analyticsEngine_header

Analytics Engine

We’re enhancing data for amazing insight at the speed of light.

The Koan Analytics Engine is the powerhouse that undergirds all our solutions. We’ve engineered it with the latest cloud technology in the Amazon Web Services environment for the utmost in scalability, security, reliability and speed.

Getting your data in shape to deliver

The Koan Analytics Engine combines six technologies to move your data from chaos to coherence:

Data transformation

Payer claims files, Medicare CCLF files, bundled payment claims files, HL7 feeds, EMR data feeds, Excel files—the data you’re wrestling with every day comes in multiple, incompatible formats. Koan Health turns chaotic input into intelligible data fast. We remove or correct duplicate records, check for valid ranges and inspect data for consistency. In half the time it would take an internal team, you’ll have useful outputs you can feed into your database or business intelligence tools.

Data enhancement

Once we’ve transformed, cleaned and imported your data into our staging databases, we add key information and markers to boost its usability and power. We use proprietary algorithms and analytics to identify specific types of risk markers based on unique patient circumstances and health system utilization. We also harness publicly available databases and leading-edge technology from other innovators, including the Johns Hopkins ACG System.

Our data enhancement process is as flexible as your needs. To enhance the power of paid claims data, for example, we can add multiple elements, including:

Episode grouping

Once the data is transformed and enhanced, we can group it into the episodes you’re actively managing. Our proprietary tool lets us build custom episodes to your exact specifications—with maximum flexibility for setting index or trigger events, inclusion and exclusion criteria and duration. Advanced episode modeling lets you see the data in multiple views, make changes and test the results. Once episode specifications are verified, our proprietary software loads them into the production system for seamless integration into the analytics workflow.

Provider attribution

If you’re operating in an accountable care environment, implementing gainsharing or simply creating provider scorecards, you have to attribute provider activity accurately—to a population, episode or network. The Centers for Medicare and Medicaid Services has established one attribution methodology for the Medicare Shared Savings Program, but you may need to go deeper—attributing specialists or other members of the care team to clinical activity. Our proprietary algorithms accelerate the development of attribution logic that links providers and their activity. You’ll easily identify cost-of-care outliers, spot areas of potential network leakage and have the information you need to improve performance and keep care in your preferred network.

Analytics Engine

Data science

We developed our own powerful approach to the art of data science—predictive modeling, statistical comparisons, data mining and descriptive analytics—to help you extract maximum insight from your data. In all of our solutions, our expertise as data scientists enables you to use sophisticated combinations of characteristics and risk factors to track patients, isolate the impact of specific interventions and improve results. One simple but remarkably significant advantage is our success in eliminating false positives—the noise that can mar your analytics and trigger unproductive “alert fatigue.”

Data visualization

At the end of the day, our efforts to maximize and analyze your data mean nothing if the results aren’t clear, beautiful and easy to grasp. We use some of today’s most sophisticated data visualization tools—RStudio, Tableau and more—to create data displays your team can understand, share and act on.