Population health analytics now embedded across almost 9 million NHS patients, supporting earlier intervention and reduced emergency care

Population health analytics are now being used across the NHS to identify and segment care needs for millions of NHS patients, supporting earlier intervention and a sustained left shift of care from hospital to community settings.

Nearly 9 million of those patients are now actively identified and stratified using the Johns Hopkins ACG® System methodology through platforms delivered by Graphnet Health - representing almost one in six people in England.

At this scale, population health analytics has moved beyond planning and reporting into routine operational use. Clinically validated patient identification and segmentation are now embedded into day-to-day operations to inform decision-making across Integrated Care Systems, NHS trusts and primary care organisations.

By identifying rising risk earlier, NHS teams are able to prioritise limited workforce capacity more effectively and redesign pathways around proactive, neighbourhood-based care.

In several regions, this approach is already demonstrating measurable impact.

In East Kent, a 12-month proactive care programme focused on people living with frailty and complex needs saw emergency department attendances fall by around 69 per cent and emergency hospital admissions reduce by approximately 70 per cent for the cohort involved. Patients reported greater confidence in managing their health at home and feeling better supported by local services.

Dr Sarah Phillips, Chief Medical Officer at Kent Community Health, Senior Lead for the programme and Chair of the Programme Board, commented:

We are delighted with these powerful results. The solution is not only making patients feel more in control of their wellbeing, but also proactively keeping people out of hospital and in the community - actively supporting the left shift in healthcare.

At the system level, Frimley Integrated Care System has applied the Johns Hopkins Patient Need Groups segmentation model to urgent and emergency care data to better understand patterns of demand. 

Analysis showed that approximately 15 per cent of total emergency department attendances - around 102 attendances per day - were driven by minor illness cases within defined cohorts, highlighting opportunities for earlier intervention and reduced reliance on acute care.

Dr Priya Kumar, a Primary Care lead in this work, said:

Urgent care is a complex system to navigate. By embedding the ACG System segmentation tool into our systems, we are further supporting our clinicians with a greater understanding of the patient in front of them. Combining this with their clinical decision-making skills means our patients are being seen by the right healthcare professional first time, which creates time and space to care and improves the overall outcomes and experiences for our patients.

Population health analytics is also supporting operational change in primary care. Practices including Brookside Surgery and Thatcham Medical Practice have used ACG-based segmentation to enhance triage processes, prioritise urgent cases more safely and route patients to the most appropriate clinician first time. This has reduced administrative burden, improved equity of access and ensured limited clinical capacity is focused on patients with the greatest need.

In several regions, including Cheshire and Merseyside, this intelligence is delivered through CIPHA - Combined Intelligence for Population Health Action - an integrated population health platform bringing together data from primary, secondary, community and social care. The platform supports hospital-to-community models of care at neighbourhood, place and system level.

Globally, the Johns Hopkins ACG System methodology supports population health decision-making for approximately 200 million patients worldwide across 20 countries, with England now representing one of the largest national deployments.

As NHS systems continue to face rising demand, workforce pressure and widening health inequalities, large-scale patient identification and segmentation is becoming increasingly important to delivering sustainable, community-based care.