Data Mining and the Implementation of a Prospective Payment System for Inpatient Rehabilitation

Published in: Health Services and Outcomes Research Methodology, v. 3, no. 3-4, Dec. 2002, p. 247-266

Posted on RAND.org on December 31, 2001

by Daniel A. Relles, Greg Ridgeway, Grace M. Carter

Read More

Access further information on this document at www.springerlink.com

This article was published outside of RAND. The full text of the article can be found at the link above.

This paper describes the development of a new Medicare Prospective Payment System (PPS) for inpatient rehabilitation care. Congress mandated such a system in the Balanced Budget Act of 1997. To help implement this system, the authors assembled four years of Medicare hospitalization data, linked it to rehabilitation hospitals' information about impairment and the functional status of patients, and developed case mix groups using the CART algorithm, a common method for determining groups in health services. While CART readily produces simple and effective rules for prediction, it adheres to a restrictive functional form and its fitting algorithm does not necessarily produce a global optimum. The authors wanted to know how these limitations affect our results. So, they compared CART's performance with methods receiving attention in the data mining community and in the statistics literature. The authors estimated that the CART models explained about 90 percent of the potentially explainable variance in individual cost and they predicted annual hospital costs that were essentially identical to other methods' predictions.

This report is part of the RAND Corporation External publication series. Many RAND studies are published in peer-reviewed scholarly journals, as chapters in commercial books, or as documents published by other organizations.

Our mission to help improve policy and decisionmaking through research and analysis is enabled through our core values of quality and objectivity and our unwavering commitment to the highest level of integrity and ethical behavior. To help ensure our research and analysis are rigorous, objective, and nonpartisan, we subject our research publications to a robust and exacting quality-assurance process; avoid both the appearance and reality of financial and other conflicts of interest through staff training, project screening, and a policy of mandatory disclosure; and pursue transparency in our research engagements through our commitment to the open publication of our research findings and recommendations, disclosure of the source of funding of published research, and policies to ensure intellectual independence. For more information, visit www.rand.org/about/principles.

The RAND Corporation is a nonprofit institution that helps improve policy and decisionmaking through research and analysis. RAND's publications do not necessarily reflect the opinions of its research clients and sponsors.