Cover: How Costly Is Hospital Quality?

How Costly Is Hospital Quality?

A Revealed-Preference Approach

Published In: NBER Working Papers / (Cambridge, MA : National Bureau of Economic Research, Jan. 2008), p. 1-37, 1-24

by John A. Romley, Dana P. Goldman

Read More

Access further information on this document at papers.nber.org

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

Abstract

One of the most important and vexing issues in health care concerns the cost to improve quality. Unfortunately, quality is difficult to measure and potentially confounded with productivity. Rather than relying on clinical or process measures, the authors infer quality at hospitals in greater Los Angeles from the revealed preference of pneumonia patients. The authors then decompose the joint contribution of quality and unobserved productivity to hospital costs, relying on heterogeneous tastes among patients for plausibly exogenous quality variation. The authors find that more productive hospitals provide higher quality, demonstrating that the cost of quality improvement is substantially understated by methods that do not take into account productivity differences. After accounting for these differences, we find that a quality improvement from the 25th percentile to the 75th percentile would increase costs at the average hospital by nearly fifty percent. Improvements in traditional metrics of hospital quality such as risk-adjusted mortality are more modest, indicating that other factors such as amenities are an important driver of both hospital costs and patient choices.

Research conducted by

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.

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.