Why you should care about episode analytics

By Jeff Alford, SAS Insights Editor

Let’s say you have a friend or family member who is scheduled to have their trick knee replaced . . . finally. They put it off for years. Marriage, kids, career – effective pain meds – have all played their part in helping them delay the inevitable.

Now, they are finally ready to bite the proverbial bullet, and they're planning surgery, probably without giving a thought about the process that takes them from that initial consultation through their post-surgical rehab? Many people in that situation probably wouldn’t either. Except to worry about the out-of-pocket expenses for all the doctor visits, tests, procedures and medication they'll need.

If, however, you work for a health care payer or provider, you’re acutely aware of the often complex process the industry refers to as an episode of care.

Think of an episode of care as all of the parts (services and events) that have to take place to move someone from new patient to former patient. Now repeat this thousands, or hundreds of thousands of times, for hundreds of treatment regimens and you’ll begin to understand why health care organizations need to be able to understand all of those moving and flexible parts in order to create models that can provide excellent care for the patient and value for payers and providers alike.

Analytics plays the critical role by taking the unrelenting flow of patient and operational big data and helping create value-based models. In fact, a whole new area of health care analytics has been developed called episode analytics.

“Any health care organization, not just those participating in bundled payment or ACO arrangements, seeking to identify and understand cross-continuum care variation will find episode-based analysis an invaluable place to start on the journey of data-driven discovery and improvement.” 
Graham Hughes, MD, SAS Chief Medical Officer


Episodic care and population health data

When you’re building a model for a particular treatment, you look at all of the services that could be provided so that you can then begin to analyze the care that will need to be delivered. That why it’s also referred to as a bundle of care.

Sure, you have at the core a set of standards and it the past, and the care methods mostly focused on those core standards. In part because it was easier to identify the total cost of treatment given the technological limitations of the time.

Now, with recent analytic innovations, health care organizations can take those core bundles and expand them to include a much wider range of patient realities to create more encompassing and flexible treatment plans for patients, payers and providers.

A typical episode of care can go something like this, for friend's knee replacement, the initial claims data lets all the involved parties know treatment has begun. Based on data, episode types are defined by which services in a given time period are related to a clinical episode – for instance, friend's knee replacement – and what care is unrelated, such as another episode occurring in the same time period.


10 ways episode analytics can improve provider operations and patient care (click each image to learn more)

Payers and providers are working to establish set fees, or bundled payments for an episode of care, which create financial incentives to work as a team to share responsibility for a patient’s health.

Episode analytics provides a way to understand the money it takes to deliver services in an episode, which portions are complications, and how an episode is related to another condition. The more data, the better, to confidently understand and manage outcomes, as well as the financial risks and rewards.

Constructing an episode

Analytics allows providers and payers to confidently understand the services that go into medical outcomes and how they should be evaluated. By better understanding the degree of complexity of patient care in a specific context payers can better recognize what services are provided to individual patients – both on a case-by-case level and in aggregate. There’s no guessing.

This data provides insights that can help reduce unnecessary admissions or readmissions, decrease length of stays, improve cost-effective prescribing, reduce variation in care and treat patients holistically.

This process of episode construction and cost alloca­tion highlights the lifecycle of a patient’s treatment (as shown in their claims data) to provide an accurate measurement of total cost of care. And because value for the patient is created by providers’ combined efforts over the full cycle of care, the patient data is structured in a way that allows tracking of patient costs and outcomes across this entire cycle.

Episode analytics allows value-based care

Once this claims data is captured you can analyze an array of factors across a patient population to develop bundled payment models, degrees of variation in care and ways to increase provider efficiency.

You can use episode analytics to look at clinically relevant episodes of care from patient service and diagnosis information. Episodes such as a joint replacement, stroke or congestive heart failure are defined collections of services spanning the care continuum over a period of time. While services related to the patient’s condition are differentiated from those that aren’t, episode analytics can also identify which outcomes are potentially avoidable – like infection, readmission or an adverse medical event. Analytics enables you to include the clinical team in the improvement process because it provides evidence that can drive change.

With the right tools, health care organizations can share the rewards of achieving a higher standard of care with a patient-centric focus – all made possible using analytics.

Additional Resources

  • Want the latest ideas from SAS experts? Bookmark our health care blog: A Shot in the Arm.

Get More Insights

Want more Insights from SAS? Subscribe to our Insights newsletter. Or check back often to get more insights on the topics you care about, including analytics, big data, data management, marketing, and risk & fraud.