Guidance

Advice for manufacturers: software and apps used in COVID-19 pandemic

Updated 17 December 2021

If you already have a CE, CE UKNI or UKCA marked piece of software or app

If your software or app has a role in responding to the current COVID-19 pandemic, but is not CE, CE UKNI or UKCA marked for this purpose, you may be able to use the MHRA’s COVID-19 related regulatory flexibilities which provides options for exemptions and expedited clinical investigations.

If there is a significant change to the intended purpose of a CE, CE UKNI or UKCA marked app or software you may also be able to apply for the device to be part of an expedited clinical investigation in order to gain clinical data. In addition, you could consider an exceptional use application to place your software or app on to the market. Review our regulatory flexibilities for further details on how to apply and the information required.

If you want to add COVID-19 as a new indication there needs to be adequate clinical evidence to support this new claim.

If your device is already CE, CE UKNI or UKCA marked and there is only a limited change, for example a change to the clinical use setting, then we would not expect to receive a request for exemption.

However, you should perform a risk assessment.

Guidance is available on what constitutes a ‘significant change’.

If you need to contact MHRA for advice, do this as early as possible info@mhra.gov.uk.

If you are developing new software or an app specifically in response to COVID-19

If you are developing software or an app which is likely to be a medical device, you may be able to apply for an exceptional use or derogation exemption prior to full UKCA or CE UKNI marking of the product.

Further details on regulatory flexibilities and how to apply for an exceptional use derogation.

If you have any questions about what is required, we recommend that you contact devices.exceptionaluse@mhra.gov.uk for advice as early in the development process as possible.

Software specifically intended to read lateral flow tests (test readers)

If you are developing an image analysis algorithm that is intended to interpret lateral flow test results, the MHRA recommends you to consider the following specific information before submitting an application for exceptional use or derogation exemption.

When developing software as a medical device, please refer to the definitions of COVID-19 tests and testing kits.

The below information is not exhaustive but provides an idea of further considerations relevant for test readers.

Before submitting your application, please ensure that the following considerations are noted and addressed:

• The MHRA typically views software that reads lateral flow tests as IVD software.
• The MHRA will not accept test readers that are described as “test agnostic”. Manufacturers should explicitly state which lateral flow test kits are within the scope of their test reader’s intended purpose. The MHRA would expect to see the supporting evidence outlined below for each lateral flow test kit that the test reader is intended to be used with. In particular, the MHRA would expect to see evidence demonstrating that the test reader results do not negatively impact the sensitivity and specificity of any of the tests kits it is intended to be used with.
• The sample size of datasets used for model development must be appropriate and justified for both the sensitivity and specificity of the test reader.
• The MHRA views the following test dataset sample sizes as the minimum necessary to merit further consideration in the exceptional use authorisation process. The MHRA would expect these test dataset sample sizes to be present for each lateral flow test kit that the test reader is intended to be used with:
  • 100 viable positive samples
  • 1000 viable negatives samples
  • 50 void samples
  • The target to be achieved is 97% sensitivity and 99.9% specificity.
  • The presentation of the results should include the calculated sensitivity and specificity along with the associated 95% confidence intervals (calculated using the exact binominal method), and relevant confusion matrices.
• The MHRA also stresses that ‘test dataset’ in this circumstance means that the test dataset has been completely unseen by the model and has not been used for model development.
• The datasets noted above must appropriately reflect the photographic conditions likely to be encountered.
• Supporting evidence on usability and accessibility should be considered. Where relevant, this should take into account the needs of the lay users who may not be technically or medically trained.
• Where relevant, the instructions for use must take into account the needs of lay users. The instructions for use should meet the relevant essential requirements.
• The instructions for use and any other promotional material must clearly set out which lateral flow test kits the device is intended to be used with.
• The MHRA would expect that the instructions for use should clearly direct users towards the MHRA Coronavirus Yellow Card scheme in the event of an adverse incident or a disputed result.