Government of Jersey

Reliability of Covid-19 test results (FOI)

Produced by the Freedom of Information office
Authored by Government of Jersey and published on 11 May 2020.
Prepared internally, no external costs.

Request

What is the known currently estimated rate of false negatives (ie people testing negative but actually positive) with the current coronavirus tests?

What is the known currently estimated rate of false positives (ie people testing positive but actually negative) with the current coronavirus tests?

Will the on-island tests being procured give a better rate of success?

Response

There are currently two diagnostic tests being used by the hospital for the SARS-CoV-2 coronavirus.

Some samples are tested by Public Health England. Data on their method is published by them and not held by the Health and Community Services Department.

Some samples are tested using the Cepheid GeneXpert method. The supplier’s data sheet is attached, which states performance criteria of the assay.

Xpert Xpress SARS-CoV-2 Data Sheet

The suppliers’ method data does not account for false negative results arising from swabs that are tested but have no virus on them in the first place. There is some evidence that not all people with COVID-19 shed the virus, and so it is sometimes not picked up by the swab, resulting in a false negative result. There is little published data on this. This issue affects all methods equally – no method can detect what is not there.

We are currently procuring a method to test for antibodies against SARS-CoV-2, rather than the virus itself. It would not be fair to say that we expect them to give a better rate of success because they are not looking for the same thing that our current methods look for. We would not be comparing like with like.

We expect the antibody tests to have an excellent rate of success in detecting antibodies against the virus, which is not what the current methods are looking for. We anticipate using the new test to identify people who have previously had COVID-19 and have now developed antibodies against it. It will not replace the two diagnostic methods referred to above which look for the virus itself and are for identifying people who currently have COVID-19.