New iHT2 report shows how payors, providers, and ACOs, can leverage episode analytics against cross continuum episodes of care to decrease avoidable complications and cut waste in healthcare.
The wide variations in healthcare cost and quality, a perennial issue for payers, have become important to providers as well with the advent of accountable care organizations (ACOs) and the increasing number of healthcare systems taking financial risk. These organizations need sophisticated software tools to analyze performance at the level of an episode of care that spans the care continuum, including episodes involving procedures, medical admissions, and chronic disease care. By using analytics to identify and measure these variations, organizations can work with clinicians to reduce potentially avoidable care variations that negatively impact care outcomes and lead to higher costs.
Previously, IT applications in this area include episode groupers that health plans have long used to profile the resource utilization of individual providers. A newer type of software can be used to more accurately define clinically meaningful episodes of care, measure variations in care, identify potentially avoidable complications, and track actual versus expected performance on a risk-adjusted basis. These new applications can also help provider organizations manage financial risk in payment bundling and other value-based reimbursement models.
The first step in analyzing health care variation is to select the right unit for analysis. Health care for the entire patient population is too broad a unit to yield useful insights; on the other hand, individual encounters, procedures, or diagnosis related groups (DRGs) provide too narrow a focus because they exclude everything that happened before and after the encounter, procedure, or hospitalization.
A better unit of analysis than either of these extremes is an episode of care that crosses care settings and includes a pre/post period of time relevant to the specific condition or medical event, such as a hospital stay and post-acute care for 30 days after discharge, or the care of a patient’s chronic condition for a year. While the definition of an episode can vary widely, it should include enough services to allow the variations in treatment choices and intensity to be measured and analyzed. The ability to perform this kind of analysis is essential to any organization that seeks to make the transition from fee-for-service to value-based payment models. An approach utilizing episode analytics represents a major advance over the traditional claims groupers. The new analytics enable providers to identify variations and opportunities for improvement, increased flexibility in defining clinical episodes, and allows organizations to assess their financial risk in value-based reimbursement agreements.
This research report is divided into two parts. The first part provides some background and a comparison of the types of episode analytics. Part two explores the real-world experiences of payers and providers in using episode analytics for payment bundling and other purposes. Finally, we offer some recommendations on how to use episode analytics to reduce variations and manage contracts that involve financial risk.
The report was put together by:
- Daniel Barchi, Chief Information Officer, Yale-New Haven Health System
- Francois de Brantes, Executive Director, HCI3
- Graham Hughes, MD, CMO, SAS
- Tricia Nguyen, MD, MBA, President, Texas Health Population Health, Education, & Innovation Center
- Gregory Poulsen, Senior Vice President & Chief Strategy Officer, Intermountain Healthcare
- David Redfearn, PhD, Advanced Analytics Senior Consultant, WellPoint, Inc.
The research report is available for download at http://ihealthtran.hs-sites.
iHT2 will also be hosting a complimentary webinar with the contributors on Thursday, February 13th at 1:00pm eastern. The contributors will be sharing their thoughts with moderator Mark Hagland, Editor-in-Chief, Healthcare Informatics. Register for the webinar at https://www2.gotomeeting.
Featured image credit: IBM Events via cc