GNS Healthcare, a healthcare analytics company and Aetna, an American managed health care organization, are collaborating to make use of GNS’ supercomputer “REFS” (Reverse Engineering and Forward Simulation). By using predictive analytics with Aetna claims and other health information, the REFS platform will create data models to help the early identification of metabolic syndrome, which increases the risk of heart disease, stroke and diabetes.
To see how REFS works with data and creating models, watch the video below:
Read more about the collaboration from the GNS Healthcare press release below:
Using claims and other health information* from Aetna, GNS will use a platform called REFS™ (Reverse Engineering, Forward Simulation) to create data-driven models. These models will define a person’s risk for developing metabolic syndrome. A person has the disorder if they have three or more of the following five conditions:
- large waist size
- high blood pressure
- high triglycerides
- low HDL (‘good’) cholesterol
- high blood sugar
A member who has one or two of these conditions is at risk for the syndrome. Based on the at-risk member’s specific health information, the model will predict which new condition the member will likely develop next and how quickly. For example, the model may show that someone with high triglycerides and low HDL will likely experience high blood pressure within 12 months without intervention. To improve or eliminate the risk factors, the model then matches each member with specific interventions that are most effective for that condition.
“Reducing or eliminating the impact of metabolic syndrome can improve the health of millions of people and reduce health care costs. Aetna offers many programs to help members understand their risks and take steps to improve their health. Using data, we will now know very quickly which of these strategies works best for specific members,” said Michael Palmer, the head of the Aetna Innovation Labs. “We will also know where we can make improvements or create new programs to help our members.”
The timelines and personalized interventions are possible because REFS discovers new knowledge directly from data. Most analytic models identify only data associations and say nothing about cause and effect. REFS is different. The model not only identifies that a data association exists, but also highlights the underlying causal relationships.
“Multiple Big Data methods can identify at-risk people. We are different because we not only identify people at risk, but we also can identify the most effective intervention for each person,” said Colin Hill, CEO and Co-founder of GNS Healthcare.
“GNS has a long history of using REFS to understand the pathways for different conditions and treatments. Working with Aetna to learn about the progression toward metabolic syndrome using massive real world data is a natural extension of our capabilities,” said Carol McCall, Chief Strategy Officer for GNS. “We expect our models, built from the millions of distinct data elements Aetna provides, to reveal causal relationships among hundreds of variables and how they evolve over time that lead to actionable insights into precursors of metabolic disease.”
Based on the success of their work in metabolic syndrome, the parties expect to address other major disease areas.
*Claims and data used in these models will be secure and comply with applicable privacy laws.