The purpose of the Strategic Case is to provide strategic justification for the Project, including outlining the background and associated objectives of the Project, as well as assessing stakeholders, opportunities and risks.
The chapter is set out in the following sections:
1. Strategic Context: Drivers for Change and Opportunities
2. Project Background
3. Mission Statement
4. Founding Principles
5. What RDS will Do and Deliver: Scope. Required organisational capability
6. Benefits of RDS
7. Business Needs – Responding to Covid-19 research requirements
8. RDS Delivery Programme
9. Strategic Risks
10. Constraints and Dependencies
1. Strategic Context: Drivers for Change and Opportunities
Scotland has a rich history of public sector data informatics, which has served to enrich and inform our most valued public services. Collaboration in Scotland between academia and the public sector has supported a national model of cross-sectoral research which has delivered path finding innovation and won numerous civil service awards, delivering key insights and furthering our understanding of some of the most pressing public policy issues.
While this collaboration has proven successful, it has grown out of the efforts of a small number of specialist technical teams and academics across Scotland to make it work. This has made it less resilient to change as data science and innovation have progressed and the system needs to grow and flex if it is to stay relevant and fit for purpose to meet the needs of a growing community of users.
Joined Up Data for Better Decisions
In 2012, following public consultation, the Scottish Government published Joined-Up Data for Better Decisions. This Strategy set out Scottish Government’s ambitions for making better use of existing public sector data sources. The focus was on building outward from the strengths and successes of health data informatics.
The Data Linkage Framework was established as a mechanism to deliver this Strategy and set out three core ambitions:
i. To build on existing successful programmes collaboratively to create a culture where legal, ethical, and secure data linkage is accepted and expected;
ii. To minimise the risks to privacy and enhance transparency, by driving up standards in data sharing and linkage procedures;
iii. To encourage and facilitate full realisation of the benefits that can be achieved through data linkage to maximise the value of administrative and survey data.
A set of Guiding Principles were consulted on and published alongside the Strategy and laid out a consistent decision making framework for data controllers and others involved in data linkage for research and statistics. These are still valid today and support all of the data linkage activity that takes place.
While the aims have not changed – the ambition has. Projects that have delivered policy insight have only served to increase interest in the use of administrative data and linkage methods to address public policy challenges. Fiscal consolidation has further heightened the need to look beyond expensive longitudinal surveys and studies toward more efficient and cost-effective methods of answering research questions.
The onset of the Covid-19 pandemic and urgent need to progress the scientific evidence relating to the virus and its transmission has also reinforced the need to transform the existing service offering. The system faces a variety of concurrent challenges that require solutions; some of these are new – others, identified earlier, require fresh thinking and renewed commitment.
Office for Statistical Research: UK Statistics Authority
In 2018, the Office for Statistical Regulation published their systematic review of data linkage, Joining up Data for Better Statistics, which set out to identify key priorities and areas for improvement to the way the UK statistical system makes use of data linkage methods to deliver public policy insight. The report identified six key outcomes designed to deliver an effective and safe data linkage system and evaluated the existing services against these. The report celebrated success stories and found impactful examples of data linkage being used to inform policy making; however it noted that this was the exception rather than the rule. It warned that the time and effort required to deliver projects was putting people off relying on this as a core research method.
In Scotland as elsewhere, it takes too long to access and link diverse datasets, for a range of reasons, and information about these datasets is not readily available. Data are dispersed both between and within public sector organisations and are not always available in a linkage-ready format.
Furthermore, the system of information governance is not designed to deliver projects that draw in data from multiple data controllers. This also leads to delays in data being available for research and innovation. For academia, this means we are not securing a suitable share of the available UK research funding, and for public bodies, this means that they do not have the data to support public service reform.
Scotland is fortunate to have some of the best data in the world. Some data are about people; their characteristics and behaviours. Other data are non-personal; such as data about climate or pollution. Bringing this data together can help address complex social and environmental issues and fill vital evidence gaps. Scotland also possesses world-leading expertise, particularly in our universities, in ethical, legal and social disciplines, as well as resources and skills in data infrastructure, data management, analysis and informatics.
Scotland is therefore well placed to harness the value and benefit inherent in our public sector data assets to deliver better outcomes for the people of Scotland whilst safeguarding the privacy of individuals.
There is a real opportunity for RDS to serve as a catalyst in unlocking the social value inherent in our data assets and research. This is coupled with potential economic benefits too; a recent study for Scottish Enterprise suggested data innovation could potentially benefit Scotland by £20bn. A report on the value of big data to the UK Economy for the Centre of Economics and Business Research (CEBR) identified similar value. Using data better supports improvements to society, productivity and organisational efficiency, attracting new businesses and highly paid jobs to Scotland. The average salary of a data professional in Scotland in 2018 was £50,000. However, due to delays in provisioning data, investment that could be happening in Scotland is beginning to flow elsewhere.
The current data linkage set-up has delivered effective research flexibly; however several challenges remain unaddressed and are further exacerbated as new demands are placed on the system. The most pressing of these challenges are set out here.
Uncoordinated data governance processes and structures
Policies and structures supporting information governance (IG) functions have come into existence over time in response to data protection and other legislation. However, this has given rise to a system that is not set-up to support cross-sectoral research. IG policies are perceived as a hurdle rather than as a springboard to ethical research in the public interest. This perspective needs to shift.
This is particularly the case where different data sources are linked for a single project, requiring several un-coordinated data access processes. This has resulted in difficulty knowing who owns the data of interest with decision-making and data linkage projects taking too long to progress, and timelines difficult to predict. There is a need to better co-ordinate and streamline data availability processes and promote sharing of best practice across the community of public sector data controllers. This builds trust, expertise and experience.
The public also hold mixed views about the use of their data in research. We need to ensure there is ongoing engagement, trust, support and feedback from the public.
Data is often of unknown or poor quality; lack of information about data
It can be unclear what public sector data are available for use in research and data can be of unknown or poor quality. This means that some data are collected more than once and significant effort is expended to find out where helpful data is and the route for access. We need to work with data controllers and users to improve the quality for research use.
There is an increasing demand for more recent and real-time data. This poses challenges for our infrastructure due to the investment and automation required in securing that data, as well as ensuring the quality of this information.
Projects require a lot of effort in preparing data where this is not already done. This means that the Business Case for using existing data sources rather than collecting fresh data is less clear-cut in many cases, again, a barrier to helpful research happening.
Less mature service model for cross-sectoral data linkage research
It can take a long time to assemble data for cross-sectoral research projects and be expensive: we need to address this such that it becomes a standardised approach. This means fitting a service model around such needs and supporting the skills, training and resources to deliver on this demand.
To date, to get projects over the line, has required several teams stepping in to support delivery with valuable and diverse skillsets but the result is a process where the roles across service providers become blurred. This leads to inefficiencies and has meant reduced resilience within each of those functions and some skill and technology deficiencies that make the current landscape sub-optimal.
Bilateral/separate commissioning arrangements in Scotland
Existing funding and commissioning arrangements for the data infrastructure and data linkage service model comprise largely of a set of bilateral arrangements and financial transfers between the various funders and the various recipients (service providers) of those funds. This arrangement does not lend itself to ready appraisal of the cost-effectiveness of the system as a whole in meeting its objectives, or of the individual parts therein. Alignment of the various funding sources through RDS with single contracts for activity with each service provider is expected to promote more efficient and effective service provision, to better align incentives and to realise efficiencies.
Ongoing feedback from stakeholders has highlighted several challenges and communicated expectations around data provisioning and access for research in Scotland. These views were further captured as part of a Discovery Phase to inform this Business Case. Some of the issues raised consistently by stakeholders, and that RDS is seeking to address, are as follows.
- It is unclear which administrative datasets are available for use in research and information is not readily available to potential users about what the process to gain access to the data involves.
- There is increasing demand for more recent and real-time data. This is currently difficult due to the investment and automation required in securing that data, as well as ensuring the quality of this information.
- Data are not always “linkage-ready” so projects require a lot of preparatory effort; some of this upstream work and curation could be done earlier so that at the point a researcher expresses interest in using the data, it is clear what is available and how that can be used.
- Assessments of the privacy/public benefit and ethics are not co-ordinated and researchers are left to navigate several processes for a single programme of work. This duplicates effort and can introduce delays in decision-making.
- There are several legal frameworks in the data and data protection space including common law, Digital Economy Act 2017, Human Tissue Act 2006, Human Rights Act 1998, Data Protection Act 2018, and UK GDPR – and their application and interaction is not always clear.
Data linkage service providers (NRS, EPCC, eDRIS/PHS)
- While there has been funding for data linkage services in recent years and many successful data linkage projects have been delivered, there has not been a sustainable business model that enables services to invest in future proofing, expands the range of data that is available for use in research, or creates ongoing efficiencies and innovation.
- The lack of distinct legal footing of the Scottish Informatics Linkage Collaborative (SILC) from its partner organisations and the collective of people, resources, technologies and platforms means that there has not been the accountability, transparency and openness which is a required of a modern efficient service.
- Government researchers and analysts need access to data too in a timely and cost-effective fashion.
- As regards data held by Government and the public sector there are also clear costs savings in terms of enabling data access via RDS instead of burdening individual data controllers with many individual requests.
- A centralised system of access/brokering point will ensure greater consistency of decision-making about what public data can be used for, who can access it, as well as more consistent management of directorates/departments risks.
- It will promote further confidence in the data hosting and provisioning system in Scotland – to seek and attain accreditation through the UK DEA for use of the National Safe Haven (NSH) to host data from UK Reserved Departments such as HMRC and DWP.
- Under RDS data controllers will have the opportunity to host linkage-ready datasets in the national safe haven (NSH).
- RDS will work with data controllers to agree clear, efficient and proportionate data access arrangements and IG processes.
- RDS will also facilitate access to non-linked datasets – for both individual level and aggregate level data, thereby relieving burden on data controllers to service these requests.
- It is anticipated that these changes will deliver efficiencies and economies of scale and will enable ongoing learning and testing of IG processes.
- Provide resources and/or skills to facilitate the holding and curation of data which some data controllers currently lack, hence: access to data is often seen as a costly and time-consuming operational exercise.
- A centralised system of access will foster greater consistency in decision-making about what data can be used for, who can access it, and how the risk management of the data can be undertaken more consistently.
- A clearer offering for data controllers in terms of benefits they will derive from sharing their data, for example by supporting a specific and well-identified research or policy challenge; helping with own evaluation and assessment /audit requirements (e.g. statutory and non-statutory); and improving operational effectiveness and service improvements.
Data professionals: statisticians/data scientists
- Improved coordination of data access and better data provision will enhance the opportunity to address important research and policy challenges and will therefore encourage more people to work with data by building capacity and capability in skills and experience that is in demand by Government and businesses, but currently in short supply.
- Building wider data science skills will benefit the wider economy.
- Building and enhancing skills capacity in the data space will also support social scientists to develop quantitative skills which is vital for the development of social science and maximise the wealth of data which already exist.
2. Project Background
Scottish Informatics Linkage Collaboration (SILC): Business As Usual Service Model
The Scottish Informatics and Linkage Collaboration (SILC) came together in 2014 with the participation of several public sector and research council funding partners. Its purpose was to realise the vision of the Data Linkage Framework through delivery of a shared service model to support public benefit research using data linkage methods in Scotland. It aimed to promote collaborative cross-sector working, the sharing of best practice and joined-up approaches to resource investment across public bodies operating within the research data landscape. Oversight was provided by the SILC Senior Management Board (SILC-SMB).
This investment supported the set-up of the state-of-the-art facility with associated national infrastructure located at No. 9, Edinburgh BioQuarter. Both Scottish nodes of the Farr Institute and Administrative Data Research Centre (ADRC) located with eDRIS, and further funders joined at later dates. The SILC-SMB acknowledged the importance of establishing a legacy for SILC with respect to the grant-based funding mechanisms of many of its component parts, and ensure its future success as a national resource.
The model comprised a set of shared resources that reflect closely how the service model delivers currently. This comprised the eDRIS team of research co-ordinators, NRS indexing team, FARR IT infrastructure located now at the EPCC.
SILC SMB discussion about the need for change, SOC to OBC
Following discussions among senior stakeholders and funders participating in the Scottish Informatics and Linkage Collaboration (SILC) and taking on board feedback from users of the SILC resources, a sub-group of the SILC-SMB was convened to scope priorities and options for transforming the system supporting public benefit research in Scotland, with the aim of addressing some of the highlighted challenges and to ensure any solution was fit for the future.
This work rehearsed the shortcomings of the existing legal and commissioning arrangements and explored how these might best be addressed. The Board reached consensus on the need to place the current national arrangements supporting cross-sectoral research on a more sustainable and formal footing and to ensure the service model within this possesses the necessary resources, skills and infrastructure to meet changes to demand, technology and legislation.
It was agreed that a key priority would be provision of a single-entry point for researchers and offering a seamless data access pathway, by bringing together and coordinating the various moving parts of the researcher access journey, including the commissioned services.
This culminated in a Strategic Outline Case (SOC), which explored the initial long list of options to determine the purpose and feasibility of establishing RDS on a more formal footing.
The strategic objective is that the future offer from Scotland becomes much stronger, with public benefit research undertaken at scale. This strategic Case outlines the model proposed under RDS.
3. Research Data Scotland: Mission Statement
The proposed mission statement for RDS is as follows:
Research Data Scotland will provide a service for accessing public sector datasets that have the potential to save time, money, and lives. It will offer safe, secure and cost-effective access to data for research, innovation and investment by enabling its users to deliver insight and understanding that will help create a more successful country through increased wellbeing, sustainable and inclusive economic growth, and improving the health and social care of the nation. We will work collaboratively with data controllers and users to develop the service while building trust and support from the public.
4. RDS Seven Founding Principles
Supporting the achievement of the mission statement are seven key founding principles:
1. RDS will only enable access to data for research that is for the public good.
2. RDS will ensure that researchers and RDS staff can only access data once an individual’s personal identity has been removed.
3. RDS will ensure that all data about people, businesses or places is always kept in a controlled and secured environment.
4. RDS will only create a dataset if it is requested for a research programme or study that is in the public good.
5. All income that RDS generates will be re-invested into services to help researchers continue to access data.
6. RDS will secure a share of commercial benefits from any firms that access our public data, ensuring benefits are returned into public services.
7. RDS will be transparent about what data it provides access to and how it is being used for public benefit.
5. What RDS will Do and Deliver
RDS will provide a “one-stop-shop” for access to a range of services and resources aimed at supporting access to data about people, places and business in Scotland.
- Help facilitate the creation of “linkage-ready” versions of high value datasets and provision for key metadata and data access criteria for each dataset.
- Deliver a service for researchers, assisting them with research design and providing secure access to “linkage-ready” datasets, with the flexibility to link to other data where required or to be used as standalone datasets in their own right.
- Commission and monitor an IT infrastructure to securely transfer, store and provide secure access to datasets, allocating resources to three services; high performance computing, indexing and customer support.
- Stay in touch with technical and methodological developments to continuously improve the service seeking ongoing feedback on performance and progress from stakeholders.
Research Data Scotland will build upon our regional and national data informatics expertise, to:
- Enhance the eDRIS service that already delivers hundreds of health and non-health data access requests and linkage projects each year.
- Utilise the Edinburgh International Data Facility (EIDF) being developed at the University of Edinburgh Parallel Computing Centre (EPCC).
- Use expertise for data indexing that exists at the National Records of Scotland (NRS).
- Build on the national data infrastructure being developed by SG: the underpinning of data policies, standards, legislation, approaches to ethics and information governance, and arrangements for cyber resilience.
- Work to make our National and Regional Data Safe Havens interoperable.
While longer term, it will be important to use data from both public and private sectors in research, our initial focus will be on getting arrangements working well for data collected by any part of the public sector providing services to people in Scotland, and should cover both linked and standalone de-identified datasets. There must be secure and carefully controlled access for different types of users from various sectors (NHS, SG, academics, third sector and industry).
Building upon existing arrangements, critical success factors for the RDS service are to:
- Have the support of the public, acting in an open and transparent way
- Deliver value for money
- Improve the service quality to users, both delivering faster and more reliably, responding to the needs of different types of user
- Foster strong relationships with data controllers, acting under clear and consistent information governance processes
- Comply with all legal requirements and protect the privacy of citizens and businesses. This will build upon the 5 safes principles (safe projects, people, settings, data and outputs).
Trustworthiness will be at the heart of everything RDS does and this includes:
- Maintaining the security and privacy of the data by removing personal identifiers
- Holding data in a very secure place
- Only allowing access to accredited researchers for linked datasets.
6. Benefits of RDS and Spending Objectives
RDS will not address all of the current challenges inherent in accessing data for research; however the approach to strengthen institutional capability, investment and profile, will enable it to work with other public sector partners on these issues.
- RDS will be able to commission services in its own right. This will improve co-ordination and promote a more effective and efficient system.
- The establishment of RDS will address the issue of service quality, working to improve the end-to-end user journey, addressing strengths and weaknesses of current processes.
- RDS will also facilitate the sharing of information about what data is available and the quality of that data via an interactive website and service.
- To provide underpinning investment for research at greater scale than data linkage current model delivers – more linkage projects through each year and projects progressing more quickly.
The RDS team and its stakeholder partners agree that the following indicators will demonstrate successful delivery against scope:
- Service – consistency in the level and quality of service
- Safety and security – compliance with legal, IG and ICT requirement
- Public trust and transparency – people trust their data is used appropriately
- How people’s data is processed will be clear and readily understandable by users and providers of services as well as the public at large
- Timeliness – data access is streamlined and efficient
- Sustainability – sustainable model operationally and financially
- Cost-effectiveness – efficient use of resources. Take Value for Money (VfM) into account when making decisions, and allocating roles, responsibilities and resources amongst service delivery partners
- Accountability – RDS held to account through external scrutiny and audit adopting a strategic, proportionate and risk-based approach.
7. Business Needs – Responding to Covid-19 Research Requirements
The vision is for a partnership, initially between SG, PHS and the University of Edinburgh, which accelerates the realisation of public value and economic advantage through data driven research and innovation. Shortcomings of the current data linkage service arrangements are likely to become more acute over time. The establishment of RDS future-proofs capability for both data linkage and non-linkage projects, addressing future business needs by thinking longer term about service requirements, resource, investment and infrastructure.
Present health challenges reinforce the need for a more strategic approach at a time where data about people, places and businesses has never been more important, supporting Government in assessing the full impact of Covid-19; evaluating strategies for re-opening the nation; and safeguarding people’s health and wellbeing.
RDS will provide a single system approach to realising greater social benefits (via research) from existing public sector datasets, bringing these together in novel ways to respond to new research questions and gaps. It is appreciably a lengthy journey to source, clean, link and assemble data for individual research projects and there are clear gains to be had from coordinating this work to serve many users, thereby realising/recouping the investment in data quality, curation and service development.
Public sector data possesses some of the attributes of a public good making it unlikely that the service of RDS will be provided by the market at the scale required to realise full social and economic benefit – that highlights the market failure. So RDS facilitates the unlocking of this potential, enabling research and acting as a trusted institution that bears risk (reputational, privacy, financial, security, cyber).
Establishing RDS will also harness the potential of the Edinburgh City Deal to foster innovation and development in Scotland and will support our ambition to make Scotland a data destination, attracting inward investment which is currently going elsewhere. A breakdown of the economic benefits is provided in the Socio-Economic Case.
8. RDS Delivery Programme
The RDS Delivery Programme comprises six integrated elements, each with a separate delivery work-stream.
1. Public engagement and provision of clear information to citizens about the use of public sector data in Scotland.
2. Transparent IG processes and procedures, these must incorporate public sector data controllers’ requirements.
3. Provision of a secure de-identification and data linkage service.
4. Provision of a secure high performance computing environment.
5. Provision of a one-stop-shop user service.
6. High-quality data assets.
These expectations will be delivered via a programme of activities covering:
- Public trust and transparency
- Commercial and procurement
- Safety and Security
The development of RDS is a specific part of the Scottish Government’s Programme for Government.
9. Strategic Risks
The RDS project will face several strategic risks including:
- Service – complexity of the “as is” model and a lack of widespread understanding as to how this works. This will require further work to understand the current service journey. Stakeholder engagement will be an important means of ensuring consistency with the level and quality of service and confidence with the new process, whilst ensuring data access is streamlined and efficient.
- Public trust and transparency – people must trust that their data is used appropriately, how data will be processed will be made clear and transparent through public engagement with users and providers of the service as well as the public at large.
- Cost-effectiveness – review of service resources to ensure VfM taken into account during decision making, allocation of roles, responsibility and resource capacity.
- Commercial and procurement – the model must be sustainable operationally and financially. An overview of the likely procurement arrangements for the preferred option is include in the Commercial Case with more details in the subsequent FBC
- Safety and security – as data will be involved in the role of RDS, sensitivity and transparency is required as concerns from data controllers around privacy, security, data storage and sharing need to be managed
- Financial – developing, implementing and supporting RDS requires agreement on longer-term funding models. That model should be acceptable to all while delivering value for money. The proposed model is documented in the Financial Case.
- Accountability – RDS will be held to account through external scrutiny and audit adopting a strategic, proportionate and risk-based approach
- Legislation – (UK and Scottish) RDS will liaise with legal teams and other stakeholders to consider perceived legislative or statutory constraints, and follow-up action where needed.
- Covid-19 – key staff across the stakeholder organisations will be diverted to responding to the pandemic in the short and medium term.
The RDS project team has developed a detailed risk register and is managing the risks associated with the establishment of RDS. A risk management plan will be included in the FBC, with a defined process for handing over risk to the newly formed RDS.
10. Constraints and Dependencies
RDS will operate within a framework of regulations governing the use and processing of personal data for research: (Digital Economy Act 2017), physical and cyber security (replacements for EU ENISA regulations, EU NIS Directive) and data protection law (Data Protection Act, UK GDPR), and other legal codes (Human Rights Law, Human Tissue Act), common law duty of confidentiality.
A key dependency is the development of the Covid-19 Research Data Service, that has been brought together from across the existing service partners and that will be commissioned under RDS. The need to urgently address priority Covid-19 research projects has drawn on existing resources and people in the short term, displacing other activity that would have taken place in its absence. While this is inevitable, there will need to be some form of debrief and review to reconsider funding requirements in order to ensure service transformation work is progressed.
A wider Covid-19 Data and Intelligence Network across the Scottish public sector is overseeing the development of a strategy to address the data requirements for the pandemic, covering both use for research and for service delivery and operations, including the information governance and information assurance frameworks for making the data available via bespoke infrastructure.
It will be important to seek clarity from service partners and stakeholders in SG and more widely about how this work interacts with RDS; where this network addresses some of the identified challenges and where it may introduce new tensions.
ADR-Scotland, Edinburgh South East Scotland City Deal
Two other key initiatives will support the delivery of RDS.
The first is partnership with Administrative Data Research UK which is the UK wide data acquisition and research programme funded by the Economic and Social Research Council (ESRC). Its aim is to develop capacity and capabilities that support innovative and cutting-edge research in the public interest using existing administrative data sources. The ADR UK investment in Scotland is administered through the Administrative Data Research-Scotland partnership.
The ADR delivery team within Scottish Government have been designing and building a new method for linking personal data within the Scottish National Safe Haven. Working with partners from the University of Edinburgh and the electronic Data Research and Innovation Service (eDRIS – part of Public Health Scotland (PHS)) a new ingest process has been designed to securely hold and manage a wide range of datasets from across the Scottish public sector.
Secondly, the investment into data-driven innovation through the Edinburgh City Deal and partnership with the Edinburgh Parallel Computing Centre will ensure RDS can harness wider resources and secure the momentum for a wider partnership between public sector and academia, building on existing data infrastructure, and Scotland’s reputation for how we manage and use data and trustworthiness in what we do.
Both the ADR-S programme and the Edinburgh City Deal taken together will kick-start and support the delivery of a new data linkage user service provided through RDS.
The existing services and investment supporting public benefit research in Scotland have delivered valuable insight through ground-breaking programmes, embedding a culture of evidence-based policy making and extending our understanding of the most complex of public policy issues facing Scotland.
However the contractual and financial arrangements that supported the initial phase of delivery of the partnership over the last seven years require to be formalised in order to take advantage of the scale of opportunity now facing Scotland. This requires a step change to the way services are supported and contracted for; and to the way the system and its parts work toward shared and agreed collective outcomes.
We need to place these arrangements on a more formal footing, to attract significant new investment and promote a more efficient and scaled-up model that can respond to future needs. The opportunity cost of not responding to this growing demand for change and reform is significant including both the direct impacts on research and public services – and the indirect impacts on the Scottish economy if a healthy and growing data innovation sector fails to take off at scale.
A Ministerial commitment in the 2019 Programme for Government, the establishment of RDS will provide a researcher support service across the whole public sector and ensure that the service is sustainable and resilient.