two work colleagues

Using health analytics to identify patients at risk of COVID-19

Designing a new segmentation model to identify patients at risk of COVID-19 in Berkshire West ICP to enable proactive contact to deliver appropriate support.

The challenge

At SCW, our Population Health Management (PHM) team has been helping CCGs and surgeries who want to identify their patients at risk of COVID-19, in order to make proactive contact with them and to support them appropriately.

The team worked in a collaborative way using clinical inputs and a variety of tools, including the Johns Hopkins Adjusted Clinical Groups (ACG®) system (through SCWs Insights Population Analytics (IPA) solution), to break down a large number of At-Risk people into more manageable segments,  while looking through the ‘lens’ of COVID-19.

The team designed a segmentation model with clinical input from Dr Steven Laitner of the National Association of Primary Care (NAPC) and Berkshire West ICP, to ensure the categorisation of the sub-segments contained a manageable number of individuals and also made sense clinically. 

Using population health analytics to drill down into more discrete groups

As Berkshire West ICP was well advanced in using a PHM approach to design interventions, the team there were in a good position to continue to use segmentation and stratification to target cohorts of their population more effectively.  Using similar techniques to the Wave 1 Accelerator programme, the team applied these to NHS England’s list of those people eligible for flu vaccines.

With certain people more at risk from COVID-19, and with the current ‘shielded’ list covering about 2% of the population - the groups are too large in number to coordinate a coherent intervention at GP practice level.

The team was able to break down the ‘at increased risk’  group into more manageable groups, based on a number of factors and successfully - using a categorisation of a number of vulnerabilities -  segmented and identified patients within a specific locality, in order to better target them for interventions and to enable enhanced patient care. 

Dr Steven Laitner, a practising GP and public health consultant said “I was delighted to work with the team at SCW. Population Health Management requires a deep understanding of the needs and preferences of patients and their carers; clinicians and other professionals; cohorts and communities. The team at SCW have that! I am proud of the holistic model we created, which whilst looking at the physical risk of COVID-19, also ensured awareness of the social, environmental, functional and mental health vulnerabilities of people. Whole people, complex needs, holistic solutions.”

Chris Morris, Associate Solution Lead, Population Health Management and Business Intelligence at SCW says: “We worked with respected academic partners and networks on population health, using a variety of tools including the Johns Hopkins Adjusted Clinical Groups (ACG) System, giving customers the opportunity to engage with integrated data in a meaningful and powerful way. Using the ACG System as an integral part of our IPA solution means we have access to a very detailed dataset to allow very specific fine-tuning, ensuring the outputs are tailored to the task in hand quickly and efficiently. This prototype was developed within five working days and the approach has now been piloted, so the principals that were used would be useful for PCNs and other healthcare groups.” 

Segmenting and stratifying the population in a measured, structured and timely way

The team used the following approach to identify patients:  

  • Worked with NHS England nationally mandated data –the Shielded Patients list and those that were defined as At Risk
  • Used the ACG System’s specification to create sub-segments of the population
  • Examined other ways to segment and decided on cross-referencing against functional vulnerabilities
  • Moving from locality level to drill down and break lists down to practice level and individual level


This created a list of older adults with vulnerabilities, who could then be targeted for proactive interventions from the GP surgery.

Segmentation was done using anonymised data. For example, the numbers of people in the current 'increased risk’ group, split by life course and vulnerabilities.   

(Please note that the current configuration places about 10% of the population into this group but we will make changes that we expect to increase this number to around 15-20%.) 

The Population Health Management (PHM) team at SCW continues to provide support and direction to Practices, Primary Care Networks and Federations in working to successfully implement new care models, develop connections and work collaboratively within Integrated Care Systems to realise the best possible outcomes for the population.

For more information on our support packages, or if you have any questions, please email This email address is being protected from spambots. You need JavaScript enabled to view it..


More case studies