Mapping the flow of healthcare data shows complex landscape

A comprehensive mapping and evaluation of electronic data flows in the NHS finds complex landscape and informs policy objectives.

Healthcare data is routinely recorded as part of the medical record. In the NHS, most interactions are recorded in electronic health records (EHRs) and can be turned into population datasets. This ‘big data’ has potential to help life sciences and artificial intelligence, improve clinical decision making, and support population health and personalised medicine. 

For the first time, a comprehensive mapping of all electronic data flows, infrastructure and assets has been undertaken, and shows a complex landscape with tens of thousands of data transactions. Mapping data flows has identified how data is shared between healthcare providers and users, and uncovers shortfalls in transparency, best practices for safe data access, and lack of value return to the NHS.  

The research, published in The Lancet Digital Health and led by researchers from the Institute of Global Heath Innovation at Imperial College London provides recommendations to support data infrastructure transformation. 

“This is the first comprehensive mapping of all electronic data flows, infrastructure, and real-world data assets across any national healthcare system.”  Dr Joe Zhang Wellcome-funded clinical research fellow at the Institute of Global Health Innovation

Lead author of the study, Dr Joe Zhang, Wellcome-funded clinical research fellow at the Institute of Global Health Innovation said: “This is the first comprehensive mapping of all electronic data flows, infrastructure, and real-world data assets across any national healthcare system.” 

“It is hard to move beyond broad concepts when talking about data strategy, because we’ve never fully understood what is out there. This new information gives an objective understanding of what gaps there are in the landscape – including in infrastructure, types of data available, and risks to patient trust. This is key information for guiding investment into new data infrastructure and for designing specific policy objectives.” 

The work in mapping data flows and assets will help inform the upcoming Sudlow Review: ‘Unifying Health Data in the UK’.

Transparency and Trust 

Accompanying the research is a website Datainsights.uk that showcases NHS data assets and promotes transparency in how data is created and used. 

The Data Insights website aims to simplify understanding of how patient data in NHS England are extracted, how data flows into research assets, and are used by consumers at the end of ‘data flow’ chains. 

Of the NHS Data consumers, the largest group of users is universities, such as Imperial, which use the information for research purposes, in an approved, regulated and ethical manner for academic research.  

One of the important findings of the mapping exercise was that reporting of data uses was incomplete or lacking in specificity, which risks going against data protection legislation. Further, it puts much of the responsibility on patients to investigate how their data is being used, when it is not clear up-front. 

The data in EHRs can be a valuable resource for improving clinical outcomes and reducing health inequalities. This research questions the utility of existing and planned data infrastructure which duplicate existing data assets and which may not add to existing capabilities. This is particularly important as the NHS plans to spend £480million on a federated data platform. 

Within the current period of data transformation in the NHS, there is opportunity to improve the value of the data we hold. “Investment should be targeted at unlocking unavailable data that is behind silos, and enabling AI deployment and effective life sciences collaboration, rather than re-shuffling existing data into new environments,” said Dr Zhang. 

Jessica Morley, Oxford Internet Institute, and study co-author said: “Maintaining the public license to access NHS data for purposes beyond direct care is extremely important. This requires concrete action and meaningful transparency. Communications and privacy statements are necessary, but not sufficient. The NHS must be able to tell all members of the public where their data is, how access to it is being controlled, what purposes it is being used for, and how it is being protected. The strategy set out in Data Saves Lives, published last year, suggests that this is the direction of travel for current policy. We hope that our research will be helpful to policymakers as they move towards a more secure-by-design and transparent-by-default system.”  

Recommendations for data transformation relate to public transparency, improvement of existing data infrastructure, and investment to increase value return to patients and providers.  

Professor the Lord Ara Darzi, Co-Director of the Institute of Global Health Innovation, said: “Mapping data flow across the breadth of the EHR landscape in the NHS in this manner can offer several benefits in our digital health maturation that allows for improved data security, operational efficiency and decision support that can hopefully be translated into better and safer outcomes for patients.” 

Explore Data Insights 

'Mapping and evaluating national data flows: transparency, privacy, and guiding infrastructural transformation', Joe Zhang et al, The Lancet Digital Health