Care Home Data Review - Full Report

Data Sharing

Current data sharing

Feedback to the review was clear that, while some data sharing does exist, we are not currently fully benefitting from the potential for data sharing data across organisations. As previously noted, the limited sharing between data producers has contributed to there being multiple data collections, leading to a confusing data landscape, increased data provider burden, and a less efficient use of both sector and analytical resource across organisations. Conversely, where we have invested time and resources in sharing data between national producers, we minimise duplication for data providers and often improve the quality and completeness of the data provided. Data providers also expressed frustration that they are not always receiving data/information shared back so cannot fully realise the benefits of data provision – which, they noted, de-incentivises provision.

Limited routine data sharing can also be a barrier to our ability to ensure that the right data can be accessed by the right people, in a timely manner, to inform decision making. Having safe, transparent and efficient data sharing is also key to improving coherence across the data landscape, reducing respondent burden and improving our understanding of, and outcomes for, those living and working in care homes.

Wider context

The data landscape around care homes is complex, with a large number of data providers supplying data to various local and national organisations, for many different purposes. This complexity can lead to a number of challenges for data sharing and create uncertainty if the sharing of some data satisfies legal requirements. Current data sharing arrangements in place may not provide for onward sharing or repurposing of data without being rewritten and agreed. It is critical that appropriate information governance is in place to ensure that any data sharing is legal, appropriate and secure, to ensure that the privacy of individuals is protected and there is confidence in the use of the data.

A number of reports have highlighted the potential reasons for, and issues caused by, a lack of data sharing between organisations^[37],[38]. The Information Governance Review carried out by Scottish Governance outlined the need for a a streamlined national information governance approach that addresses:

• Inconsistencies in decisions over the delivery and use of data and digital technologies
• Variation in interpretation of current regulations
• The risk appetite among organisations, which can obstruct or facilitate the realisation of benefits from digital and data-driven health and care innovation

In addition, the Office for Statistics Regulation (OSR) recently published a report on Data Sharing and Linkage for the Public Good. This report focuses mainly on data sharing and linkage within government for research purposes, however OSR note that their findings are relevant for the wider public sector. The report looks at 4 main themes:

• Public engagement and social licence: The importance of obtaining a social licence for data sharing and linkage
• People: The risk appetite and leadership of key decision makers, and the skills and availability of staff
• Processes: The non-technical processes that govern how data sharing and linkage happens across government
• Technical challenges: The technical specifics of datasets, as well as the infrastructure to support data sharing and linkage

The report uses these themes to produce a number of recommendations and a roadmap to better data sharing and linkage.

Data sharing and information governance are also key themes of the Data Strategy for Health and Social Care:

“We want a trusted, secure health and care ecosystem where data is shared, managed and stored securely, consistently, efficiently and transparently.”

The strategy notes that demonstrating our ability to collect, store and share data safely and securely are crucial to public confidence and effective delivery of care, whilst ensuring that data is available to the right people at the right time for clearly defined purposes.

The strategy also notes the additional complexities associated with social care being provided by a large number of different organisations, using different systems/standards and much of the social care data being generated outside of the public sector.

Impact at national level

Limited data sharing across data producers has contributed to the cluttered data landscape and has resulted in increasing the burden on data providers and duplication of effort amongst data producers, whilst potentially also creating inconsistent and sometimes contrasting evidence.

Part of the reason for this is it is important to have appropriate information governance in place to ensure confidence in the use and security of the data. This can take time to develop if there are multiple organisations involved and the length of time that it can take to agree data sharing arrangements can contribute to the need for separate collections, as organisations are unable to obtain data through sharing in the required timescales. This means that every organisation that requires to know information about residents in care homes asks for the data separately.

Data sharing was also seen as critical for the rationalisation of data collections, as by agreeing questions and definitions between national and local organisations / data collections, and sharing data between producers, the number of data returns could be significantly reduced.

The frustration with the current lack of sharing was clear:

“Data to be shared rather than telling 3 folk [different national and local organisations] how many beds are available in 3 different forms - phone, email and online.”

“[We need] robust data sharing agreements to allow sharing between organisations working towards the same cause.”

“Collect once and reuse many times.”

CHDR Workshop Feedback

Impact at local level - Inequality of data access

Data providers expressed frustration that it can be difficult to access data which the local authority or HSCP have provided. It was noted that part of the rationale for supplying the data in its current format is that it could be linked to other data from the area and information shared back to increase the wealth of information available for local/national decision makers. However, many local areas reported that linked health and social care data was not always available in an appropriate timescale / they had never received any linked data back. As noted above, the Local Intelligence Support Team (LIST) provide tailored support to HSCPs, which can include facilitating access to health and social care data as well as providing bespoke analysis.

Feedback to the review was that the data people need is not always accessible to them in an appropriate timescale. Delays with data sharing can mean that by the time data is available, the point at which it may influence a meaningful outcome has passed.

The future local government data platform will aim to deliver tangible benefits in terms of timeliness, by giving data providers and requesters a secure, real-time access to their own data. The data will also be enhanced, providing access to analyses and data visualisation directly within the platform, allowing Local Authorities to benchmark themselves with others, or explore trends over time. The platform will also aim to allow the linkage of datasets where possible, increasing the knowledge and insight generated from the data.

Local Authority, HSCP and care home data providers also noted that they are not seeing the full potential benefit of providing data as they noted that very little data was being provided back to them, for example for benchmarking purposes. Some care home providers requested a more meaningful feedback loop, as they noted they can “input data into the systems multiple times” without seeing any evidence of its use or impact on their day.

The lack of consistency in data access is highlighted in the British Geriatrics Society’s Smarter Data, Better Care report. The report notes:

“NHS and local authorities hold various data about people living in care homes that is often inaccessible to care home staff, and data collected within care homes goes unseen by external health and care staff supporting residents”.

In feedback to the review, care homes noted that having access to social work and health information on people using their services “would be transformative on outcomes.”

Analytical capacity

The current lack of data sharing restricts the analysis that can be done by both local areas to inform local decisions, and national bodies to inform planning and policy development. At present, the lack of sharing between organisations means that there is a siloed understanding of the different data sources, which both limits the understanding of the whole data picture and the depth and range of possible analysis.

Sharing of appropriate data could help harness the full analytical power of staff working on care home data across organisations and allow access to consistent, coherent and timely data across the system. There are a number of organisations / people who can analyse care home data and to maximise the benefits to be made from it, we need to ensure they have appropriate access to data in a timely manner. This would both help to fully utilise the analytical capacity in the sector, and also ensure all data users were using, and basing decisions on, consistent data.

Data sharing for research and innovation

The research and innovation sector have an integral role in providing analytical capacity, evidence and expertise to inform policy and practice, to ultimately improve care. To identify and answer meaningful questions about care homes, residents and the care home workforce, researchers need to be able to access good quality timely data and associated metadata.

National care home data which are aggregated or anonymised (to reduce the risk of being able to identify residents or workforce) are increasingly being made available via open data platforms, for example Scottish Health and Social Care Open Data platform and Statistics.Gov.Scot platform.

For more sensitive data, or data where there is a higher risk of identifying residents or workforce, the information governance landscape is more complex and time-consuming, with no single acknowledged pathway^[[39]. For requesting access for research purposes to nationally held data by SG, PHS, CI, and SSSC there are the following Information Governance procedures in place. For research/statistics data held by SG or PHS, researchers can apply to the Public Benefit and Privacy Panel for Health and Social Care (HSC-PBPP) or the Statistics Public Benefit and Privacy Panel (SPBPP) (Scottish Government data only). These panels have been established to scrutinise applications for public benefit and information governance requirements. However, it was raised during the review that despite the HSC-PBPP being in place:

“Ensuring there is sufficient expertise [on the Public Benefit and Privacy Panel for Health & Social Care] versed in the spectrum of data linkage research topics which are needed in Scotland to address key priority areas for health, but also for social care is essential to make use of the data we have and contribute constructively to improving our understanding of the needs of the population and challenges they face.”

For research/statistics data held by CI or SSSC, researchers can apply for access direct to these organisations.

For access to local care home data (for example data held by care homes), the information governance routes are more complex (care homes cover public, private, and not for profit sectors), there can be a lack of clarity around data ownership, and a lack of consistency around information governance^[40]. There can also be additional considerations around commercial sensitivities^[41].

Furthermore, depending on the nature of the research / innovation project, researchers may have to apply for additional ethical review (see Ethics chapter).

The gaps in the research infrastructure (e.g. a lack of research ready / linkage ready care home data, lack of clarity and consistency around information governance, and a lack of timely information governance integration) are hindering care home (& social care) research and innovation in Scotland. These have been highlighted not only in the published literature^[42][43][44], but also during discussions with stakeholders, and outlined in the current health and social care data strategy^[45].

Making data research ready

National data producers will work with Research Data Scotland (RDS) and Administrative Data Research Scotland (ADR-S) to make national care home (& social care) data more accessible via the Researcher Access Service. The Researcher Access Service is a digital service that makes it quicker and simpler for approved researchers to access public sector data for research. Launched by RDS in April 2024, the Service enables faster access to public sector datasets held in Scotland’s National Safe Haven – a Trusted Research Environment which securely houses some of Scotland’s most useful datasets for research.

Case study: Understanding pathways into care homes using data – data access challenges

In 2015 funding was secured from the Scottish Informatics and Linkage Collaboration to investigate a very practical applied question to better understanding pathways into care homes in Scotland. Scotland was thought to be well-placed as a location to undertake this work due to the existence of the annual national data resource, the Scottish Care Home Census (SCHC). Despite this however, the process was fraught with challenges and delays which have ultimately impacted on the timeliness of this important work.

The first challenges related to the need for a new data source, not previously used for linkage, to undergo indexing to the Community Health Index (CHI) database, so that individual-level data submitted to the SCHC could be joined to other national data sources. The technical skills required to undertake this process are specialist and time was required to facilitate this process. Along with colleagues working on another social care data resource (The Scottish Social Care Survey), a paper summarising the methods and the key components of the data resources was published.

Further challenges were faced when securing permission to use the data for research. As this project started early within the Public Benefit and Privacy Panel process, additional ethical approval was required for the study, a process which has now been simplified. However, although the NHS Research Ethics Committee process was relatively uncomplicated, the Public Benefit and Privacy Panel process was lengthy and the lack of opportunity to engage directly with the panel, to present the work and resolve queries in real-time resulted in further delays. It was not always clear what was expected or why decisions were made when applying the principles of working with data in a secure environment with the safeguards which have been built-in, particularly around disclosure control. For example, initially it was advised that Ethnicity (a demographic variable in the SCHC) could not be used. This decision was ultimately reversed when challenged, but raised concerns in terms of equalities and inclusion.

Finally, although all data for the project are held in the National Safe Haven, when the project lead changed universities within Scotland, a further long period of delay of significant months occurred with new contracts and updated data sharing agreements required and thus suspension of access to the data and research opportunities. This reflects that existing academic approaches to the ownership and oversight of research have not yet adapted to the landscape of data linkage, particularly when it uses national infrastructure for hosting and management.

Working with a novel data source which has been collected for a different purpose, national statistics generation, requires effort on the part of the research team to work with key stakeholders to understand the context of data collection and use. This work has been kindly supported by a multiprofessional stakeholder group to support the process and is a key recommendation to others embarking on data linkage work using a novel data source. Ultimately, the Understanding Pathways into Care Homes using Data (UnPiCD) study has yielded important results for practice, for policymakers and for data researchers. However, this case study highlights key challenges for those considering working with national novel data sources and the impact this can have on timely research delivery and progress.

Ongoing work

The ongoing work to improve and streamline information government across health and social care is outlined in the Data Strategy for Health and Social Care – 2024 update.

In addition, the Social Care Data & Intelligence Executive Group has set up a separate workstream on data sharing and information governance, to manage the overall delivery of data sharing across social care data producers and the wider social care sector. It is acknowledged that care home data (and the types of data in scope of this review) forms only one part of the much wider health and social care data landscape. This workstream will align to, and complement, the wider work being taken forward to implement the commitments outlined in the Health and Social Care Data Strategy, and the National Information Governance Plan.

The Data Sharing and Information Governance Workstream will co-ordinate the overall investigation and delivery of the data sharing potential across social care data producers and the wider social care sector. The workstream will consider a number of strands for investigating the potential for sharing of care home data and what changes may be required to enable this:

• Sharing data and what changes may be required to enable this between national / local bodies to facilitate the streamlining of the data collections and improving coherence of evidence base.
• Sharing back to care homes for the purposes of service delivery, planning and improvement
• Sharing with research bodies to facilitate research and innovation

A key aspect of this work will be engaging with data owners and stakeholders to understand and manage the barriers and risks associated with data sharing, along with identifying and advertising more widely the benefits of sharing, both in terms of reduced burden and improved understanding of, and outcomes for, those living and working in care homes.

Data sharing: Aims and recommendations

Aims: We will work to agree and publish a plan to share data securely and efficiently.

Recommendations: CHDR partners will establish a Data Sharing Working Group, containing both information governance and data professionals, to progress the safe sharing of care home data across organisations, where appropriate. The first task of the group will be to consider short, medium and long term opportunities.

Working Group will engage with stakeholders and manage barriers and risks associated with data sharing, and publish / advertise more widely the benefits of, data sharing. Produce documentation about how we improve sharing in social care, and how we do it safely and responsibly.

Ensure the Working Group links into the National Information Governance Plan for Health and Social Care.

Aims: Work to share added-value data and analysis back to data providers to maximise benefits of data provision.

Recommendations: Work with stakeholders to understand what data providers would find beneficial to have and take forward as a workstream of the Data Sharing Working Group.

Aims: Establish and publish a timely and efficient approvals process for accessing care home data for research.

Recommendations: Work with the Public Benefit and Privacy Panel for Health and Social Care (HSC- PBPP) and Research Data Scotland (RDS)/Administrative Data Research (ADR) to facilitate sharing of data for research with appropriate information governance in place.