New directions for Medicare payment systems.
During the past decade, the Health Care Financing Administration (HCFA) has developed and implemented two innovative payment systems. The Social Security Amendments of 1983 (Public Law 98-21) changed the method of payment for inpatient hospital services provided to Medicare beneficiaries from a cost-based, retrospective reimbursement system to a diagnosis-specific prospective payment system (PPS). Likewise, the Omnibus Budget Reconciliation Act (OBRA) of 1989 (Public Law 101-239) dramatically changed the method of paying for physician services provided to Medicare beneficiaries from a charge-based system known as customary, prevailing, and reasonable (CPR) to a resource-based Medicare fee schedule (MFS).
Since their implementation, both of these systems have continued to be refined by HCFA and have been adopted by numerous non-Medicare payers.
The focus of this issue of the Health Care Financing Review is on the ongoing development of Medicare payment methodologies, their adoption by non-Medicare payers, and issues to be addressed in the development of other systems based on these methodologies. This article provides a brief history of the development, implementation, and refinement of these payment methodologies. It also briefly describes the ongoing efforts to facilitate the adoption of these methodologies by non-Medicare payers.
Since October 1983, HCFA has used PPS to reimburse hospitals a fixed price for inpatient episodes of Medicare patients, regardless of how long the patient stays or the resources that the patient actually uses. By setting rates prospectively, incentives are built into the reimbursement system to promote efficiency in the provision of inpatient services.
PPS was developed largely in reaction to rising hospital expenditures and concerns about their impact on the Medicare Part A trust fund (Altman and Young, 1993). PPS continued the movement away from retrospective, cost-based reimbursement that began with earlier cost-containment efforts. These included the Nixon Economic Stabilization Program (ESP), Carter-era hospital cost containment, and hospital ratesetting programs in various States (Gold et al., 1993). The article by Ozminkowski, Gaumer, Coit, and Gabay discusses some of the lessons learned from ESP's wage and price controls on hospitals. Among these lessons, the authors discuss the relative merits of wage and price controls versus ratesetting methods like PPS. They point out that, while wage and price controls do not allow relative prices to change, ratesetting methods typically try to set the "correct" price for individual services.
Under PPS, the basis for payment is a national standardized amount that represents an average payment for a typical case. Diagnosis-related groups (DRGs) based on principal diagnosis and other factors are used to group medically similar cases that require comparable resources for treatment. Each DRG is assigned a weight based on its resource costs relative to the national average. These relative weights are recalibrated each year using the latest available charge data for Medicare discharges. To determine the Medicare reimbursement for an individual episode in a particular hospital, the standardized amount is adjusted by the relative weight of the DRG classification of the patient, the area wage level, the extent of the hospital's teaching activity, and the degree to which the hospital serves low-income patients. In addition, PPS established categories of outlier, or extremely costly, cases which receive supplemental payment amounts.
In the decade since PPS was implemented, HCFA has continued to refine the payment system to help ensure equitable payment for services and to address issues raised by the provider community. Some major concerns that have been addressed include the following:
Under PPS, the DRG assignment based on principal diagnosis is one of the main factors in determining the payment made for hospital inpatient services provided to Medicare patients. A homogeneous grouping of patients facilitates the mapping of the type of patient treated to the average resource cost involved in care. To the extent that the classification system results in the grouping of patients with dissimilar resource costs, the equity of payment based on average resource costs may be problematic. Specifically, hospitals that treat more severely ill patients may be undercompensated. As a result, one of the objectives of refinements to the DRG groupings has been to reduce the variance within these groupings.
Despite the expansion in the number of DRGs from 468 in fiscal year (FY) 1984 to 489 in FY 1994, there continue to be concerns that the current classification system should be improved to compensate hospitals more, equitably for treating severely ill Medicare patients. As described in the article by Edwards, Honemann, Burley, and Navarro, HCFA has recently completed work on developing an expanded set of 652 DRGs that incorporates a severity measure based on secondary diagnoses, which have a major effect on the resources used in treating patients across DRGs. The proposed methodology incorporates aspects from two previously developed systems, the Yale refined DRGs and New York all-patient DRGs. The article includes the rationale for developing severity-adjusted DRGs for the Medicare beneficiary population and a description of alternate severity measurement instruments. A paper describing the severity refinement methodology was announced in the Federal Register (1994). HCFA is currently evaluating comments received and will incorporate changes into the methodology as feasible and appropriate.
Originally, the standardized amount differed according to a hospital's classification as urban, located in a metropolitan statistical area (MSA), or rural. Analysis of 1981 Medicare Cost Report data revealed that the average cost per case was about 20 percent lower in rural hospitals than in urban hospitals after adjusting for differences in case mix, labor costs, and indirect teaching activity. The payment differential reflected this unexplained difference in urban and rural hospital costs. In October 1987, Congress created two separate urban standardized amounts: one for large urban areas (located in an MSA with more than 1 million inhabitants) and one for other urban areas (O'Dougherty et al., 1992).
Beginning in FY 1988, rural hospitals were granted higher updates in the standardized amount than urban hospitals. OBRA 1990 (Public Law 101-508) phased out separate standardized amounts for rural and other urban hospitals between FYs 1991 and 1995. As of October 1, 1994, there is a single standardized amount for other urban and rural hospitals and a higher rate for hospitals in large urban areas. The relationship between costs for urban and rural hospitals has changed very little since the original decision to have separate rates. The decision to eliminate the differential for hospitals outside large urban areas was based on several factors, including the declining financial conditions of many rural hospitals and the potential adverse consequences on rural beneficiaries' access to care (O'Dougherty et al., 1992).
PPS statutes created a category of specialized hospitals and hospital units that continued to be paid based on the cost of providing services subject to a limit on the rate of increase in per case costs. Excluded facilities include psychiatric hospitals and units, rehabilitation hospitals and units, pediatric hospitals, and long-term hospitals. The motivation behind their exclusion was the belief that no existing case-mix measurement system provided a good correlation between diagnosis and resource use for these specialized services. In addition to the statutorily excluded hospitals and units, alcohol and drug treatment facilities and units were granted a 4-year exclusion from PPS. During this period, HCFA refined alcohol and drug abuse DRGs to address concerns raised by the provider community that these DRGs did not adequately capture the differences in length of stay required by detoxification and rehabilitation services. On October 1, 1987, alcohol and drug treatment services were included under PPS.
The implementation of PPS continued cost-based reimbursement for capital costs. The cost-based reimbursement for capital was believed by many to provide hospitals with an incentive to substitute capital for operating costs. Many policymakers felt that Medicare payment policy should encourage the efficient use of resources and not influence hospitals' choices between capital and operating inputs. As a result, OBRA 1987 (Public Law 100-203) mandated the implementation of prospective payment for capital starting October 1, 1992. On October 1, 1991, HCFA initiated a 10-year transition to full prospective payment for capital costs (Cotterill, 1992).
Specially Designated Facilities
From the outset, PPS singled out specific categories of hospitals, such as rural referral centers and sole community hospitals, for more generous payment. OBRA 1989 established another category of essential access small rural hospitals: Medicare-dependent hospitals (MDHs), for which Medicare patients account for more than 60 percent of total inpatient days or discharges. The Consolidated Omnibus Budget Reconciliation Act (COBRA) of 1985 (Public Law 99-272) established additional payments under PPS to hospitals that serve a disproportionately large share of low-income patients. This adjustment was believed necessary to ensure access to care for low-income Medicare beneficiaries (U.S. Congressional Budget Office, 1990).
In January 1992, HCFA implemented the MFS to reimburse physicians for services delivered to Medicare beneficiaries. Under MFS, services delivered to Medicare beneficiaries are classified by the HCFA Common Procedure Coding System (HCPCS), and, in turn, are assigned relative value units (RVUs). The RVU for each current procedural terminology code has three components: a resource-based value for physician work required to produce the services, charge-based values for practice expense, and charge-based values for malpractice expense. The RVUs are converted into payment amounts by applying a conversion factor (CF) and adjusting by a geographic adjustment factor (GAF) to account for geographic differences in the cost of producing physician services.
In moving to the uniform, national, resource-based MFS, a number of perceived inequities in the physician-specific charge-based CPR system were addressed. These included the overcompensation of physicians performing procedures, the undercompensation of physicians for evaluation and management services, and an inappropriately large variation in payment across geographic areas. The MFS system also features restrictions on the ability of physicians to balance-bill Medicare beneficiaries for charges exceeding the fee schedule, and Medicare volume performance standards (MVPSs) designed to control the growth in expenditures.
Several issues related to the development and implementation of the MFS continue to be addressed. They include the following:
The MVPS system is comprised of two components: the volume performance standards (VPSs) and the CF updates. OBRA 1989 specifies the process by which these standards are set and the update factors are determined. VPSs are target rates of growth for physician services that are recommended by the Secretary of the Department of Health and Human Services (DHHS) and approved by Congress. Increases in physician expenditures that differ from the standard can be offset by an adjustment to the CF update, which is also recommended by the Secretary and approved by Congress. OBRA 1989 includes a fallback formula that automatically sets standards and update factors in years for which Congress does not legislate them. While VPSs have been set since 1990, they were not incorporated into the update factor until the implementation of the MFS in 1992.
The MFS was implemented with one CF for all types of services. Since OBRA 1989 specified that VPSs be set separately for medical and surgical services, separate update factors were established for these services resulting in two CFs in FY 1993. As a result of this process, surgical services received a higher update factor than medical services. Since one of the MFS objectives was to increase compensation for cognitive services, OBRA 1993 (Public Law 103-66) created a third service category that separated primary care services from other medical services.
Geographic Adjustment Factors
MFS payments for individual services differ across MFS payment areas called localities according to locality-specific GAFs. The GAF is a weighted average of three geographic practice cost indexes (GPCIs) that measure geographic variations in the cost of physician work, practice, and malpractice expenses. For physician work, the index was based on geographic variation in hourly earnings for non-physician professions with 5 or more years of college. For practice expenses, the index was based on geographic variation in housing rents and in the hourly earnings of nurses, technicians, and clerical workers. For malpractice expenses, the index was based on geographic variation in malpractice premiums (Levy and Borowitz, 1992).
Since the implementation of MFS, these indexes have been criticized for several reasons. These include the age of the 1980 census data underlying many components of the indexes and the use of residential rents as a proxy for commercial rents. For 1996, the GPCIs will be based on more recent data: 1990 census data, 1994 Department of Housing and Urban Development data on fair market rent and a 3-year average (1990-92) of malpractice premium data. The transition to these update GPCIs begins in 1995. The use of residential rents as a proxy has been supported by studies showing a high correlation with indexes of commercial rents (Dayhoff and Pope, 1994; Gillis, Reynolds, and Willke, 1991).
Relative Values for Practice and Malpractice Expenses
OBRA 1989 specified that relative values for practice and malpractice expenses be calculated based on historical Medicare charge data. While OBRA 1993 imposed some constraints on the relative values for practice expenses, HCFA has initiated a more extensive effort to acquire data and develop methodologies needed to generate cost-based relative values for both practice costs and malpractice expenses. The Social Security Amendments of 1994 (Public Law 103-432) require that cost-based relative values for practice expenses be implemented in 1998.
Although MFS will not be fully implemented until 1996, impact studies prepared by DHHS and the Physician Payment Review Commission (1994) suggest that the MFS has already been successful in achieving the objectives of narrowing the disparity in compensation between medical and procedural services and among geographic areas. These studies have also assessed preliminary physician response. So far, physician participation and assignment rates have increased since MFS implementation (Health Care Financing Administration, 1994b).
Another area of concern is whether physicians react to fee reductions by increasing service utilization. A potential clue is offered by the episode-of-care analysis described in the article by Lee and Mitchell. They explored whether physicians responded to the surgical fee reductions for six procedure groups included in OBRAs 1986 (Public Law 99-509) and 1987 by providing more services as part of the surgical episode. The surgical groups involved are cataract extractions, total hip replacement, total knee replacement, coronary artery bypass graft (CABG), endoscopy of the upper gastrointestinal tract, and prostatectomy. Results show that two of these procedures (CABG and cataract) give evidence for the existence of a service volume offset to the fee reductions. Both CABGs and cataract extractions account for nearly one-half of Medicare revenues for cardiothoracic surgeons and ophthalmologists, compared with the other four procedures, which account for about 19 percent of the volume for the associated specialties. As a result, Lee and Mitchell's findings are consistent with the hypothesis that an intervention impacting a greater proportion of provider revenues is likely to elicit a larger and more immediate offset response.
EMERGING LONG-TERM CARE PAYMENT METHODOLOGIES
While reimbursements for nursing facility services account for only 3.5 percent of Medicare program expenditures, they represent a much larger proportion, 25 percent, of Medicaid program expenditures (Health Care Financing Administration, 1994a). As a result, both Federal and State governments have participated in efforts to develop payment systems that encourage the efficient delivery of institutional long-term care without compromising access to those services.
A primary emphasis of HCFA's work in the last decade has been on designing prospective systems that adequately respond to the varying types of beneficiaries receiving different types of services in skilled nursing facilities (SNFs) (Health Care Financing Administration, 1991). The development of such prospective, case-mix adjusted, per diem systems was hampered by a variety of data and methodologic problems (Liu et al., 1986; Fries et al., 1987; Morris et al., 1990). Many of these obstacles have now been overcome, so that it has been possible to develop and test several patient-level case-mix classification systems (Fries et al., 1987). One of these is Resource Utilization Groups (RUGS), which classifies patients based on costs according to the relationship of patients' various medical, functional, and personal characteristics, and their daily use of staff time. The use of staff time is measured according to direct and indirect time spent by all levels of nursing staff, social workers, and therapists. RUGS originally were developed for reimbursement of care received by Medicaid residents in nursing homes. More recently, the concept was adapted and refined for paying for Medicare-covered patient care in certified SNFs.
RUG-based classification and payment systems are appealing for several reasons. First, they could provide greater incentives for SNFs to accept more Medicare patients, particularly those requiring technical services or heavy care. Second, because patient classifications are well defined, providers would have greater confidence that all patients would be appropriately covered for payment. Finally, a RUG-based payment methodology could lead to integrated payment systems for Medicare and Medicaid patients, which would reduce facilities' administrative burden (Health Care Financing Administration, 1988).
RUG-I was developed by researchers associated with Yale, based on their experiences in developing DRGs and using an extensive professional standards review organization long-term care data base. It was successful in explaining 37 percent of variance in staff time costs across all residents, but was never implemented. The article by Schultz, Ward, and Knickman presents an evaluation of the next generation of RUG-type classification systems, RUG-II, over its first 5 years in New York State. It was implemented by the New York State Department of Health (NYDOH) in 1986 in a movement away from uniform per diem reimbursement based on facility-specific historical costs. With RUG-II, NYDOH hoped to match more closely payment with intensity of care and provide strong financial incentives for the admission of more patients requiring a high intensity of care. The authors conclude that RUG-II apparently was successful in encouraging the admission of patients requiring a high intensity of care. This may, however, be at the expense of access to long-term care for patients with low-intensity care needs because of a perception that reimbursement for these patients is inadequate. In addition, Schultz, Ward, and Knickman found that the financial status of long-term care facilities deteriorated over the 5-year period. The authors offer several possible causes for this decline, including the system of cost corridors incorporated into RUG-II, which sought to limit payment to high-cost facilities.
Building on research done to develop RUG-II, a RUG-type classification for Medicare SNF patients was developed using high-volume providers in New York, Pennsylvania, Florida, Illinois, and California (Fries et al., 1987). RUG-III goes beyond RUG-II by including more patient types and different kinds and levels of care to accommodate both Medicare SNF and Medicaid long-stay populations. For example, RUG-III recognizes residents with cognitive impairments as well as those with behavioral symptoms. RUG-III also subdivides and expands the 16 RUG-II resource groups into 44 groups, with 12 groups for those needing special rehabilitation services. These 44 groups assess differing dimensions and levels of care in such areas as rehabilitation services and the special needs of clinically complex patients and those requiring extensive care. RUG-III is expected to provide incentives for facilities consistent with the statutory requirement of OBRA 1987 to improve or maintain Medicare and Medicaid residents at their highest practicable level of physical, mental, and psychosocial well-being.
The article by Cornelius, Feldman, Marsteller, and Liu analyzes the expected impact of the RUG-III classification system using HCFA's claims history data for Medicare SNF patients nationwide. The analyses were undertaken to estimate the Medicare case mix for facilities in a six-State demonstration (Burke, 1990). The result is the first longitudinal assessment of case-mix change on a national basis. The authors found that applying the RUG-III classification scheme to all Medicare SNF stays, unadjusted for facility type, accounts for 20 percent of the variance in average covered charges per stay.
ADOPTION BY OTHER PAYERS
Over the past few years, several health care reform proposals have included provisions for optional provider payment rates, based upon Medicare payment methodologies, that could be adopted by other public and private payers. As a precursor to the development of these rates, HCFA sponsored two studies to assess the extent to which non-Medicare payers already use Medicare payment methodologies and to describe how these methodologies have been modified to accommodate the needs of other payers.
The article by Carter, Jacobson, Kominski, and Perry reports widespread use of Medicare hospital payment methodologies by other governmental and private payers. They include two-thirds of Blue Cross/Blue Shield Association Plans that use DRGs for at least 1 of their hospital insurance products and 21 States that use DRG-like systems for their Medicaid programs. Rather than just adopting the Medicare weights and payment rates, DRG users for non-Medicare populations have developed widely varying diagnosis-related, per discharge prospective systems. The authors discuss the extent to which PPS payment rules, regulatory procedures, and data sources have been modified to accommodate the needs of other payers. A very flexible payment system has emerged in which the only obvious constant is the use of DRGs as a measure of output.
Similarly, the article by McCormack and Burge reports the rapid diffusion of the resource-based relative value scale (RBRVS) to non-Medicare payers in the short period of time since Medicare's implementation of the MFS. Among the respondents to a large mail survey of payers, one-third reported having adopted Medicare's RBRVS. Like DRGs, most non-Medicare payers adopting RBRVS are not implementing other elements of MFS, including the Medicare conversion factor. Instead, they are tailoring the RBRVS payment system to their own circumstances. For example, several payers use RBRVS as a fee screen to set maximum allowed charges for preferred provider organizations (PPOs). Another 40 percent of survey respondents reported that they were considering adopting RBRVS. This leads the authors to conclude that the potential for increased use of RBRVS is substantial.
In both of these studies, the investigators asked non-Medicare payers about their motivation to either adopt or not adopt Medicare payment methodologies. One of the principal attractions to non-Medicare payers of both the Medicare hospital and physician payment systems is their flexibility. As a result of this flexibility, DRGs and RBRVS have been integrated into broader cost-containment strategies by non-Medicare payers and tailored to the needs of local providers and of the local patient population. Another frequently cited motivation for adoption is the ability to minimize cost shifting by public payers. The most frequently cited reasons for not adopting these systems relate to the adequacy of the patient classification systems and the relative weights (or values) for the equitable reimbursement for services delivered to non-elderly patients.
DIRECTIONS FOR THE FUTURE
HCFA is pursuing several activities to facilitate the ongoing development and evaluation of Medicare payment methodologies and the adoption of these methodologies by non-Medicare payers. These developmental activities include: a national, standard, and integrated claims processing system; severity-adjusted DRGs; all-payer DRG weights and RVUs for services not covered by Medicare; and information on case-mix adjusted costs and charges by payer and geographical area.
Payment Systems for Other Payers
Several factors led HCFA to consider its appropriate role in facilitating the adoption of Medicare payment methodologies by other payers. First, several States have included provisions for all-payer or multi-payer systems in their health care reform initiatives. For example, Minnesota will implement a regulated all-payer option, which includes mandatory provider fee schedules based on Medicare payment methodologies for insurers and providers furnishing services outside a capitated managed care system. Second, PPOs negotiating discounted rates with providers can use DRGs and RVUs to set relative prices across services. Negotiations can then focus on the overall level of prices. Finally, as mentioned previously, several national health care reform proposals have included provisions for optional provider payment rates based upon Medicare payment methodologies.
Since the DRG system used by Medicare was originally developed by Yale researchers using data from all types of cases, it could theoretically be appropriate for other payers. Nonetheless, many payers claim that Medicare DRGs do not provide sufficient discrimination among non-elderly patient groups. Areas of specific concern include neonatal, psychiatric, substance abuse, rehabilitation, and transplant cases. Non-Medicare payers could use an alternate patient classification system that explicitly recognizes differences in patient severity of illness. Several of the currently available systems involve a large number of classification groups, raising the issue of low-volume groups and the instability of weights across years. As an alternative, the methodology used by HCFA to develop the severity-adjusted DRGs for Medicare patients could be adapted and expanded by non-Medicare payers for their patient populations. This would provide severity-adjusted DRGs for non-Medicare patients with fewer DRG categories than the 1,437 DRGs contained in the all-patient refined DRGs.
Previous empirical studies have shown that costs within DRGs differ systematically according to the patients' source of payment Morpe, 1987). Non-Medicare payers could calculate payer-specific or multi-payer (i.e., all patients outside Medicare) relative weights. To assist non-Medicare payers in choosing appropriate weights, HCFA funded research by the Urban Institute to develop all-payer and payer-specific DRG weights using all-payer hospital discharge data from 19 States. These weights will be developed for both the Medicare and the New York all-payer DRGs, which are an expansion of the Medicare DRGs to include newborn and neonate categories. In addition, this research will provide information on how the average case-mix adjusted costs vary by geographic area, payer type, and provider type.
Most payers accept the methodology underlying the development of RBRVS and the developers' intent to describe resource use by a typical patient, regardless of payer. Nonetheless, non-Medicare payers are concerned about the absence or inadequacy of RVUs for some frequently used services not typically provided to the Medicare population, such as obstetrics/gynecology and pediatrics. Using data from several payers, the Urban Institute will fill RVU gaps for services not covered by Medicare and develop CFs by type of service, geographic area, and payer type. These CFs will provide information on how case-mix adjusted payments vary by geographic area and payer type. Both the physician and hospital components of this research project will be completed in 1995.
In addition to decisions about the appropriate patient classification system and relative weights (or values), the choice of a CF or payment rate is another issue facing any payer or group of payers adopting Medicare payment methodologies for both hospitals and physicians. The article by Kominski and Rice discusses this choice in the context of aff-payer systems. Two fundamental decisions must be made. First, an appropriate price level must be chosen. Second, a decision is needed about whether the current, substantial payment differentials among payers should be retained, reduced, or eliminated. Based upon an analysis of California hospital discharge data, the authors argue that a single set of payment rates should not be applied to all payers because resource use varies within DRGs. Consistent with a previous analysis by Thorpe (1987), the authors advocate the use of payer-specific weights and payer-specific CFs to adjust for differences in resource costs across payers.
Medicare Transaction System
In the past, the development and evaluation of Medicare payment methodologies were hindered in part by a fragmented claims processing system. Medicare currently has 79 contractors using 10 independent systems at 62 sites across the country. The article by Warren, Jackson, and Veiel describes the development of a new claims processing system, the Medicare Transaction System (MTS). MTS will consolidate all claims processing for both Part A and Part B services into one national, standard, integrated claims processing system.
Besides improving services to beneficiaries and providers by supporting a single point of information, the MIS will also facilitate the development and evaluation of payment policies by providing more timely, accurate, and uniform data. The integrated processing of claims for Part A and Part B services could simplify the development of bundled payment systems that help to align physician with hospital incentives. The development of a national provider and payer identification system will be especially helpful for evaluations by allowing the verification of the same provider in multiple health plans.
The articles presented in this issue describe some recent or ongoing developments in the creation, implementation, and refinement of Medicare payment methodologies. The adoption of Medicare payment systems by non-Medicare payers is also described. Overall, the articles underscore Medicare's major contributions to how hospitals, physicians, and nursing homes are paid in the United States. The extensive adoption and adaptation of PPS and the MFS by non-Medicare payers nationwide indicates the systems' general flexibility and their credibility among various private insurers and States. In addition, the high adoption rates of PPS and the MFS reinforce their utility as a platform for the development of new payment policies, such as all-payer systems under State health care reform. Similarly, Medicare-developed classification schemes, such as RUGS and severity DRGs, are leading the way toward more precise and equitable payment systems for more severely ill patients. The development of payment systems and their associated classification systems has been facilitated by HCFA's unique national patient data base. The development of the state-of-the-art MTS is expected to further facilitate the development and evaluation of payment policies.
In sum, Medicare has led the development and implementation of innovative payment systems for hospital and physician services in the past decade. HCFA has also fostered the development of innovative payment systems for nursing home services. This leadership will continue in this decade as exemplified by the development of MIS and with continued efforts to facilitate the adoption of Medicare payment methodologies by other payers.
The authors gratefully acknowledge the contributions of Sherry Terrell, Ph.D., Elizabeth Cornelius, Ph.D., and Philip Cotterill, Ph.D. of HCFA in the preparation and review of this article.
Altman, S., and Young, D.: A Decade of Medicare's Prospective Payment System--Success or Failure? Journal of American Health Policy 3(2):11-19, March/April 1993.
Burke, R: The Multistate Nursing Home Case Mix and Quality Demonstration. The Circle, Inc. February 28, 1990.
Cotterill, P.G.: Prospective Payment for Hospital Capital. Health Care Financing Review 1991 Annual Supplement. Pp. 79-86, March 1992.
Dayhoff, D. and Pope, G.: Comparison of GPCI Rental Index to Three Sources of Commercial Rent. Prepared for the Health Care Financing Administration. Waltham, MA., 1994.
Federal Register: Medicare Program; Hospital Inpatient Prospective Payment Systems and Fiscal Year 1995 Rates. Proposed Rule, Vol. 59, No. 102, 27715-27716. Office of the Federal Register, National Archives and Records Administration, Washington. U.S. Government Printing Office, May 27, 1994.
Fries, B., Schneider, D., Dowling, M., et al.: A Case Mix Measure for Medicare Skilled Nursing Facilities. Cooperative Agreement Number 18-C-98581/2-01. Prepared for the Health Care Financing Administration. Baltimore, MD. 1987.
Gillis, K, Reynolds, R., and Willke, R.: Assessing the Validity of Geographic Practice Cost Indices. Chicago. American Medical Association Center for Health Policy Research Discussion Paper 91-5. 1991.
Gold, M., Chu, K, Felt, S., et al.: Effects of Selected Cost-Containment Efforts: 1971-1993. Health Care Financing Review 14 (3):183-225, Spring 1993.
Health Care Financing Administration: 1994 HCFA Statistics. Baltimore, MD. 1994a.
Health Care Financing Administration: Report to Congress: Monitoring the Impact of Medicare Physician Payment Reform on Utilization and Access. Baltimore, MD. 1994b.
Health Care Financing Administration: Report to Congress: Prospective Payment for Medicare Skilled Nursing Facilities. Baltimore, MD. 1988.
Health Care Financing Administration: Report to Congress: Skilled Nursing Facilities Prospective Payment Demonstration: Report on Research Findings. Baltimore, MD. 1991.
Levy, J.M., and Borowitz, M.: The Medicare Fee Schedule Unveiled: An Account of Physician Payment Reform. Health Policy 22:263-286, 1992.
Liu, K, Weiner, J., Schieber, G., and Doty, P.: The Feasibility of Using Casemix and Prospective Payment for Medicare Skilled Nursing Facilities. Inquiry 23(3): 365-370, 1986.
Morris, J.N., Hawes, B.E., Fries, B., et al.: Designing the National Resident Assessment Instrument for Nursing Homes. The Gerontologist 30 (4):293-307, 1990.
O'Dougherty, S.M., Cotterill, P.G., Phillips, S., et al.: Medicare Prospective Payment Without Separate Urban and Rural Rates. Health Care Financing Review 14(2):31-47, Winter 1992.
Physician Payment Review Commission: Annual Report to Congress, 1994. Washington, DC. 1994.
Thorpe, KE.: The Distributional Implications of Using DRG Relative Prices in DRG Payment Systems. Inquiry 24(1):85-95, Spring 1987.
U.S. Congressional Budget Office: Medicare's Disproportionate Share Adjustment for Hospitals. Washington. U.S. Government Printing Office, May 1990.
Reprint Requests: Brigid Goody, Sc.D., Office of Research and Demonstrations, 2-B-14 Oak Meadows Building, 6325 Security Boulevard, Baltimore, Maryland 21207.
The authors are with the Office of Research and Demonstrations, HCFA The opinions expressed are those of the authors and do not necessarily represent the views policy positions of HCFA
|Printer friendly Cite/link Email Feedback|
|Title Annotation:||Medicare Payment Systems: Moving Toward the Future|
|Author:||Goody, Brigid; Friedman, Maria A.; Sobaski, William|
|Publication:||Health Care Financing Review|
|Date:||Dec 22, 1994|
|Previous Article:||Publication weighs hard choices in paying for long-term care.|
|Next Article:||Hospital wage and price controls: lessons from the Economic Stabilization Program.|