Managing Whole Population Health Needs

Population risk stratification, predictive modelling and resource allocation with Adjusted Clinical Groups®

Our consultants have extensive experience in over 30 NHS localities with patient-specific whole-population case-mix systems combining primary and secondary care clinical, utilisation and cost data. They have led or played a key role in the deployment of the Johns Hopkins University ACG tool in 14 health communities across NHS England.

Now deployed across 5 continents, the ACG® system is a 3rd generation, evidence-based predictive modeling and risk adjustment tool whose applications we are tailoring to the needs of NHS commissioning consortia and whole-system management. We are one of only two formal partners of Johns Hopkins. We are unique in that we work with health organisations such as CCGs and CSUs to develop locally bespoke solutions on local business intelligence platforms. We also train clinical staff as users of these systems employing expert case managers to lead this training.

The policy framework for the NHS  empowers Clinical Commissioning Groups with the commissioning of up to 70% of healthcare. Accordingly it includes the following prescriptions:

  • CCGs will be held to account for their expenditure and outcomes
  • They will be incentivised to improve outcomes
  • Commissioners will accordingly need to set equitable hard budgets at practice level
  • Personalised budgeting will be taken forward for some patients with long-term conditions


All this has the following implications for commissioners and/or providers

  • They need to assess risk at an individual patient-level which can aggregate to practice, locality and CCG level
  • The risk assessment should be clinically-based informatics which connects need to current and required costs
  • They need to identify all patients who are high risk and high cost currently and, in the future. to inform systematically the commissioning and impact assessment of best-practice care and case management
  • They need to understand how the consortium and each constituent practice is using resources relative to its relative morbidity profile
  • Although the consortia are allocated resources using a formula generated by the Commissioning Board, resource allocation and management below consortia level needs to be case-mix adjusted. This is to reflect the fact that morbidity and utlisation of small populations such as practices can vary significantly even when the age, gender and deprivation are the same. This confirms or otherwise the common response of GP practices that “we consume more resource because our patients are sicker than the average or the practice down the road”.
  • This data should also be mapable to post-code to allow small-area analysis such as lower super output areas, the “currency” in use in public health.


ACGs ® has delivered significant quality and cost improvements in both the NHS in England and internationally particularly in other European Countries (such as Sweden)  as well as not-for-profit Health Maintenance Organisations such as Kaiser Permanente in the US. This includes:

  • Evidence-based identification of specific patients groups with specific needs profiles and expected high resource needs and utilisation for more pro-active management
  • Quality and productivity gains in the short and medium term (up to 2 years)
  • Clinically-informed risk management
  • Risk adjustment of resource allocation and management
  • Morbidity adjusted benchmarking of practices resource utilisation across a balanced score card which contextualises use of referrals and rate of hospitalisation against other expenditure such as pharmacy and GP utilisation
  • Developing meaningful dialogue with Practices, CCGs and Area Teams on resource needs and deployment


Our team includes expertise on:

  • Clinical applications of various tools including successful case management programmes
  • Case-mix and predictive modelling expertise including drawing on our senior colleagues from Johns Hopkins University ACG international development and deployment team
  • Relevant informatics