Chair
Andrew George
Community Committee Members
Alisha Aman (AA), Anne-Laure Donskoy (A-LD), Christine Vial (CV), Eleni Chambers, Jan Speechley (JS), Louise Vale (LV), Sandra Duggan (SD), Sarah-Jayne Ambler (S-JA), Simon Kolstoe (SK)
HRA Staff Attendees
Becky Purvis (BP), Eve Hart (EH), Jonathan Fennelly-Barnwell (JF-B), Kat Evans (KE), Reshma Raycoba (RRa), Teagan Allen (TA - Secretariat)
HRA Guest Presenters
Karen Williams (KW), for agenda item 4
Susan Merriman (SM) for agenda item 3
Zoher Kapacee (ZK) for agenda item 1
External Attendee
Helen Mansfield (HM), Tony Curtis (TC), Fruitful Analytics for item 3
Apologies
Margaret Cheng, Stephanie Ellis
Welcome and apologies
AG welcomed the members to the meeting of the Community Committee and introduced HRA staff in attendance to observe and the guest speakers.
AG with a declaration of interest – AG is now Chair of Oxleas NHS Foundation Trust. AG noted that we have reimagined our agenda for this meeting – starting with papers for review and advice.
AG confirmed that Simon & Sarah-Jayne will attend the HRA May Board meeting as representatives of the Community Committee and will liaise with them regarding speaking at the Board in the next week.
AG announced that Jan Speechley has been appointed as the Community Committee Co-Chair following a recruitment process. Jan will work with Andrew and the HRA team in the coming weeks as we look to establishing the Co-Chair role and responsibilities moving forward.
- Action – TA to arrange a meeting with Jan and the HRA team to discuss the Co-Chair role.
It was agreed that the members would meet during the summer, for an informal, agenda less ‘get to know you’ meeting.
- Action – TA to work with members to arrange an informal meeting for summer 2024.
It was noted that the post arrived late for some. Members ask that Royal Mail do not require a signature to deliver papers.
- Action – members to send TA any possible solutions that could improve their experience getting printed copies of papers ahead of the meeting.
AG thanked Jan for her March Meeting blog which is now live.
Item 1 - REC members using Artificial Intelligence (AI) to review applications: for review and advice: Paper 1
SK presented the contents of Paper 1.
Excerpt from Paper 1 summarising the proposed approach:
“Existing policy over the handling of sensitive data has shaped the HRA’s approach to date. The HRA has reminded all staff and HRA Community members who have access to sensitive data that such data, including but not limited to the following, should only be processed by HRA-managed systems, or systems for which a valid data sharing agreement exists. It must not be shared with public AI systems (such as ChatGPT):
1: IRAS forms and associated documents.
2: Commercially sensitive information, for example drug names, protocols defining methodologies, licensing arrangements, or any other information that would be deemed to be confidential by the applicant.
3: User data (for example completed feedback forms or comments in response to calls for input)
4: letters to and from the HRA.
5: audio recordings or minutes of any HRA or REC meetings.
This ensures that our systems and data are used in compliance with the Data Protection Act 2018 (DPA 2018) and the UK General Data Protection Regulation (UK GDPR), as well as new guidance from the Government Communications Service which says that we must be explicit if outputs like emails or responses to enquiries have been produced with the help of AI.
Work underway to understand the potential of AI technologies for REC review:
AI technology offers the potential to automate and enhance complex processes such as research ethics review, which is one of the approvals that health and social care research projects need to earn before they can go ahead. There is huge value to streamlining the research ethics review process. However, it is very important that any application of AI is well-understood and maintains public confidence in the approval process.
The HRA is currently considering a pre-pilot evaluation to test use cases as a guide to possible broader adoption of AI technologies. We are purposely taking a phased approached because:
Building internal expertise in AI development, implementation, and regulatory oversight requires time and resources, necessitating a careful risk assessed approach. The use of AI tools for regulatory purposes is nascent, with no mature fully tested options readily available. This necessitates an experimental phase to develop appropriate solutions and ensure responsible adoption. Transparency and involvement of stakeholders, including researchers, patients, and the public, are essential for earning trust in AI use for the regulation of research – the pre-pilot will foster collaboration and learning before any large-scale implementation.
The HRA pre-pilot will evaluate an AI tool using past research ethics committee (REC) data with the aim to assess the feasibility and potential benefits of applying AI generated efficiencies into the review process.
Specifically:
To test AI outputs in predetermined aspects of REC review against existing standards and frameworks (e.g. SOPs, guidance on participant information sheets) and compare with standard review findings from RECs and staff.
Explore other features of the REC review process that could be automated using AI tools.
The results of pilot projects will inform the planning and development of clear guidelines for any AI adoption in the HRA. Continuously monitoring and evaluating the experimental phase ensures the HRA remains sighted on and, together with others such as the Medicines and Healthcare products Regulatory Agency (MHRA) make contributions to, responsible AI use across the sector in the regulation of health and social care research.
Additionally, the results will inform a collaboration with an NIHR-funded project to better understand how AI can be applied to the research ethics review process to identify: a) the potential of such technologies to streamline the process of ethics review and increase consistency (as appropriate given that ethical review is not a binary decision), and b) identify how and where AI may be used, while maintaining people’s confidence in the process.
The findings of this work may inform changes to the process research ethics review to make it easier to do research that people can trust.
Further developing the HRA approach to AI technologies:
The HRA is developing proposals that will inform future spending bids for resources to support activity to ensure that the HRA is prepared for the increasing use of AI technologies and has appropriate processes and policies in place, enable the HRA to scope and use AI technologies effectively in the work of the HRA.”
Committee members were asked to review and advise on:
The approach the HRA is taking.
Anything that the HRA should be considering/aware of/may be missing from this approach.
The Committee members discussed the need to keep pace with the development of AI, whilst ensuring that the use of AI is appropriate and proportionate for the work the HRA does.
Another member noted that AI is needed, particularly as humans can make mistakes. But there would need to be an assurance that if we do not build an independent HRA system, that privacy and confidentiality is adhered to when utilising a third-party company.
The members discussed the possibility of licencing options and collaborating with another body of the government.
A member stated that ethics, by its very nature, is a human concept and process. Ethics can be perceived as being dependent on being human. Trust of the public in research is essential and the use of AI may put this trust at risk. The member recognised the appeal of AI: reduce the time it takes to review a large volume of papers and noted that this may be tied to the good will of reviewers and the provision of payments for such work.
The members discussed how AI can support an ethics review process but cannot make an ethical decision.
A member noted the possibility of programming bias – how do we handle this as part of a review. Ethical discussions are often personal and subjective.
Whilst it can streamline and speed up the review process, AI ideally should run alongside the current process. An AI system will only be as good as the data it is trained on – there will need to be a lot of work to make sure the training dataset is on-point and free of bias in the first instance.
A member suggested AI should be used in parallel to an ethics reviewer rather than replacing the human aspect of it. It should be the REC members who discuss the application and its ethical implications - but AI can help reduce the work and mental load in doing this.
A member stated that using AI for certain tasks carries a risk that researchers will no longer have the skill sets to complete certain aspects of research – writing lay language summaries, using plain language etc. The use of AI could potentially de-skill a researcher’s data literacy.
Another member noted the need to be quick and stay up to date with a fast-moving industry whilst recognising we need to be sure who and what company etc, we are trusting with data. There is a potential danger that we over-trust AI, that we do not question the accuracy enough, that we use the evolving technology in our work too quickly, an example of the potential damaging impact of which can be seen with Horizon computer system used in post offices to manage in-branch sales.
A member asked about the reflexivity element of the review process – there is a danger of reinforcing the tick box aspect of putting an application through approval using AI.
A member also asked how AI would read and assess something like a “non-standard” Participant Information Sheet, e.g., one using infographics.
The Committee advised that the HRA must be transparent in how it will use AI in the future. The HRA needs to be able to explain and justify whatever decision you make to the public.
Member comments via email, out of session - A member stated that although they understand the HRA stance I would also want to flag up that AI should not be forbidden, we need to find the right way of living with it. Especially if it helps people who might find reading all the paperwork difficult. In an ideal world, the new HRA Research Systems Programme should have built into it a subset of AI for doing this job, but we still need to address the current situation.
Item 2 - How we enforce standards to earn people’s trust in research: for review and advice: Paper 2
BP presented the contents of Paper 2.
Excerpt from Paper 2 summarising the proposed approach:
“The HRA sets expectations for what health and social care research should look like. These include requirements that will be set out in legislation for some of the research that we review – e.g. research transparency requirements for clinical trials of medicines – and requirements and expectations that are set out in guidance – e.g. current transparency requirements for all clinical trials, expectations around public involvement in research.
Where an application to a Research Ethics Committee does not meet these requirements and expectations, Research Ethics Committees can choose not to approve the research, give a provisional opinion but ask for further information, set conditions that need to be met for it to go ahead or choose to approve the research.
Much of the HRA’s work is focused on encouraging researchers to meet the HRA’s requirements and expectations before they make an application. However, the HRA needs to strike a balance between enabling research to go ahead and ensuring that it is research that people can trust.
In the course of our work, we continually reflect on where that balance should be. This informs our approach to research applications where our requirements and expectations are not met and how we support our Research Ethics Committees in their decision-making.
This paper provides some examples to illustrate this issue.
Participant information quality standards – taking action where there is an expectation:
In 2023 we rolled out our Participant Information Quality Standards (QS) and Design and Review Principles (DRP) with the aim to improve the quality of participant information materials and to make the ethical review process undertaken by RECs more consistent. The QS and DRP were soft launched in July 2023 and applied to new submissions from September 2023, at which time any areas of non-compliance were raised as recommendations only by the REC. From 1 December 2023, compliance with the QS and DRP became mandatory for any new applications.
Research ethics staff undertake a check against the QS when a research application is validated for review by a REC. The staff check extends only to documenting the presence or absence of the necessary information to address the QS, which is passed to the REC to consider as part of its ethical review. As the QS are ethical expectations, it is the REC which is responsible for determining whether the QS have been met and the participant information materials are considered compliant. This is an important distinction to make as the participant information materials may include all relevant detail, but in a way which is unsuitable within the parameters of the specific study being reviewed.
The DRP provide a framework to support RECs to make these decisions. The DRP provide information to both RECs and researchers about the important ethical considerations when preparing high quality information materials for use in research.
The DRP encourage a proportionate and pragmatic approach to the development of participant information, which is appropriate to the presented study and patient population and is informed by public involvement.
Following the REC review, researchers receive a single outcome in relation to the application, which will include all outstanding requests in relation to the participant information materials, including where the QS have not been achieved. This provides a consistent and streamlined process for all, as follows:
REC Staff – request changes to participant information at a single timepoint, preventing delays or duplication of review from the submission of multiple revisions,
REC Members – are able to discuss issues or concerns with the presented information materials within the REC meeting and provide clear requests for information or revisions, explaining why they are necessary to achieve expectations,
Researchers – are provided an opportunity to discuss queries and provide clarity and reassurance in the REC meeting, following which they will receive a single request for further action in order to address the expectations.
Both the QS and DRP are publicly available so the expectations of what high-quality participant information materials should be and included are clear to researchers, RECs, and members of the public. It is recognised that any new initiative can take time to bed into usual practice, but it is expected that researchers will use the DRP as a framework to design information materials which are compliant with the QS.
An evaluation of the rollout and implementation of the QS and DRP has provided early evidence to support the view that these expectations are beginning to bed into both the research community and the REC review process.
A 10% sample of applications reviewed at REC meetings in October 2023 (26 applications) was analysed to determine a baseline immediately after the soft launch of the QS and DRP. This was compared against a 10% sample in February 2024 (18 applications) to determine whether the QS requirements were bedding in within the research community.
The number of points being identified by staff in the QS proforma at validation (initial review prior to the application being sent to the REC) has reduced from a range 0-8 findings in October 2023 to 0-3 findings in February 2024.
It was also recognised that throughout the analysis, incomplete information around the public involvement activities which had been carried out in the development and review of participant information materials remained the most prevalent issue identified by staff in the validation check. This was identified in 53% of the sample in October 2023 (14 applications) and was still identified in 28% of the sample in February 2024 (5 applications). An outcome of the evaluation was a recommendation to provide further guidance to staff, REC Members and the research community about the requirements in this area.
The number of points raised following the REC review as part of the Request for Further Information (RFI) relating to participant information materials reduced from a mean of 5.4 points (range 0-14) in October 2023 to 2.83 points (range 0-10) in February 2024.
REC staff capture key information post-REC meeting relating to the finalised REC meeting minutes. During the rollout and implementation phase of the QS and DRP, some additional questions were added to this data capture to collect real time information from staff on how the QS and DRP were impacting the quality of information materials received and the ethical review process.
Staff were asked to record whether the RECs ethical review of participant information materials was focussed and aligned to the DRP. In October 2023 it was reported that 91% of reviews were aligned wholly, or to a degree to the DRP. By February 2024, this figure had increased to 97%. This data highlights a positive culture shift for RECs undertaking reviews of information materials which are wholly or partially aligned to the DRP, as these further bed into business-as-usual practice, and RECs become increasingly familiar with reviewing in this manner.
The evaluation suggests that both the QS and DRP are being adopted as standard practice across the research community and encouraging the positive changes they were intended to create, by improving the quality of information materials used in research and streamlining the review process they undergo.
There have been instances since the implementation of the QS and DRP where researchers and RECs have had a difference of opinion or interpretation of how to apply the QDS and DRP. These have all been handled via usual practices with REC staff facilitating discussion between the two parties to reach a resolution.
Where the participant information materials are deemed non-compliant with the QS and DRP, the application will be issued with a provisional opinion with a requirement for necessary changes to be made before a favourable opinion is issued.
Requiring registration of clinical trials of medicines – taking action where there is a legal requirement:
The HRA has a statutory responsibility to promote transparency in research.
Our Make It Public strategy sets out how we will increase transparency in research, by which we mean:
registration: making it public that a study has started.
reporting results: making it public what the study has found.
informing participants: letting those who took part know what the study has found.
It is a condition of a Research Ethics Committee (REC) favourable opinion that a clinical trial is registered. We currently require that researchers should register their trial before the first participant is recruited and no later than six weeks after.
The law for clinical trials of medicines is currently being updated to make this registration a legal requirement (clinical trials of medicines make up some of the applications to do clinical trials that RECs review – this will remain a condition of a favourable opinion rather than a legal requirement for the rest of the clinical trials the RECs review).
In 2019, we asked people to share their views on how we can best promote transparency. You can read the findings of the online consultation and workshops here: Feedback from the consultation - Health Research Authority (hra.nhs.uk)
They were largely supportive of us publishing the registration status of the research that a REC has approved, which we have done for the first time this year. You can see this here: Clinical trial registration report 2022 - Health Research Authority (hra.nhs.uk)
We asked for views about an option to take into consideration the extent to which sponsors have fulfilled their transparency responsibilities in relation to their previous studies, when reviewing new studies for approval. The majority of online respondents (75%) supported this approach with significantly more patients and members of the public (92%) agreeing than research managers (67%), researchers (71%) and Contract Research Organisations (CROs) (37%).
We also asked for views about whether we should impose a financial penalty on sponsors with very poor transparency rates, noting that it would need a change to the law to make this possible. Response to this option was more divided than for other options for sanctions. 47% of online respondents did not support this sanction whilst 39% did. There was greater support from patients and the public (57%) completing the online survey for the use of fines than those employed to conduct research such as research managers, researchers, and Contract Research Organisations (CROs) (30%).
We decided not to seek the power to impose financial penalties and will focus instead on making performance information public – as we have done for registration this year - and taking into account past performance when reviewing new studies for approval. Work is currently underway to develop and agree approaches to do this.”
Committee members are asked to review and advise on
- what people/perspectives we should involve?
- anything else that the HRA should consider?
A member noted that the HRA can flag issues such as registration requirements but can be clearer in what REC members have the power to do and say no to in terms of approval – the HRA can provide clear guidance and guidance and set expectations for RECs to apply and follow.
It is important that people know what, how and why a study is approved in a transparent manner.
The members discussed the need to be transparent with how public money is being spent and how the HRA monitors and surfaces compliance to increase public trust in research. A suggestion being that the HRA might consider looking at approaches like the Athena Swan Charter. (Athena Swan Charter is a framework which is used across the globe to support and transform gender equality within higher education (HE) and research) which uses a gold, silver, bronze accreditation system.
The members discussed how significant work has already been carried out over the years by the HRA, with a suggestion that the HRA implement an ‘ethics framework’ using this discovery work to support training for researchers before an application is submitted for approval. The approval stage often feels as though it comes too late in a study life cycle when a study design is almost at the point of completion. A framework must therefore be delivered as early as possible and be a two-way training process with a researcher and the HRA and REC’s.
Diversity of ethics groups and others involved in research is essential to building trust.
SK shared a “code” vs “ethics” consistency article published in the Journal of Medical Ethics with the members.
- Member comments via email, out of session – a member noted that the REC they work with does raise PPI on a regular basis. However, situations do arise that can complicate the process. The first is when the researchers are using a PPI group in which the members are people who have been in the group for a long time and are experts in their condition. Whilst there is nothing wrong with this, they may miss what might be confusing to others: they may no longer see technical language as bewildering. It is important to remember that for many researchers the patient facing documents are not the explanation of the research, so much as the framework of how the researchers will discuss the research with a potential participant - we need to give researchers a better explanation of how they need to explain their public involvement to us.
Item 3 - Optional session: Community Survey 2024: for information: Paper 3
Guests: Helen Mansfield (HM), Tony Curtis (TC), Fruitful Analytics, Susan Merriman (SM), HRA Project Manager
HM presented slides to the Community Committee members detailing the early findings from the recently completed survey of HRA Community members.
Committee members asked how the information collected by the survey will be used: HRA team confirmed that this work is ‘in train’ to share the findings publicly. The HRA plan to bring an updated action plan, acting on the findings to the September meeting of the Community Committee.
HM confirmed that the full findings include tables showing what individual groups in the HRA Community said.
Committee members agreed it is important to look at the granularity of responses.
- Action – HRA team to analyse survey results and update current action plan (created by Community Insight Group) and will incorporate Community Committee observations.
- Action – HRA team to circulate the updated action plan to the Community Committee at our September meeting.
Item 4 - HRA Annual Report 2024; for review and advice: Paper 4
Guest: Karen Williams (KW), HRA Deputy Chief Executive and Director of Finance, Procurement and Estates (Executive Director).
KW updated on progress in developing and publishing the HRA’s business plan for 24-25. The HRA chose to delay publication of this to enable it to be informed by the HRA’s financial settlement for 2024-25 and the completion of the foundation stage of the Research Systems Programme. This will now go to the HRA Board in May.
We lay our annual report and accounts before Parliament on a yearly basis.
Our annual report describes the work of the HRA and outlines some of the most significant achievements and sometimes challenges. A draft of the report was shared with Committee members.
Committee members were asked to review and advise on:
a) The approach that the HRA is taking.
b) Anything that the HRA should be considering/aware of/may be missing from this approach.
Committee members noted the need to be clear with public involvement expenditure. This spend should be detailed separately if possible. It would be good practice to include reference to money spent working with the public as part of the performance analysis. This should be highlighted and celebrated.
Committee members agreed that the document must be made available in an accessible form. Suggestions to improve accessibility includes adding links to the document and implementing the use of ‘talking heads’ as part of the supporting content – which refers to the HRA strategy and demonstrates its performance within the report.
Committee members also suggested that the HRA uses the report to define ‘success’ and define how we are measuring success and how we report it.
KW undertook to share the HRA Annual Report with the Public Involvement Network when published.
- Action – HRA to consider producing an easy read version of the HRA Annual Report 2024.
Item 5 - Agreeing Community Committee Terms of Reference: for approval: Papers 5a (track changes) & 5b (clean)
The draft Terms of Reference for the HRA Community Committee and formal appointment letters for members were developed following the discussion at our 19 December 2023 meeting and modelled on Terms of Reference for other similar HRA Committees.
It was shared with the papers for the 01 March 2024 meeting for review and advice. We requested any comments to be provided in writing by 01 April 2024 or offered to meet and discuss any feedback if this was preferred. We undertook to revise the Terms of Reference further in the light of these comments, with a view to approving them at our May meeting.
At this meeting we proposed the following amendments:
- Make-up of the Community Committee:
Section 2:2
When we established the Community Committee, the Community Insight Group recommended that this should be made up of at least half people who do not have professional expertise in clinical research or health or social care.
In reviewing the Terms of Reference, questions have been raised about both whether this requirement is needed, and the definition used to distinguish between members.
We have proposed to maintain this requirement – noting that we should review its effectiveness when we next review the Terms of Reference – and use the current definition for ‘Lay Plus’ as used by Research Ethics Committee members.
The language that the HRA currently uses in our appointment documentation when referring to the classification of REC members is as follows:
Expert
You will be an expert if you are currently a registered health or social care professional, (e.g. dentist, doctor, pharmacist, midwife, nurse, chiropractor, ophthalmic optician, osteopath, social worker or registered on the HCPC register), a retired dentist or doctor, or someone who is qualified or has experience in conducting clinical research.
Lay
You will be a lay member if you are a retired health or social care professional, (excluding retired dentists and doctors), someone who is involved with the management, monitoring, regulation or administration of clinical research, or someone who has been a chair, member or director of a health service body or any other body providing health care.
Lay plus
You will be a lay plus member if you have never been a registered health or social care professional, involved in clinical research (other than being a participant recruited to a research project), a chair, member or director of a health service body or any other body providing health care. (these definitions are in the process of being updated and we would propose that the Committee might consider updating the Terms of Reference to continue to mirror them when this process is complete).
Committee members discussed the need to continue to work on clarification of the definition of lay and expert. Some members noted that it is difficult to know which category they belong to using the current definitions and discussed how the use of ‘clinical’ can cause misunderstanding.
RRa confirmed that the HRA is required to use the definitions contained within the current Clinical Trials Regulation, which is in the process of being updated, but noted that the HRA has had similar feedback in the past and the HRA team is currently looking at how we use these definitions in practice going forward.
Members proposed that the terms refer to ‘health and social care’ if possible.
BP confirmed that equality, diversity, and inclusion training can be made available for Committee members to complete.
The members discussed the need for the Terms of Reference to reflect the ‘include’ pilar of the current strategy – in that we ensure we include all, do not ignore the minority and that we include everyone regardless of number or the majority.
AG proposed that the Terms of Reference utilises the current wording and definitions with an action to implement a one-year review.
- Action – HRA team to continue to look at the current categorisations of REC members as part of the wider work alongside the update of the Clinical Trials Regulation.
- Action – HRA to explore providing equality, diversity, and inclusion training for Committee members.
- HRA staff attendance at Community Committee meetings
Section 3
To date, the following staff have attended Community Committee meetings:
- Director of Policy and Partnerships and Senior Engagement Officer, providing the secretariat.
- Staff leads for the support teams for REC members, CAG members and the public involvement team and the Company Secretary - to listen to the discussion, answer any questions, ensure Community members are supported and act on any issues raised.
- Other HRA staff as required to present items for discussion and received advice.
We have proposed to continue this attendance in the Terms of Reference.
Members agreed with this proposal.
- Period of appointment
Section 10
The HRA Board and Audit and Risk Committee appoint members for a period of 2 years with the option to extend by a further 2 years, twice. This seeks to ensure a balance of experience and new perspectives among the membership and a manageable time commitment for members.
We have proposed harmonising with other HRA governance committees and taking the same approach for this Committee (in practice for the current members, we would seek to flex this as it is introduced so that membership changes are staggered).
Members agreed with this proposal.
- Action - HRA team to update the Terms of Reference in line with this discussion and publish on the HRA website.
Item 6 - HRA 2023-24 Q4 Strategic Performance Reporting
Community Committee members received drafts of the HRA’s quarterly strategic reports ahead of HRA Board, to highlight and advise the Board on issues that they thought important.
Members received the following reports:
- Highlight Performance Report Q4 2023 – 2024
- HRA Strategic Risk Update
- Portfolio Dashboard Q4 2023 – 2024
- Strategic Engagement Update for HRA Board - May 2024
Members noted that CAG have been doing a lot of work recently to speed up and streamline processes and members have asked that this be highlighted and presented to Board where possible.
Item 6 - AOB
Next meeting: 06 September 2024, 10:00 – 13:00 (Zoom)
At our next meeting we would like to get your views on:
- Public involvement in Research Ethics Committees
- Action plan following the HRA Community Survey 2024