FOR REVIEW
Tom McCann, director of Sopra Steria, outlines the findings of research on how to ensure public sector data meets peoples’ needs
While the value of data is now widely understood throughout the public sector, to really make it work for the public good requires a precision in how it is structured and used, ensuring that human needs are placed at the centre of activities.
This can be a significant challenge, complicated by the rapid pace of change in the demand for and delivery of services. But if data is structured from the start in a way that it can shared and used by organisations to provide people with better targeted and personalised support, it enables a proactive approach to identifying and supporting those most in need.
Enabling a human approach to data has been the subject of a research project carried out by UKAuthority in collaboration with Sopra Steria, aimed at identifying which measures can help the most.
Discussions with a number of public sector officials highlighted several problems, and it came as no surprise that data quality is a major issue, often related to organisational shortcomings. Senior officials can be prone to assume that the quality of their organisation’s data is better than is actually the case, and they often see data as an IT issue, rather than one for the whole organisation that is core to everything they do.
Legacy lock-in and APIs
Another factor is that a lot of data is still locked in legacy systems, with some suppliers not making it easy to build in the application programme interfaces (APIs) for it to be shared and made more accessible. And its quality often reflects the original use case for which it was collected, which means it may be structured in a way that is not suitable for a different context decades later.
There is also a complication in the lack of data on people who are hard to reach, either because of personal backgrounds or a desire to remain ‘off the grid’ – when these are often some of those most in need of public services. Then there are the longstanding issues around the details of information governance and legal framework for data protection. The UK General Data Protection Regulation enshrined in the Data Protection Act 2018 has often been perceived as a blocker to data sharing rather than an enabling framework, and while there is a gradual change in the perception to seeing it as an enabler, there is still a mixed understanding among organisations with differing levels of data maturity.
Overcoming all these issues to deliver a human-focused view of people will not be easy, but the discussions produced a series of ideas to support progress.
The first is to build on the experience of the Covid-19 pandemic, when there was a surge in data sharing for health and social care encouraged by the Government’s COPI notices that relaxed the relevant rules. While these have now been withdrawn, there is now a strong evidence base of the benefits of such sharing This, coupled with a data provision notice for client-level adult social care and guidance on information sharing from NHS England’s Transformation Directorate has built confidence that it is possible to find the right balance for appropriate data sharing across sectors to work for the public good while respecting privacy.
Risk stratification
It was suggested in the discussions that this approach could usefully come with an element of risk stratification. While the term currently has a relevance mainly to healthcare – referring to categorising patients into low, medium and high risk groups – it could be applied to people in a wider context to establish where there are stronger grounds for sharing their personal data.
There is also a clear need for data standards – related to the data quality issue above – and interoperability. This includes creating the conditions for routine use of identifiers for linking data from different sources. Progress on the latter has been slower than many would like, but there is scope for wider use of the NHS number, national insurance number, and unique property reference numbers (UPRNs), the identifiers for each addressable location in Great Britain.
Routine provision of APIs in digital systems was highlighted as a key to integration and sharing of information. However, this has been something that some suppliers have been reluctant to provide, which has raised the question of whether there should be a central set of procurement clauses used across the health and public sector. There is a trend in this direction but progress has been slow.
Other factors include: the need for a broad and clear picture of where relevant data lies within the public sector; taking advantage of AI to make more use of qualitative data; building data literacy among public servants; and encouraging innovation in the use of data, possibly through ‘sandbox’ environments.
Finding optimism
It amounts to a complex and challenging outlook, made more so by the pace of change in technology and public services, but there are causes for optimism.
These include a shift in perceptions around information governance, with more public officials seeing the regulations as an enabler rather than a blocker to sharing. There are also efforts to deal with specific issues, such as flags indicating vulnerability and protecting children from harm; and examples of good practice that could be scaled up for wider use.
Awareness is building and supported by a growing consensus that data is there for the benefit of individual citizens, and that it is legitimate to find ways to make it work accordingly to meet real peoples’ needs.
These issues are covered in greater depth in the briefing paper which can be downloaded here or below. <<TO ADD>>