Under the current Medicare payment system, hospitals are reimbursed for inpatient stays on the basis of each patient’s assigned Medicare severity diagnosis-related group (MS-DRG). The Centers for Medicare and Medicaid Services (CMS) attempts to set reimbursement rates so that the average hospital has the same profit margin for the care they provide regardless of a patient’s MS-DRG. The finding that some services, such as orthopedic surgeries, are systematically more profitable than others, however, suggests that there is still mispricing in payment rates.1 Such mispricing has encouraged providers to increase capacity and patient volume for the most profitable services and reduce capacity for the least profitable ones — ultimately reducing the overall efficiency of the health care system.2
Although Medicare is moving toward new payment models, such as accountable care organizations (ACOs) and bundled payment, the MS-DRG–based payment rates continue to serve as the building blocks for these new models. As a result, mispricing in the current payment rates will continue to create incentives for hospitals to favor services with higher profit margins.
CMS sets Medicare payment rates to reflect the relative costs that hospitals incur in treating patients in each MS-DRG, but its cost-measurement system is fairly crude. In response to increasing pressure to control costs, many U.S. hospitals have adopted superior internal cost-measurement systems that enable them to estimate costs with greater accuracy and to identify services with high and low profit margins.
We believe that CMS could substantially improve Medicare’s payment system for inpatient care by drawing on the highly detailed and accurate cost data that are produced by these sophisticated internal systems. In addition to yielding more accurate payment rates, this arrangement would synchronize the cost data used by hospitals and those available to CMS, thereby eliminating any possibility for hospitals to use their superior data to cherry-pick the most profitable services. As a result, the overall efficiency of the health care system would be increased. The wide adoption of internal cost-measurement systems in recent years would allow for fairly rapid implementation of this model.
Since cost data aren’t available at the MS-DRG level, CMS currently uses the charges billed to the agency for the services delivered to each Medicare patient in estimating the relative cost of treating patients in each MS-DRG. Charges are set by the hospital and are summarized in the chargemaster, a list of all the clinical services that the hospital offers. The charges are aggregated into 19 departments for which cost data are available from the Medicare Cost Report, such as “operating room” and “laboratory.” To approximate the costs associated with delivering care in each of those departments, CMS assumes a uniform markup rate for all services in a given department. The cost estimate for an MS-DRG is the sum of the cost estimates for care in all 19 departments.
The current method is crude, in part because, as studies have shown, markups actually vary a great deal within individual departments.3 Hospitals’ newer internal cost-measurement systems are expected to produce more accurate relative cost estimates for the MS-DRGs because they are much more granular.4 These systems, which are typically licensed from external vendors, produce a cost estimate for each item on the hospital’s chargemaster — which typically includes tens of thousands of services. The fact that many hospitals have made substantial investments in these systems indicates that their leaders believe that the resulting cost information is valuable above and beyond the information that must be provided in the Medicare Cost Report.
Using 2013 survey data collected by the Healthcare Information and Management Systems Society,5 we estimate conservatively that 37% of the 3580 U.S. hospitals whose cost data are currently included in the Medicare payment-rate estimation have adopted a granular internal cost-measurement approach. Although larger hospitals are overrepresented, many smaller hospitals have been adopting such methods through their membership in multihospital systems.
CMS would not need to expand reporting requirements in order to make use of the data collected through these systems, but a representative sample of hospitals would have to submit cost data in a standardized format. The agency could encourage hospitals to participate by providing subsidies that offset the costs of compiling these data. Key to promoting participation would be convincing hospitals that the confidentiality of their data would be protected.
Germany has effectively used data from hospitals’ internal cost-measurement systems to calculate payment rates in its version of a DRG-based payment system for inpatient hospital care. Its system reduces the burden on hospitals by eliminating the requirement that they prepare cost reports that have value only for the payer. Moreover, the German model has been successful in fostering the development of a high-quality infrastructure for measuring and managing costs.
We believe that by drawing on the systems that many U.S. hospitals have adopted, CMS could substantially increase the accuracy of the Medicare payment rates currently used for inpatient hospital care and, by extension, the accuracy of new payment models such as ACOs and bundled payment. Building new payment models on a foundation that makes all MS-DRGs equally profitable and that fosters the development of a high-quality cost-measurement and cost-management infrastructure will increase the likelihood that providers will respond to the broader incentives in these new models.
From the Clinical Excellence Research Center, Stanford University School of Medicine, Stanford (M.E.), and the Sol Price School of Public Policy, University of Southern California, Los Angeles (P.B.G.) — both in California; and the Center for Health Policy, Brookings Institution, Washington, DC (P.B.G.).
1. Gamble M. What 4 factors should determine a hospital’s service line strategy? Becker’s Hospital Review. April 5, 2012 (http://www.beckershospitalreview.com/strategic-planning/what-4-factors-should-determine-a-hospitals-service-line-strategy.html).
2. Berenson RA, Bodenheimer T, Pham HH. Specialty-service lines: salvos in the new medical arms race. Health Aff (Millwood) 2006;25:w337-w343. CrossRef | Web of Science | Medline
3. Dobson A, DaVanzo J, Doherty J, Tanamor M. A study of hospital charge setting practices. Falls Church, VA: The Lewin Group, December 2005 (http://220.127.116.11/documents/Dec05_Charge_setting.pdf).
4. Labro E, Vanhoucke M. A simulation analysis of interactions among errors in costing systems. Account Rev 2007;82:939-962. CrossRef | Web of Science
5. The Dorenfest Complete Integrated Healthcare Delivery System Plus (IHDS+) Database and library. Chicago: Healthcare Information and Management Systems Society Foundation.
This Perspective article originally appeared in The New England Journal of Medicine.