CASE STUDY

Case study 5: Predicting worsening health status within populations

We design innovative and impactful data-driven solutions for clients, which we deliver in the form of user-friendly applications and tools. We have extensive experience in implementing and assessing the impact of these solutions in partnership with our clients.

While employed by Medscheme, Cristina and Stefan were part of a 3-person team who won the inaugural Starfield Award presented at the 2014 Johns Hopkins ACG System® conference. This was awarded for our work on developing a predictive model to identify individuals within populations with emerging health risks 5-10 years earlier than ordinarily would take place with traditional models that identify individuals at high risk. 

This Emerging Risk model was born from research regarding the transition in health status observed amongst individuals within a population over time. 

What differentiated this approach from other methods of identifying individuals for care management programmes, was the recognition that individuals at lower current morbidity levels, but who are at risk of deteriorating health status, present a significant financial risk to health systems. 

The Starfield Award, established in honour of Barbara Starfield, M.D. – a Johns Hopkins University Distinguished Professor and co-developer of the Johns Hopkins ACG System – honours her legacy by recognising the best use of the ACG System in three key areas of her work: addressing multi-morbidity in individual patients and populations; reducing disparities in the delivery of primary health care; and improving population health. 

We were integrally involved in all aspects of commercially available care management programmes aimed at individuals with high and emerging risk of worsening health. 

Our contribution included:

  • Conceptualisation of the population health strategy
  • Development of the predictive models
  • Design of the interventions aimed at mitigating the risk factors identified
  • Monitoring and evaluation of the care programmes implemented based on these population health models
  • Randomised controlled trials and retrospective case-matched cohort studies (as part of our aforementioned evaluations)