| |||||||||||||||||||||||||||||||||||
|
However, research also shows that there are limitations to the usefulness of cost sharing for improving efficiency. Most important, patients with high levels of cost sharing appear equally likely to cut back on essential health care services as on services of low or no value. Such findings suggest that consumer cost sharing ought to be selective, or value-based: low cost sharing for high-value services and high cost sharing for low-value services. Some progress has been made in designing value-based insurance for pharmaceuticals, but the application of this concept to other aspects of medical care has lagged, in part because of a lack of comparative-effectiveness information.
Many efforts to improve health care markets involve providing consumers with technical decision support — helping them to determine when to seek care, how to manage chronic conditions, and how to participate in decisions regarding treatment choices. Payers have invested billions of dollars in efforts to engage patients in self-care and making decisions about treatment. Evidence of the efficacy of such efforts, however, is in short supply. Disappointing results from the recent Medicare experiment with disease-management services provided by third parties have called into question the approaches most commonly used to engage consumers.3
Substantial effort has also gone into arming consumers with information to help them select physicians and hospitals that are high-quality, low-cost, or both. The combined goals of this effort are improving average quality or efficiency by directing patients to the best providers and creating competition among providers by inspiring them to improve care along the dimensions that are assessed and reported. Provider report cards have been available and evolving since the 1980s, but research has cast doubt on their value, since patients frequently do not understand, trust, or rely on such reviews in selecting providers.
On the supply side, by contrast, there is some evidence of a response by providers to public reporting about their performance. For example, in a controlled experiment in Wisconsin, hospitals whose quality data were reported publicly engaged more actively in quality-improvement activities than did hospitals receiving the same information privately.4 Public reporting of providers' performance also appears to yield less desirable supply responses, including avoidance of patients perceived to be high-risk. Such adverse effects might be diminished through measure selection, risk adjustment, and denominator-exclusion policies that reduce incentives for avoiding nonadherent or high-risk patients.
Payers, for their part, increasingly recognize that payment incentives can affect supply in the realm of medical services in the same way that informed consumers do in other markets. Conceptually, reimbursement reforms involve changing the level of payment, the risk that providers face in relation to utilization or costs, or both. Payers might use the level of payment to encourage more or less utilization overall or to encourage providers to deliver specific services. Causing providers to hold financial risk related to the cost of services is known as risk sharing. In practice, risk sharing varies along a spectrum from withholding of payments or paying of bonuses related to utilization or spending to capitation (i.e., full risk).
Increases in relative prices (for one group of patients or type of service versus another) have been associated with increased supply. Yet in some cases, providers may increase the volume of services in response to an overall reduction in prices — as they did during the Economic Stabilization Program of the early 1970s. Such "offsets" have been found to reduce the savings from fee reductions by as much as 40%.
Although the magnitude of providers' response to increased risk sharing is less well understood than that of consumers' response to increased cost sharing, providers do respond predictably to incentives for cost control: risk sharing reduces health care utilization and spending. The introduction of Medicare's Prospective Payment System, for example, resulted in a substantial decline in lengths of stay in hospitals. Of course, the quality of care is also of concern, and there are conflicting findings regarding the effect of risk sharing on quality. There are also concerns about incentives for avoiding high-cost patients, but though such "cream skimming" has been found among health plans, there is weaker evidence in the case of providers. There is a paucity of well-controlled studies of risk sharing, however, and most of them confound provider payment with other features of coverage, organization, and care management.
Performance-based payment is increasingly being used to address quality problems in health care. Studies have found modest effects of pay for performance on process measures of quality and intermediate health outcome measures. There are few published data regarding the effect on costs, though one study showed that a pay-for-performance program targeting diabetes quality measures generated savings of 1.5 to 2.5 times the program's cost.5 Savings from pay for performance should not be presumed, however, as demonstrated by Britain's experience with spending overruns due to its Quality and Outcomes Framework for general practitioners.
Although reformers point to economic principles to support market-oriented policies, in many cases there is a gap between belief and evidence. Yet research and practice with such policies provide some lessons. First, consumers and providers respond predictably to policies requiring them to share the burden of health care costs: they find ways to reduce utilization and spending. Second, there is strong evidence that increased consumer cost sharing causes unintended reductions in the use of services that are important for better health — a problem unlikely to be solved by simply offering consumers information or advice. But no similar body of evidence documents negative effects on quality from increased risk sharing by providers. Moreover, providers' information advantage over consumers, along with professional ethics, suggests that essential or effective services are less likely to be underused as a consequence of provider risk sharing than consumer cost sharing. Pay for performance and risk adjustment could minimize any negative effects on the quality of care.
On balance, I believe that reforms focused on provider rather than consumer behavior are more likely to yield lower costs and higher quality while minimizing unintended consequences. To be sure, many questions remain about the ideal design of provider incentives. For example, how can risk-sharing arrangements be used effectively when providers are not part of integrated systems? And more research is needed to establish the appropriate scope and magnitude of pay for performance to complement enhanced risk-sharing regimes.
The focus on providers should not mean an absence of involvement by patients in improving the effectiveness and efficiency of care. Along these lines, there may be promise in efforts to reward consumers for the same care-quality processes included in provider pay for performance. Consumers, too, must have some perceived financial stake — and choices — in cost control. Offering them a financial incentive to entrust their care to a provider team with the capabilities and incentives to deliver coordinated, effective, and efficient care might be a near-term way of accomplishing this goal.
No potential conflict of interest relevant to this article was reported.
Source Information
Dr. Rosenthal is an associate professor of health economics and policy at the Harvard School of Public Health, Boston.
References
| |||||||||||||||||||||||||||||||||||
This article has been cited by other articles:
HOME | SUBSCRIBE | SEARCH | CURRENT ISSUE | PAST ISSUES | COLLECTIONS | PRIVACY | TERMS OF USE | HELP | beta.nejm.org Comments and questions? Please contact us. The New England Journal of Medicine is owned, published, and copyrighted © 2009 Massachusetts Medical Society. All rights reserved. |
Previous
