Follow Us

Our Work

Over the past few months, we've been rapidly repurposing existing programmes of work in order to address the immediate needs of the COVID-19 pandemic response. Here's an overview of our current progress so far. For more information on how we can help, get in touch via the 'Contact Us' page on our website!


To support the focus on population health management in the Long Term Plan (LTP), the Data, Analysis and Intelligence Service (DAIS) in NHS England and NHS Improvement (NHSE and NHSI), Public Health England (PHE), Outcomes Based Healthcare and Arden & GEM CSU have developed a national data-driven approach to population segmentation, and a national framework for segmentation that supports delivery of this approach by local systems using local datasets.

A national person-centred segmentation dataset for population segmentation has been developed within NHSEI’s data environment using data sources from the National Commissioning Data Repository (NCDR). This is an adaptation of the internationally recognised ‘Bridges to Health’ (B2H) segmentation model.


      The National Segmentation Dataset
  • Incorporates data from every person registered to a GP in England (apart from people who have opted out of data sharing), over 10 years of Secondary Uses Service (SUS) data (which includes A&E data, Admissions data and Outpatient data), several years of maternity data, Mental Health, Community Services, SLAM, IAPT, Assuring Transformation data and Hospital Frailty Risk Score data.
  • The Segmentation Dataset includes the segments and sub-segments (conditions) for the 62 million people in the GP registration list, for each month over the last four years (April 2016 – January 2021).
  • Further information about the Segmentation Dataset can be found in the Population and Person Insight (PaPI) Workspace on FutureNHS.

COVID-19 Vulnerable Population Analysis

Using the existing NCDR segmentation model, OBH have identified and quantified those who are ‘clinically vulnerable’ to risk of severe illness from COVID-19, at Primary Care Network/CCG/STP/National levels. This information has been used to inform the national response to COVID19.

One example of this being the use of the Bridges to Health Segmentation and ethnicity model as part of a significant piece of research exploring the risk of Diabetes and ethnicity on mortality from COVID-19.

How does this fit alongside the current national COVID-19 response?

The CMO (and NHS Digital) have identified two key groups most ‘clinically vulnerable’ to severe complications from COVID-19:

• Those ‘At High Risk’ (c. 2.5 million) on the Shielded Patient List (SPL)
• Those normally ‘At Risk’ of flu (c. 19 million)

The wider list of people normally at risk of flu contains groups of people not known to be at significantly increased risk of COVID-19 complications (eg. pre-school children), and excludes people understood to be at risk of complications (eg. people who have had a stroke, or history of cancer, or hypertension). Conditions in the Red Box are included in the CMO’s wider ‘At Risk’ list of conditions, and the conditions in the Yellow Box are additional conditions where the evidence suggests an increased risk of severe illness.

Draft Report – Estimated ‘vulnerable populations ‘at risk’ of severe illness from COVID-19 in the National Commissioning Data Repository (NCDR) England

This report provides estimates of the ‘clinically vulnerable’ populations ‘at risk’ of severe illness from COVID-19 using the National Commissioning Data Repository (NCDR) England – as at 30th June 2021, at a national level and for individual STPs and CCGs. The output of this work is being incorporated into NHS Viewpoint, and visualised area by area, down to Primary Care Network level.

Key Findings:

  • 16.3 million (c.26.7% of England’s population) are ‘at risk’ of severe COVID-19 illness
  • 67% (c.10.9 million) of the ‘at risk’ cohort have a known COVID-19 risk factor condition, rather than just being ‘at risk’ due to age alone
  • Over half (c.5.3 million) of people aged 65 and over have no currently known risk for COVID-19 complications
  • Wide variation at CCG level (between 19-37%) in size of the ‘core older and clinically vulnerable’ cohort