are usually the first to appear following an infection. A positive result from an appropriately validated serology test that detects IgM is likely to indicate that someone currently has or has recently had the virus.

However, if an infected person has not developed antibodies at the time of testing, the test result will be negative (false negative)[This is more likely to happen early in the infection]. Similarly, the presence of antibodies to coronavirus strains other than SARS-CoV-2 may cause a positive test result (false positive).

Generally, tests to detect IgM antibodies are not used to rule out SARS-CoV-2 in a person.

IgG antibodies

These antibodies usually appear around 28 days after infection, but may vary considerably from person to person. Although these antibodies are more specific for COVID-19, they cannot be used to rule out infection. However, these antibodies are useful for serological surveillance.

Other concerns regarding antibodies are:

1. The timeline of appearance, peaking and disappearance of antibodies against SARS-CoV-2 is not established
2. It is unclear if the presence of antibodies indicates recovery
3. It is unclear if persons with SARS-CoV-2 antibodies who continue to shed the virus are infective
4. The duration of immunity (lifelong or shorter) is unknown
5. It is not known if immunity against one strain of SARS-CoV-2 protects against other strains of the virus.

The major features of the test options are shown in the table below

	RT-PCR	Antigen test	Serology
Detects	Viral RNA	Viral surface proteins	Antibodies (IgG, IgM)
Specimen	Nasal/ Oral	Nasal	Blood (Serum)
Time to obtain results	Up to 2 days	Up to 30 minutes	Minutes
Positive result indicates	Virus RNA present in sample- does not indicate if it is viable	Virus surface protein present in sample	Antibodies against virus present in blood (Past infection)
Precaution(s)	Nasal/throat swab reliable in 1st week of disease only	Test may be negative if amount of antigen material is low	Immune response not well/fully understood; no validated test available

Serological tests are not recommended as the sole test(s) for diagnosis of COVID-19.

The factors that may influence test results include:

Patient: The history provided by a patient is crucial. False negative test results may occur if the patient provides inaccurate information regarding

1. Onset and duration of symptoms (symptomatic patient)
2. Travel
3. Contact
4. Treatment (some treatments may interfere with tests)

This would be on account of delay in performing a test like RT-PCR.

Doctor: The choice and timing of test are crucial. Ideally, RT-PCR should be performed in the first 7 days of disease; testing for IgM antibodies after at least 5 days of illness; testing for IgG antibodies after the end of minimum 1 week.

When sending serological tests, ideally paired sera should be sent for testing.

Technician: Perhaps one of the most important factors is sample collection, storage and transportation.

• Anything that might dilute a serum sample (like drawing serum sample from an arm receiving intravenous fluids) will cause a false negative test result
• Sample collection to testing. Regardless of the test, one is dealing with live tissue. The viability of sample is affected by several factors:
  1. Storage temperature (the actual storage temperature must be verified to determine if it is in compliance with recommendations)
  2. Storage time
  3. Transport medium
  4. Transport temperature (the actual temperature within the transport container must be verified to comply with recommendations)
  5. Transportation time (should be within 2 hours)

Over time, antigen and antibody levels will decrease till they are no longer detectable.

• Very low volume of (nasal/throat) sample. This will render the ability to detect antigens ineffective. Sample volume is not an issue with RT-PCR, since amplification is possible.
• Source of sample: As mentioned earlier, nasal/throat samples are useful only in the first week of disease. The source is important for PCR, but not for serology.

Testing procedure:

• Choice of primer. This is crucial for molecular tests like RT-PCR.
• Chemicals used: Various chemical reagents are required to perform laboratory tests. Factors that may influence the test outcome include:
  1. Quality of reagent
  2. Standardisation of reagent (should be performed daily before commencement of testing)
  3. Expiry of reagent
  4. Chemicals prepared in-house. Sometimes, manufacturers require the use of additional chemicals (that are not supplied with reagents) to perform tests. These must be prepared in-house. All such preparations must be standardised.
• Quality control: The machine used for testing must undergo regular calibration and quality control (preferably external quality assurance).
• Lysed blood samples: Rolling the specimen tube, delay in processing, and spinning the tube at higher revolutions per minute (rpm) may cause lysis. This may cause false negative results.
• Decontamination of testing equipment before each test: If not performed diligently, may cause cross-contamination and false positive results.
• Changes in testing procedure: This may affect the test results when the person reporting test results is unaware of modifications to the testing procedure.
• Testing Kit: The kit used should be within expiry date, and must be stored at an appropriate storage temperature. The test kit should have successfully passed a rigorous validation process.
• Use of unauthorised tests: Certain specific tests (test kits) are authorised for laboratory diagnosis of COVID-19. Using unauthorised kits may yield non-valid (false) results. (For instance, the use of serological tests to diagnose pulmonary tuberculosis is prohibited. Nevertheless, these tests are used to diagnose pulmonary tuberculosis.)
• Reading time: The test results must be read within the specified ‘reading time’ (The reading time for HIV is 20 minutesàthe test result must be read within 20 minutes. Beyond this time the test will become ‘positive’).
• Error in interpretation of test result

Clerical error: This may include:

1. Mislabelling of sample with another person’s name
2. Typing error in the test report
3. Mixing up test results and identities (wrong test result to the wrong person)

Can a positive test result quickly become negative?

The short answer is, yes.

The long answer is, Yes if

1. The timing of the test was wrong (the first test was performed on day 6, the second on day 9 [RT-PCR])
2. A different test was employed
3. Guidelines for sample collection, storage, transportation were not followed
4. There were technical issues with the testing procedure followed
5. There was a clerical error
6. The source of sample was different
7. There was a technical error, etc.

As you might have realized, a similar response will be appropriate for the converse question (Can a negative test result quickly become positive?). However, generally one encounters disbelief when a positive test result rapidly becomes negative, not the other way around.

Acknowledgement: I am grateful to Dr. Riyaz Sherrif for his technical inputs.

Useful Links:

Link to the CDC page on COVID-19 testing:

https://www.cdc.gov/coronavirus/2019-ncov/symptoms-testing/testing.html

Link to the US FDA page on testing:

https://www.fda.gov/medical-devices/emergency-situations-medical-devices/faqs-diagnostic-testing-sars-cov-2

Link to WHO statement on use of point-of-care tests for COVID-19:

https://www.who.int/news-room/commentaries/detail/advice-on-the-use-of-point-of-care-immunodiagnostic-tests-for-covid-19