What tests and why?

There are two different types of testing, both very different in time and objectives.

PCR test

The so-called “PCR” test (Polymerase Chain Reaction) is a screening for the presence of the virus in an organism.

It is the only test currently validated by WHO. “The only biological diagnostic technique for COVID-19 recommended to date is the molecular test by RT-PCR allowing the direct detection of the RNA (viral genome) of the coronavirus SARS-CoV-2.

It is performed through nasopharyngeal swabs deeply immersed in the nostril: This test is considered now very reliable once performed correctly.

Laboratories uisng the PCR test specifically for SARS must apply strict procedure for positive results confirmation, especially in areas of low prevalence, where the positive predictive value may be lower.

The sensitivity of PCR tests used for SARS will vary depending on both the sampling type and the period in which the test will be performed during the sickness. This can only be performed when genuine cases of SARS show a negative PCR test (false negatives). The use of tests on multiple sampling and/ or at different sites may improve the sensitivity of the test.

The specificity of the SARS PCR test is excellent as the technical procedures follow quality control guidelines. We may observe false negative results due to technical reasons (e.g. due to laboratory contamination) and therefore, each positive PCR test should be checked.

Serological tests: Detection for antibodies

Serological tests allow the detection of antibodies produced by the body and directed against the virus. They may attest of the immune response of the tested person and can thus show whether the person is infected or has been infected.

Serological tests are not recommended for the early diagnosis of COVID-19 infection within the week following the onset of first symptoms as antibodies production might be delayed until day 7. (See figure below).

While PCR remains a very suitable test at the onset of symptoms, the addition of IgM serology turns out to be of great help for the diagnosis diagnosis [1]

In the second week following disease onset, positive rates were 54.0% for PCR and 89.6% for antibody tests.

The combination of PCR and antibody testing improves diagnosis during the different stages of the disease.

Scientific knowledge has been progressing the past weeks, particularly thanks to the end of May publication about a study showing that even patients with minor forms develop antibodies that neutralize the virus in the laboratory[2], however, we still ignore whether this neutralizing activity is associated with protection against reinfection.

Will this immunity be the end of the general confinement?

Serology data will be an important parameter, but not the only one. Other measures must be carried out, wilaya by wilaya. And the end of the quarantine will not mean people will be authorized to freely move unprotected. Quite the contrary.

A positive serology does not give permission to people to go for a walk in the streets, or shop in stores without wearing masks. It cannot be a pretext for a “relaxation of barrier measures within the population”.

Serological tests meet an urgent need to help with the end of quarantine, for all the people who work in collectivity, exposed nursing and non-nursing staff.

Performed on representative samples of the population, these tests will represent though a valuable indicator for epidemiologists, who will be able to have some insight regarding the prevalence of the coronavirus within the Algerian population. A particularly useful data to assessing the time group immunity could be achieved in – that is, the fact that a sufficient proportion of immune individuals (estimated at around 60% of the population) will prevent circulation virus.

Without serological data, specialists can only make estimates and assumptions so far. According to them, only 10 to 15% of the population have encountered the virus. This explains why it will not be easy to end up quarantine without redoubling efforts on barrier gestures.

Type Serological tests ELISA

It is an immuno-enzymatic detection technique performed in laboratories and that allows the visualization of antigen-antibody reactions thanks to a colored reaction produced by the enzyme substrate action previously linked to the antibody. The use of monoclonal antibodies makes the detection specific and the achievement of a range in parallel (reference line performed by diluting serially with a positive control) allows the quantification of the patient’s antibodies in the blood. Several tests are marketed including anti-SARS-CoV-2 ELISA IgG, Euroimmun, Germany, EDI New Coronavirus COVID-19 IgG ELISA, Epitope Diagnostics (EDI), USA, and recomWell SARS-CoV-2 IgG ELISA, Mikrogen, Germany (6). However, an enzymatic reaction makes this technique dependent on temperature, pH and lightening. More precisely, the ELISA requires the achievement of different successive steps: the specific antigen of the SARS-CoV-2 virus (the N protein contained in the viral nucleocapsid or the binding receptor of the virus known as RBD (Receptor Binding Domain)) is stuck overnight in the bottom of a well of a 96-well plate (“coating”); the antibodies within the patient’s plasma sample will bind specifically to the antigen. A detection antibody will then fix the human antibodies for dosage. Those detection antibodies are coupled to an enzyme which, in the presence of its substrate, transforms it into a detectable and measurable reaction product thanks to the appearance of a coloration. The intensity of the latter matches the amount of enzyme present and therefore to the concentration of the required antibody.

Some of these steps (including coating) take several hours. The ELISA test cannot be performed with a drop of blood. It is therefore not a quick test and it cannot be performed at the patient bedside.

The ELISA sandwich is a less common (clinical) variable of this technique, used to detect an antigen sample in serum or any other sample. This technique is, however, very common in research.

Rapid immunochromatographic tests on a nitrocellulose strip (also commonly named “Lateral Flow Assay” or “LFA”)

In contrast to the ELISA test, these are rapid tests that allow a diagnosis in a matter of minutes ( < 15 minutes). The blood drop sampling is made from the fingertip at the patient’s bedside, in the laboratory, but also at the doctor’s office. With these tests, it is not by dosing, but the detection of the presence or absence of antibodies (IgM and IgG). Their purpose is to identify people who have had infectious contact with the virus (with or without symptoms) and who have developed antibodies.

The results interpretation should be done by a physician.

A quality rapid screening test performance

A screening test is used to sort among an apparently healthy target population the people who are more likely to have the disease

A screening test must have the following qualities: Valid, simple, specific, sensitive, reproducible, acceptable and affordable.

  • – A test is considered valid in its ability to pick up within the target population those who are likely to have the disease from those who are likely not. This capacity depends both on the specific performance of the test and on the characteristics of the population tested.
  • – The specific (intrinsic) performancesof the screening test are its sensitivity and its specificity: They define the intrinsic validity of the test. They are defined and calculated under experimental conditions and are therefore independent from the type of person tested.
  • – The characteristics of the population tested, in particular the prevalence of the disease, condition the extrinsic performance of the test. These ones are the positive and negative predictive values. They are related to the use of the test on a given population and differ according to the characteristics of the population tested. They are defined and calculated in a screening situation and show the importance of the use of the test on this specific population.

Intrinsic performance: sensitivity and specificity

The two main qualities of a test, that define the internal validity (“accuracy”) of the measuring instrument, are:

– Sensitivity:Ability of the test of identifying subjects with the disease

– Specificity:Ability of the test of identifying healthy subjects.

  • The specificityof a test is the probability that the test will be negative if the person tested is not infected with COVID-19. This is the number of true negatives (negative tests in people with no COVID-19) divided by the total number of people with no COVID-19 (d / b + d).

The more specific a test is, the fewer false positives it causes (positive tests in people free from COVID-19) and the better it allows, if it is positive, to confirm COVID-19.

  • A test sensitivity and specificity are interdependent: increasing the sensitivity of a test always comes at the expense of its specificity and vice versa. This is the number of true negatives (negative tests in people with no COVID-19) divided by the total number of people with no COVID-19 (d / b + d). The more specific a test is, the fewer false positives it causes (positive tests in people free from COVID-19) and the better it allows, if it is positive, to confirm COVID-19.A test sensitivity and specificity are interdependent: increasing the sensitivity of a test always comes at the expense of its specificity and vice versa.
Test Present Absent
Positive a
true positive (VP)
false positive (FP)
a + b
total positive tests
Negative c
false negative (FN)
true negative (VN)
c + d
total negative tests
a + c
totally sick
b + d
total not sick
total workforce

Extrinsic performance: Positive and negative predictive values

In daily screening practice, the questions that arises for a doctor is to assess in an apparently healthy person the probability of being sick, as per the positive (or negative) result of the test. This probability is also called a posteriori probability or post-test probability. It depends on the characteristics of the test (sensitivity and specificity) and on the a priori probability (pre-test probability) that the person has a disease, that is to say on the prevalence of the disease in the population considered. The prevalence of the disease is the a priori probability that the disease is present in a random person in a population (a + c / a + b + c + d).

  • The positive predictive value (PPV) of a test is the likelihood (probability) that a person is actually sick if their test is positive. This is the number of true positives (positive tests in people with the disease) divided by the total number of people whose test is positive (a / a + b). Bayes’ formula calculates a test PPV according to its sensitivity (SE), its specificity (SP) and the prevalence of the disease (P)
VPP = SE x P
SE x P + (1 – P) x (1 – SP)
  • The negative predictive value (NPV) of a test is the likelihood that a person will not have the disease if their test is negative. This is the number of true negatives (negative tests in people without the disease) divided by the total number of people whose test is negative (d / c + d).
VPN = SP x (1 – P)
SP x (1 – P) + P x (1 – SE)


  • Reproducibility is the ability of a test to give the same result when tested repeatedly on the same subject. It is measured by the proportion of concordant cases. Reproducibility depends on the measuring instrument (dispersion of the results), the observer (interpretation of the results) or the observation conditions (dosage time for example).
  • Affordable and acceptableto the target population.

Sensitivity and specificity are proportions: they can therefore only vary between 0 and 1 (0 and 100%). A test is said to be “sensitive” if the number of false negatives is low, it is said to be “specific” if the number of false positives is low.


Serologic testing is a valuable indicator of the prevalence of SARS-CoV-2 and collective immunity in a population. They will help defining the kinetics of antibodies and the diagnostic performances of serological tests, which is an essential but lacking data due to the emergence of SARS-coV-2.

To date, preliminary data suggest a figure of 6 to 10% of the population in contact with this virus, without taking into account the end of quarantine and a wider circulation of the virus in all layers of the population and on the unknown of humoral immune response quality among the youth, the asymptomatic, and the pauci-symptomatic. We should also consider the choice of diagnostic kits and their reliability, as well as the conditions under which the assays were performed. The convergence of the results of studies at all stages of the disease, including asymptomatic, pre-symptomatic and symptomatic, is the prerequisite for a good understanding of the individual contribution of serology in the detection of COVID.

[1] Li Guo, Lili Ren, Siyuan Yang, et al. Profiling Early Humoral Response to Diagnose Novel Coronavirus Disease (COVID-19), Clinical Infectious Diseases, ciaa310, 21 March 2020

[2] https://www.pasteur.fr/fr/espace-presse/documents-presse/covid-19-tres-grande-majorite-malades-atteints-forme-mineure-developpent-anticorps-sero