UK STATISTICS AUTHORITY
RESEARCH ACCREDITATION PANEL
Tuesday 9 March 2021
Present Committee Members Professor Paul Boyle (Chair) Andrew Bolton (Deputising for Paul Lodge, Department for Work & Pensions) Mark Brewin (HM Revenue & Customs) Siobhan Carey (NISRA) Chris Dibben (Independent member) Tricia Dodd (Independent member) Andrew Garrett (Independent member) Emma Gordon (UK Research and Innovation) Roger Halliday (Scottish Government) Sarah Henry (ONS) Sarah Mathieson (Independent member) Alex Singleton (Independent Member) Misa Tanaka (Independent member) Advisors Andy Wall (Chief Security Officer, ONS) Ross Young (Data Protection Officer, UK Statistics Authority) UK Statistics Authority Grazia Ragone In Attendance Andrew Austin (Security and Information Management, ONS), for item 4 Bill South (Research Services & Data Access, ONS), for items 4 and 5 Karen White (Data Governance, Legislation & Policy, UK Statistics Authority), for item 6 Apologies Stephanie Howarth (Welsh Government)
Simon Whitworth (Data Ethics, UK Statistics Authority)
Professor Paul Boyle (Chair)
Andrew Bolton (Deputising for Paul Lodge, Department for Work & Pensions)
Mark Brewin (HM Revenue & Customs)
Siobhan Carey (NISRA)
Chris Dibben (Independent member)
Tricia Dodd (Independent member)
Andrew Garrett (Independent member)
Emma Gordon (UK Research and Innovation)
Roger Halliday (Scottish Government)
Sarah Henry (ONS)
Sarah Mathieson (Independent member)
Alex Singleton (Independent Member)
Misa Tanaka (Independent member)
Andy Wall (Chief Security Officer, ONS)
Ross Young (Data Protection Officer, UK Statistics Authority)
UK Statistics Authority
Andrew Austin (Security and Information Management, ONS), for item 4
Bill South (Research Services & Data Access, ONS), for items 4 and 5
Karen White (Data Governance, Legislation & Policy, UK Statistics Authority), for item 6
Stephanie Howarth (Welsh Government)
1.1 The Chair welcomed the members to the twentieth meeting of the Research Accreditation Panel.
1.2 Members approved the minutes of the meeting held on 7 December 2020
1.3 Lily O’Flynn updated the meeting with progress on actions from previous meetings. All actions were complete or otherwise in progress.
1.4 At the December meeting, ONS colleagues were actioned to provide an update to the RAP on progress towards recommendations following the ONS’s DEA processor annual review. As the ONS is now working to create an operationalise the Integrated Data Programme (IDP), which may include technological and policy changes that impact the ONS’s current DEA accreditation, it was agreed that the ONS will outline its plan for the IDP June RAP meeting, including what this means for the ONS and SRS’s accreditation as processors under the DEA.
ACTION: ONS representatives to present ONS’s plans for the IDP and its meaning for ONS’s accreditation as a processor under the DEA at the June RAP meeting.
- Updated Terms of Reference
2.1 Lily O’Flynn presented the Panel with the latest update to the Research Accreditation Panel’s Terms of Reference, as the RAP is required to review its terms of reference annually. The following changes were reflected in the updated Terms of Reference:
i. Changes to the project accreditation process agreed at the December 2020 meeting, whereby projects are now considered via correspondence by default;
ii. The RAP is now required to only meet quarterly;
iii. Previous references to the ONS SRS have been updated to include reference to all DEA accredited processing environments;
iv. Reference to the precedent process has been removed, as recently agreed project accreditation process have superseded use of the precedent process; and,
v. The membership of the RAP has been updated to reflect current membership.
2.2 The RAP requested that the new Terms of Reference make clear that the Panel assesses each project’s public good individually, without comparisons with other ongoing research. Subject to this final addition, the Panel approved the new Terms of Reference and agreed for these to be published.
ACTION: The Secretariat to update the new Terms of Reference to clarification on public good judgements. The Secretariat to publish the updated Terms of Reference on the UK Statistics Authority website.
2.3 The RAP requested that the UK Statistics Authority appoints an individual with expert knowledge of public authority health data to the Research Accreditation Panel. The Panel noted that data held by public authorities for the provision of health services or adult social care is outside of the DEA Research strand’s scope, however recognised that health-related projects are enabled in part by the DEA whereby multiple legal gateways are used to provide access to DEA and non-DEA data in a single project. The RAP also quested that a health data governance body is invited to present at a future RAP meeting to explain the process, governance and legal basis regarding health datasets which are not specifically covered by the DEA.
ACTION: The Secretariat to work with the Chair and the National Statistician to appoint a health data expert to the Research Accreditation Panel. The Secretariat to invite a representative of a health data governance body to a future RAP meeting to inform the RAP on the process by which health data access requests are reviewed and approved outside of the DEA process.
- Project Accreditation: New Process Feedback Discussion
3.1 Lily O’Flynn presented the Panel with an overview of the benefits and successes of the new project accreditation process via correspondence, which was implemented following the RAP’s agreement at the December RAP meeting. The following points were raised:
i. Project accreditation decision times were reduced from 6 weeks to 8 working days on average, which improves the accreditation service for the research community through enabling more timely access to data;
ii. Early feedback from the ONS SRS suggests that this process allows processors to work in a more streamlined fashion and manage applications more smoothly, due to the removal of strict deadlines for meeting papers; and,
iii. The Secretariat recognises that Panel members have consistently provided extensive and useful comments through this process, which provides added value for researcher that use the DEA through the ability to benefit from the Panel’s expertise.
3.2 The Chair thanked RAP members for their patience and commitment as this new process was operationalised. Lily O’Flynn informed the Panel that, due to a number of highlight calls for new datasets that went live in early 2021, the number of projects submitted to the RAP for consideration each month is likely to increase further. In the short term, to prevent a bottleneck in the number of projects requiring RAP consideration, the RAP members agreed to increase the number of projects seen via Confluence per month. In the long term, the Secretariat agreed to undertake a piece of work to understand how anticipated levels of project applications can be managed efficiently, without creating a significant workload for RAP members.
ACTION: The Secretariat to undertake a piece of work to understand how anticipated levels of research projects can be managed efficiently, while maintaining appropriate scrutiny.
3.3 The Panel welcomed the benefits of the new system and recognised that members now have more time to review project applications, therefore enhancing the quality of accreditation decisions and comments on research project applications. The RAP raised the following as areas for improvement:
i. The ability to flag to the Secretariat when a RAP member is unavailable for project review;
ii. Ensuring all automatic project notifications are correctly received;Where possible, matching Panel members’ skillset knowledge and expertise to projects for review, whereby expertise could improve a project’s methodology and analytical techniques;
iii. Increased ability to view and track all projects that are being considered by the Panel at a given time; and,
iv. Increasing the clarity of the regional/national scope of projects (e.g., GB or UK).
The Secretariat agreed to provide both short- and long-term solutions for the issues raised by the Panel.
ACTION: The Secretariat to implement short- and long-term solutions for the above issues raised by the RAP in relation to the new accreditation system via correspondence.
3.4 It was recognised that the majority of data made available under the DEA gateway continues to be ONS data. Given the successful developments in improving the efficiency of services offered under the DEA gateway, the RAP considered why other government departments continue to use legacy department-specific legal gateways to make data available for research. The Panel commissioned the UK Statistics Authority to undertake a review of the DEA Research strand to further understand the reasons for limited use of the DEA Research powers across government departments.
ACTION: The Secretariat to undertake a review of the operationalisation of the DEA Research powers, focusing on both success thus far and opportunities for improvement that may impact wider government use of the DEA Research powers. The Secretariat to present this review at a future RAP meeting.
3.5 Lily O’Flynn presented the Panel with research project 2021/012: Investigating epidemiological insights for the COVID-19 infection across the UK, which required further review by the Research Accreditation Panel, following initial consideration via correspondence.
3.6 The project proposed the addition of academic partners to the research project on an ad-hoc basis to assist with COVID-related analysis to support government decision-making during the pandemic. The RAP was not content to accredit this project without knowing which organisations will be accessing data as part of this project in the future. It was agreed that the RAP must retain oversight of the project scope and data accessed as part of all accredited project, to uphold the required oversight and governance of data accessed via the Research strand of the DEA.
3.7 The RAP accredited the project, and wished to make clear that access to data under this accredited project should be limited to only those listed on the application, for the defined research purposes that are set out in the application. The RAP acknowledged the potential requirement for additional future research in this area by the lead organisation and its academic partners, especially as the COVID-19 pandemic and its impacts evolve. The RAP requested that future, related uses of this data or additions to this project are subject to the RAP’s existing data governance arrangements, which are expediated by default through project accreditation via correspondence, so as to ensure timely access to data for public good research.
ACTION: The UK Statistics Authority to discuss the Panel’s accreditation decision with the lead research organisation undertaking this project.
- Processor Accreditation: SAIL/UKSeRP Annual Review Report
4.1 Andrew Austin (UKSA Security) and Bill South (UKSA Capability) presented the Panel with an evidence report of SAIL/UKSeRP’s first DEA accredited processor annual review. The Accreditation teams recommended that the RAP continues SAIL’s accreditation for the preparation and provision of data and UKSeRP’s accreditation for the provision of data under the DEA. Levels of compliance with the security and capability controls selected as part of this review have either remained at the level ‘Good’ or progressed to ‘Mature’.
4.2 The Panel agreed to validate the continued accreditation of SAIL/UKSeRP under chapter 5 of part 5 of the Digital Economy Act, based on the evidence provided in the annual review report.
ACTION: The Secretariat to write to SAIL/UKSeRP to confirm the continuation of accreditation under the Digital Economy Act, following the successful completion of this annual review.
4.3 Andrew Austin and Bill South provided the RAP with an update of the potentially upcoming accreditation reports for NHS Wales Informatics Service (NWIS) and National Records Scotland (NRS). UKSA is continuing to engage with these organisations, for which DEA accreditation has been prioritised to enable the flow of ONS held COVID-19 Infection Survey data for analysis in the Devolved Administrations, under the Research strand of the DEA.
4.4 As the RAP now meets on a quarterly basis, the RAP agreed to review these accreditation reports via correspondence to enable timely review, accreditation and access to data available under the DEA Research powers, where applicable. The Panel agreed that RAP members can request that the accreditation decisions are deferred to a full meeting of the RAP in cases where additional scrutiny is required.
ACTION: The Secretariat to make the NWIS and NRS’s accreditation reports available for review via correspondence in the intervening period between quarterly RAP meetings to enable expediated accreditation, where possible.
- Processor Accreditation: Update on SRS and UK Data Archive Secure Lab Home Access Arrangements
5.1 Bill South (ONS SRS) provided the Panel with an update on the process by which DEA accredited processors ONS SRS and UK Data Archive Secure Lab enable home access to secure environments throughout the pandemic, due to limits on travel and social distancing restrictions.
5.2 Bill South provided the RAP with assurance that the way in which the ONS SRS and the UK Data Archive had operationalised home access during the pandemic was compliant with the requirements set out in the DEA’s Research Code of Practice and Accreditation Criteria.
5.3 The RAP welcomed this assurance and supported the ongoing work across accredited processors to facilitate the continuation of vital research throughout the pandemic. The RAP agreed that the UK Statistics Authority should work with accredited processing environments to bring together knowledge on such new processes that have been operationalised to support ongoing public good research during the pandemic, to enable the wider research community to learn from such measures, and continue to work to facilitate better access to data for research.
ACTION: The Secretariat to work with accredited processors to draw together key learnings from continuing to enable, and improve, access to data for research throughout the pandemic.
- Research Code of Practice and Accreditation Criteria Compliance Review: Update and Forward Plan
6.1 Karen White, in the Data Protection Compliance team at the UK Statistics Authority, presented the RAP with initial findings and a forward plan from the Research Code of Practice and Accreditation Criteria Compliance Review, commissioned by the RAP at the October 2020 meeting.
6.2 Karen White assured the Panel that processes continue to be operationalised to manage the accreditation of processors and projects under the Research strand of the DEA, which are compliant with the Research Code of Practice and Accreditation Criteria. The RAP welcomed the following recommendations which were identified as part of this review to help improve processes further:
i. Revision of the DEA processor accreditation application form to transparently illustrate how the DEA accredited processor controls align to the Research Code of the Practice and Accreditation Criteria;
ii. Implementation of a standardised feedback approach to DEA accredited processor site visits and reviews, to provide additional clarity to processors as to why they have been graded as either ‘good’ or ‘mature’ across controls reviewed; and,
iii. Potentially periodically inviting an independent expert to advise on the DEA accredited processors’ application and accreditation process to continue to learn from best practice and show the DEA process’s commitment to upholding organisational best practice.
6.3 The UK Statistics Authority agreed to provide the RAP with an updated Compliance Review Report at a future meeting of the RAP to provide assurance that report recommendations had been implemented through management actions.
ACTION: The UK Statistics Authority to provide an updated Research Code of Practice and Accreditation Criteria Compliance Report to a future meeting of the RAP.
- Any Other Business
7.1 The RAP welcomed confirmation that the UK Statistics Authority had agreed to extend the provisionally accredited researcher accreditation period from 1 year to 3 years, following a proposal by the ONS which was supported by SAIL, UK Data Archive and NISRA. This change allows researchers the time for build up the required skills and experience to apply for full accredited researcher status (5 years) by the time provisional researcher status has elapsed.
7.2 Simon Whitworth informed the RAP that the UK Statistics Authority recently launched the Centre for Applied Data Ethics to provide applied data ethics guidance and support for researchers accessing and using public authority data for research and statistical purposes. The Panel welcomed this development and requested further updates as this work develops.
7.3 The Panel noted the report of ongoing accreditation processes undertaken outside of Panel meetings, presented for information.
7.4 The next meeting of the Research Accreditation Panel is on Tuesday 8 June 2021.