Skip to main content
Innovative uses of data and data science

Our Innovative Uses of Data (IUoD) team aims to improve our information analysis and reporting, by using novel data science techniques. This will enable new insights from data that work to improve health and social care. Products will be focused on the needs of patients, clinicians and organisations within the health and social care sector, to increase the likelihood of delivering real benefits that will improve patient outcomes.

How to make better use of data

This will be achieved in 3 ways:

  1. getting better value from data – providing analytical services for the NHS and researchers, reducing the costs of procuring these services from somewhere else
  2. doing more with data – increasing the types of analysis of data to develop additional insights
  3. increasing access to data – using the power of technology to give insights to users and stakeholders in a way that they can understand, without relying on advanced technical or analytical skills

Analytics is critical to improving health and care and there is a shortage of skilled analysts in the sector. Most large NHS organisations have a dedicated informatics unit and most have additional contracts with external analytics providers. Much of their time is spent producing baseline analytics and most are only scratching the surface of how they might use data to improve care for patients. For example, find out how Cambridge University Hospital Foundation Trust is using data to improve patient care.

Many smaller organisations would like to use more data, but do not have enough analytical capability or systems in place to keep patient level data secure. Our IUoD team are tackling these issues.

Baseline analytics

We undertake baseline analytics, such as:

  • benchmarking
  • dashboard preparation
  • prediction
  • data linkage

These can be used by many organisations, freeing them to focus on more bespoke work on improving care or efficiency.

Cutting edge analytics

In collaboration with our partners, we will demonstrate cutting edge analytics to tackle complex clinical and administrative challenges in health and care.

Increased access to data

We will increase access to data by providing technology to deliver intelligence to stakeholders who do not have analytical skills and systems. This might include graphical interfaces to data and the ability to create natural language queries.

When we look at the data cycle, these areas of work will provide a significant increase in analytical capability, new functionality at the interpretation stage and quicker, more accessible, and more efficient ways of feeding data back into the system. This will help drive change to health and care.

Patient safety

Health and care organisations, clinicians and patients will use our new tools to help them make decisions about the health and care of groups and individuals. We will make sure that data is accurate and interpreted correctly, to avoid the risk of suboptimal care or harm to patients. To do this, we will develop systems to make sure patient safety and information governance issues are addressed throughout the planning, development, deployment and maintenance of our new tools and analysis.

We will also make sure there is early and ongoing clinical input, so that interventions are clinically effective and relevant, making front line teams more likely to use them.

Professor Trish Greenhalgh at Oxford University, in collaboration with NHS Digital, has created a self-assessment version of her framework for Nonadoption, Abandonment and Challenges to the Scale-Up, Spread and Sustainability of Health and Care Technologies. We will use this framework to assess interventions, determine how likely they are to be adopted, and simplify them where possible, to improve uptake.

Project themes

We aim to look at the following themes:

  • demand management – making use of linked data to evaluate the level of demand on STPs and Vanguards
  • predictive models – using novel analytics, including analysis of NHS111 data, to help providers understand the likelihood of rehospitalisation, and to improve cost benchmarking
  • data driven subscriptions – analysing care pathways, and using technological advances to provide data to stakeholders
  • global analysis – developing tools that make international comparisons, enabling us to learn from worldwide best practice
  • machine learning techniques – exploring their potential applications
  • mortality dashboards – improving the accessibility of data
  • Trust picker/peer finder tool, which we are developing with NHS Improvement
  • Pareto analysis – a new pareto analysis dashboard for winter preparedness has been built and quality assured
  • risk stratification – addressing the legal, technical and data barriers of sharing data, to enable sharing of identifiable risk stratified patient information with GPs, and so improving patient care

Collaboration with central bodies

The Virtual Data Science Centre, hosted by our organisation, is a collaboration between the Department of Health and Social Care (DHSC), Care Quality Commission (CQC), NHS England, Public Health England and others. Members meet once a month to share learning and progress and plan collaborations.

Support for the life sciences

Our data has huge potential for research, and we want to support the life sciences to use this to the full. We are working with other organisations in the health and care system to make sure the value of our work, and the data we hold, can be maximised.

Increasing the range and availability of data

We are working collaboratively to provide life sciences and research access to a wider range and volume of data. This includes:

  • Direct care data – a Life Sciences Direction is being prepared, that would make available data including Spine data (1 billion messages a month), NHS e-Referral Service (e-RS) (550,000 referrals a month) and pathology messaging, via the Data Access Request Service (DARS). This provides a new dimension for researchers to study
  • GP data – we are supporting the cross-system 'GP data for research working group', set up by the Office for Strategic Coordination of Health Research (OSCHR) Informatics Board in response to feedback from researchers, such as from the recent UK Biobank International Phenomics Conference
  • Prescribing data – we are actively working on methods to make both hospital and community prescribing data available, as access to hospital data is currently limited
  • Public Health England (PHE) data – an implementation board has been formed to take forward findings of the recent McNeil review, on bringing PHE data together with our collections
  • National clinical audits and registries – we are working to make more of this data accessible for research
  • NHS data registers – we've launched an external service of reference data, in partnership with the Government Digital Service, which lets researchers access this data via an Application Programming Interface (API)
  • NIHR Health Data Finder (HDF) – we support this initiative to help researchers find out about data sets that can be used for research, and how to access them.

Streamlining legal and ethical approvals

We are working with the Medical Research Council Regulatory Centre and the Health Research Authority (HRA) to simplify the process for researchers to get legal and ethical approvals, by:

Supporting clinical trials

We already provide support for clinical trials, such as the ORION-4 clinical trial at Oxford University, and are consulting with partners including National Institute for Health Research (NIHR), HRA, Cancer Research UK, and CPRD, on providing more services to help clinical trials find out about feasibility and recruitment (identification and optional invitation) of participants. We've had positive feedback welcoming these proposals, and we're looking into whether it would be useful to flag participation in a clinical trial on a patient's health record.

Last edited: 9 October 2019 11:50 am