This site uses cookies. By continuing to browse the site you are agreeing to our use of cookies.

Improving our review of research using AI and data-driven technologies

Last updated on 12 Jan 2022

We want to improve the approvals process for people applying to start health and social care research involving artificial intelligence (AI) or data driven technologies. To identify common issues with our existing approvals process, we conducted qualitative research, a literature review, and a series of workshops with patients; industry; clinicians; academics; other regulators and HRA staff and volunteers.

This work showed we need to help developers of AI and data driven technology get access to data quickly and to clarify which activities are research and what approval they require. There is more information about what we did on the Streamlining Data Driven Research page. We’re now exploring a number of solutions which can then be rolled out as business as usual from 2022.

Coordinated review for Medical Devices with the Medicines and Healthcare products Regulatory Agency (MHRA)

AdobeStock_48872017_Preview.jpeg

We are working with the Medicines and Healthcare products Regulatory Agency (MHRA) to look at how we can streamline the overall approval process of clinical investigations of medical devices, with the aim of reducing the time taken to approve studies.

We are testing a coordinated assessment where the MHRA Medical Devices review and the Research Ethics Committee (REC) review are completed in parallel, with some information shared between the two organisations. We expect that the improvements identified will become our usual way of carrying out reviews later in 2022.

This is just one aspect of our joint working with MHRA to streamline research approvals.

We have also worked with MHRA to improve the review of clinical trials of investigational medicinal products (CTIMPs), by developing combined review. Combined review is now the way to seek approval for all new CTIMPs and offers a single application route and a joined up review, leading to a single decision in a faster overall timeline than the previous separate processes.

Access to confidential patient data without consent

Doctors giving results back to patient - REVERSE.jpg

Working with colleagues we have mapped, analysed and designed a new and improved route for applications to the HRA that involve research using identifiable patient data without the patients’ consent. This new pathway, which coordinates applications for both our Confidentiality Advisory Group (CAG) and our Research Ethics Committee (REC) review, will be piloted in the first half of 2022. By aligning approvals, improving communications between our CAG and RECs and consolidating communications with applicants, we hope to reduce the number of queries we receive from applicants after they have had their review.

As part of this project we are working with NHS Digital Data Access Request Service to look at ways to improve how we work together to make it easier for applicants to get access to NHS data.

Adding HRA Approval for databases and other data holding infrastructures

Evidence from our research showed that many organisations want to use health and social care data, at scale, for the development of AI. We do not currently review or offer HRA Approval for databases or other data-holding structures, but we will now explore adding this service. As part of this work, we will look at the way’s organisations are currently storing data. By adding HRA Approval for databases and other data-holding infrastructures, we hope we will be able to reduce the need for defined kinds of future project specific HRA Approval to use data for research.

Woman looking at Xrays.jpg

Supporting developers to access data and get the right approvals

Another clear message from our research was that developers and researchers want more clarity on which activities are research and what approval they require when developing AI and data-driven technologies. We will be developing some resources to address these areas and will be looking for people to help us test these.

How you can get involved

We will be undertaking and seeking a wide-range of opinions and feedback on this piece of work. If you are working in the development of data driven technology and /or AI please register your interest with us.

We are particularly interested in those people who are preparing research applications to CAG and would be willing to help us pilot a new coordinated review. If you are a researcher or sponsor it would be great to hear from you.

Back to streamlining data-driven research (sddr)