Learning from Medicare: Prospective Payment

Print

In the wake of the passage of the Patient Protection and Affordable Care Act (Affordable Care Act or ACA) in 2010, policymakers are grappling with implementation of several Medicare-related provisions in the legislation. These include demonstrations and pilots on payment bundling, pay-for-performance, and value-based purchasing that have the potential to realign incentives toward high-quality health care. In addition, rising concern over the federal deficit and debt has led various leaders and commissions to recommend mechanisms to control the rates of spending increases in mandatory programs such as Social Security, Medicaid, and Medicare.¹ These two developments are likely to shape Medicare policy significantly in the years to come. As policymakers consider specific Medicare payment reforms, it is useful to reflect on past efforts to change what and how the program pays for health care services, as well as HCFO’s long-standing role in supporting research examining the effects.

This is the first of three Hot Topics over the next few months that will evaluate previous Medicare reforms and identify some lessons that may be relevant to the current policy environment. The series will review three aspects of the Medicare program that have influenced the larger health care system—the development of a prospective payment system, the coverage decision process, and the growth of Medicare managed care within Medicare Advantage. Through its grant making function, HCFO has funded research on major changes to Medicare over the years and drawn insights into what these changes might mean for the larger health care system.

Background: Prospective Payment

In 1982 Congress mandated the development of a prospective payment system (PPS) to control Medicare costs. Previously, the program utilized a retrospective, cost-based reimbursement system to pay hospitals for services. The prospective payment model had been successful in several states, and Congress hoped to control escalating hospital costs—between 1967 and 1983, annual payments to hospitals rose from $3 billion to $37 billion.² Under the PPS, Medicare pays hospitals a set amount for each hospital stay. The goal is to reward efficient hospitals and incentivize less efficient hospitals to improve their operations. 

The PPS established a per-case reimbursement method based on diagnosis-related groups (DRGs). Each of the almost 500 DRGs has an assigned payment weight based on the average resources required to treat Medicare patients in that group.³ Payment rates are also adjusted for several factors including geography, co-morbidities, sex, and some specific procedures performed during the hospital stay.⁴ Hospitals may receive additional subsidies through the PPS if they help train physicians or treat large numbers of low-income patients (“disproportionate share payments” or DSH).⁵ While prospective payment began with inpatient hospital care, different forms of it have been applied to all other types of Medicare services including physician, hospice, outpatient hospital, and post-acute care.⁶

Since the inception of prospective payment, providers and policymakers have raised concerns about adequacy of Medicare reimbursement and the effect of these cost controls on quality of care. In its most recent study of the impact of the inpatient PPS on hospital financial performance, the Medicare Payment Advisory Commission (MedPAC) concluded that the system was functioning largely as intended. MedPAC found that while certain payment components may need to revisited, a substantial portion of hospitals’ Medicare inpatient margins are attributable to operating characteristics at least partially under management control. Accordingly, hospital managers have the ability to influence the efficiency and cost of care.⁷ Research from the RAND Corporation on the early effects of the PPS also found that it did not lead to significant declines in the quality of care. The research did find, however, that more Medicare patients were discharged in unstable condition than before the implementation of PPS.⁸ 

Analysis of the PPS payment policy has implications beyond Medicare. Due to the immense size of the program, Medicare’s actions affect the behavior of providers and private payers throughout the Unites States.⁹ Many private payers have adopted methods of prospective payment for most of their covered services. Several Medicare payment provisions in the ACA also have the potential to exert influence beyond the Medicare program. A pilot program in bundled payment, beginning in 2013, will take the efficiency incentives of prospective payment a step further—certain conditions will be subject to a single payment for the complete range of necessary treatment over a period of time. A single payment will cover inpatient hospital services, outpatient hospital services, and post-acute services for an episode of care.

Although bundled payment is intended to encourage efficiency in much the same way that prospective payment does, it introduces greater risk for Medicare providers by creating the potential for both shared savings and revenue loss.¹⁰ In recent years Medicare has undertaken smaller bundling experiments, including the Acute Care Episode (ACE) Demonstration.¹¹ The private sector has also engaged in some limited experimentation with bundling.¹² However, the payment bundling pilot mandated by the ACA is much larger in scale and has the potential to influence the whole health care system.

On April 29, 2011, the Department of Health and Human Services (HHS) announced the launch of another ACA-mandated Medicare payment provision.  The Hospital Inpatient Value-Based Purchasing (Hospital VBP) program will connect reimbursement to the quality of patient care rather than just the quantity of services received. Under the Hospital VBP program, payment will now be tied directly to performance scores on certain conditions including heart attack, pneumonia, and healthcare associated infections. Additionally, the patient experience of care will also be part of the performance measures.¹³ This change and the bundling program both represent attempts to pay for performance and incentivize quality and efficiency in Medicare beyond the current PPS. Because the implementation of the PPS also represented a large change in Medicare payment policy, earlier research on the effects of the system may help shed light on the consequences of these broader changes to payment policy for cost, quality, and access to care.

HCFO Research on Prospective Payment

HCFO grantees began evaluating the inpatient PPS not long after it was first introduced.  Early HCFO-funded work by Frank Sloan, Ph.D, of Duke University, utilized data from the first two years of the system (1984-1985) to examine its impact on hospitals. His work found several trends in the early years of the PPS. Medicare discharges as a percent of total discharges remained constant between 1983 and 1985, but the number of Medicare and total discharges per hospital declined. The number of complex DRGs also increased for both Medicare and non-Medicare patients.¹⁴ Sloan also analyzed the early effects of the PPS on the likelihood of hospitals discharging Medicare beneficiaries to skilled nursing facilities (SNFs), intermediate care facilities (ICFs), and home health agencies. He and his colleagues found that after controlling for hospital and patient characteristics, the probability of transfer increased in almost all DRGs and discharge destinations studied. The effect was particularly strong for the probability of transfer to SNFs. Additionally, length of stay in the hospital before transfer to sub-acute facilities declined in almost all DRGs. The results of this work suggested that payment experiments with bundling were warranted in order to ensure that patients were not being discharged from hospitals too quickly.¹⁵ The bundling provisions of the ACA represent an attempt to eliminate hospitals’ financial incentives to discharge patients prematurely.

Other early HCFO-funded work explored hospital profitability and cost-shifting. Research led by Jack Hadley, Ph.D., and colleagues examined hospitals’ profits and losses to see if they were consistent with the PPS’s intent to reward efficiency and penalize inefficiency. Their work assessed the relationship between both total hospital profits and Medicare PPS profit margins and efficiency. Efficiency was defined as the lowest hospital cost for hospitals of a given size, adjusting for several factors including payer mix, quality, and Medicare patient characteristics. Their results showed that the least profitable hospitals constrained the growth in their expenses more than higher profit hospitals and experienced a slight decrease in inefficiency. The researchers concluded that payment reforms that put financial pressure on hospitals can result in improved efficiency without significant cost-shifting.¹⁶ 

The Balanced Budget Act (BBA) of 1997 extended prospective payment for Medicare patients in nursing facilities. With the support of a HCFO grant, David Gifford, M.D., of Brown University, examined the effect of the PPS on SNFs in terms of access to care, case-mix changes, and outcomes. Using data from nursing facilities in Ohio, Gifford and his colleagues found that after the implementation of the PPS there was only a small decrease in the in the proportion of the costliest patients discharged to nursing facilities compared to the years before the BBA. While the reduction was small, it was concentrated within specific types of facilities, and the researchers recommended close monitoring of ongoing policy updates to ensure adequate access.¹⁷

The BBA of 1997 also gave Medicare the authority to establish a PPS for outpatient hospital services.¹⁸ A recently awarded HCFO grant addresses the implications of this system for the number of Medicare patients in hospitals and for the volume of outpatient care. Jennifer Mellor, Ph.D., of the College of William and Mary, is examining whether the outpatient PPS led hospitals to substitute away from Medicare patients toward those with other coverage and if it resulted in a shift of care from outpatient departments to inpatient stays for conditions that could be treated in either setting.

Additional HCFO-supported analysis explores the responsiveness of particular service lines to changes in DRG reimbursement. Research from Kevin Volpp, Ph.D., University of Pennsylvania, is examining the quality of care that patients hospitalized under more and less profitable conditions received in 1995, 2000, and 2006. He is testing the effect of the changes in reimbursement on mortality for these service lines, with a goal of providing policymakers with a better ability to measure profitability and quality of different service lines.¹⁹

Conclusion

While prospective payment began almost 30 years ago as a way to control rising inpatient hospital costs, it has grown to encompass all forms of Medicare fee-for-service payment as cost containment concerns have become more widespread. As Medicare payment continues to evolve through the ACA and other pilots in bundling, pay-for-performance, and value-based purchasing, research will continue to help decision-makers understand the intended and unintended impacts of these policy innovations. Details on HCFO studies and related publications can be found at http://www.hcfo.org.

Related HCFO Grants

Title: CCRCs: An Efficient Alternative for Long-Term Care Provision and Financing?
Grantee Institution: Duke University
Principal Investigator: Frank A. Sloan, Ph.D.
Grant Period: January 1, 1992-December 31, 1994
Are Continuing Care Retirement Communities (CCRCs) an efficient alternative for the provision of long-term care, and can public programs be adapted to assist in financing this model? This study examined who enters CCRCs and why, which factors influence nursing home utilization in CCRCs, and financial stability of CCRCs by conducting a survey of CCRC residents and analyzing data from the 1989 National Long-Term Care Survey and the American Association of Homes for the Aging survey of CCRCs. Investigators also gathered primary data from selected states to evaluate CCRC financial solvency. Finally, by studying specific regulations, they assessed the potential of CCRCs as mechanisms for providing and financing long-term care for the low and moderate income elderly.
 

Title: Efficiency/Quality/Outcome Trade-Offs in Medicare’s Prospective Payment System
Grantee Institution: Georgetown University, Center for Health Policy Studies
Principal Investigator: Jack Hadley, Ph.D.
Grant Period: June 1, 1991-October 31, 1993
Are hospitals' profits and losses consistent with Medicare's Prospective Payment System's (PPS) intent to reward efficiency and penalize inefficiency? How do hospitals respond to their different financial positions and what is the impact of their responses on quality, process, and outcomes of care? This study assessed the relationship between hospital profits (both total profit and PPS profit margins) and efficiency (defined as the lowest hospital cost for hospitals of a given size, adjusting for payer mix, quality, input prices, Medicare patient characteristics, and outcomes). Using 1987-1989 hospital and Medicare data, the project examined whether differences in profit rates are associated with changes in the proportion of Medicare cases flagged for potentially poor quality; Medicare patients' outcomes; volume and mix of both Medicare and non-Medicare patients treated and services offered; staffing levels and capital spending; average case-mix, severity level, and demographic characteristics of Medicare patients treated; and efficiency and profitability in the next year.
 

Title: Impact of Profitability on Hospital Responses to Financial Stress
Grantee Institution: University of Pennsylvania School of Medicine
Principal Investigator: Kevin Volpp, Ph.D.
Grant Period: December 1, 2007-November 30, 2010

The researchers will examine the impact of financial pressure on hospitals on the quality of care provided. They will test whether hospitals' responses to a change in the level of reimbursement is likely to vary by DRG-specific incentives, using the Medicare BBA as an example. They would: 1) calculate the generosity of Medicare payment by diagnoses and service lines for 1995, 2000, and 2005; 2) compare the quality of care received by patients hospitalized with conditions from more vs. less profitable service lines in 1995, 2000, and 2006; 3) test the effects of changes in reimbursement on mortality for more vs. less generously reimbursed diagnoses and service lines; and 4) test the effects of changes in reimbursement on Patient Safety Indicators (PSIs) for more vs. less generously reimbursed diagnoses and service lines. The researchers note that existing studies on the effects of financial stress on hospitals have examined acute myocardial infarctions, one profitable condition that may provide a misleading sense of the overall impacts on quality and cost/quality tradeoffs. The objective of this study is to provide policymakers with better ability to measure profitability and quality for hospital service lines (e.g., neurosurgery, cardiology, etc.), the level at which many important decisions about resource allocation are made.
 

Title: The Impact of the Prospective Payment System on Nursing Home Care
Grantee Institution: Brown University
Principal Investigator: David Gifford, M.D.
Grant Period: June 1, 2000-May 31, 2002
What are the effects of the Medicare Prospective Payment System (PPS) for skilled nursing facilities (SNFs) on access to, and case-mix changes in, SNFs, and on outcomes for SNF patients in Ohio? The SNF PPS creates a fixed, all-inclusive, per diem reimbursement rate per patient, based on where that patient fits within a resource utilization group (RUGs) classification system. For some high-need RUGs, the cost of care may be higher than the per diem rate set by the SNF PPS due to increased pharmaceutical use, the costs of which may not have been fully assessed when calculating the per diem. The researchers hypothesized that instituting a prospective payment system may give SNFs the incentive to block access to care for patients who fall into more severe RUGs classifications, potentially reducing care options and increasing the risk of negative outcomes for frail elderly. The researchers 1) examined the effect of the SNF PPS on patient-level indicators, including access to SNFs, utilization of costly care (including pharmaceutical therapies whose costs go above the per diem cap) and re-hospitalization during high acuity episodes; and 2) examinedg the effect of the SNF PPS on facility-level indicators, such as case-mix, changes in SNF staffing and bed availability. The objective of this project was to better inform policymakers about the implications of prospective payment cost-reduction strategies on access to and quality of care through skilled nursing facilities.
 

Title: Evaluating the Impact of the Outpatient Prospective Payment System on Hospitals’ Medicare Volume and Outpatient Care
Grantee Institution: College of William and Mary
Principal Investigator: Jennifer Mellor, Ph.D. 
Grant Period: February 15, 2011-February 14, 2012

Using data from a Florida database on ambulatory and inpatient discharge data from 1997 to 2008, together with data from Medstat’s MarketScan database and the American Hospital Association, the researchers will study the effects of the Medicare Outpatient Prospective Payment System (OPPS) on payer mix and site of care. Focusing on surgical procedures, they will estimate two regressions. First, to examine the impact of OPPS on payer mix, they will regress the volume of payer-hospital-quarter specific outpatient procedures on Medicare's share of total outpatient volume interacted with the Medicare reimbursement rate, controlling for county and hospital characteristics. Second, to examine the impact of OPPS on inpatient care, they will regress the volume of procedure-hospital-quarter specific inpatient Medicare discharges on a variable that measures the extent of the hospital’s “exposure” to the Medicare shift to OPPS. The goal of this project is to determine whether the payment system led hospitals to (1) substitute from Medicare patients toward those with other coverage, and (2) substitute from outpatient toward inpatient care for conditions that can be treated in either setting.
 

1. Kaiser Family Foundation. “Comparison of Medicare Provisions in Deficit Reduction Proposals,” Publication Number 8124, April 2011. Also see: http://www.kff.org/medicare/8124.cfm.
2. Department of Health and Human Services, Office of the Inspector General (2001). Medicare Hospital Prospective Payment System: How DRG Rates are Calculated and Updated (OEI-09-00-00200). Also see: http://oig.hhs.gov/oei/reports/oei-09-00-00200.pdf.
3. Centers for Medicare & Medicaid Services, Overview Acute Inpatient PPS, 2011. Accessed on May 2, 2011. https://www.cms.gov/AcuteInpatientPPS.
4. Department of Health and Human Services  Office of the Inspector General, 2001. 
5. Centers for Medicare & Medicaid Services, Overview of Acute Inpatient PPS, 2011.
6. Centers for Medicare & Medicaid Services, Overview of Prospective Payment Systems. Accessed on May 2, 2011. https://www.cms.gov/ProspMedicareFeeSvcPmtGen/
7. Medicare Payment Advisory Commission. (2003). Report to the Congress: Variation and Innovation in Medicare. Washington, DC; MedPAC. 
8. RAND. “Effects of Medicare’s Prospective Payment System on the Quality of Hospital Care,” RAND Research Highlights, 2006. Also see: http://www.rand.org/content/dam/rand/pubs/research_briefs/2006/RAND_RB4519-1.pdf. 
9. Mayes, R. and Berenson, R. 2006 Medicare Prospective Payment and the Shaping of U.S. Health Care. Baltimore: Johns Hopkins University Press. 
10. Health Reform GPS. “Bundled Payments: Medicare Pilot Program,” Implementation Briefs, 2010. Also See: http://www.healthreformgps.org/resources/bundled-payments-%E2%80%93-medicare-pilot-program/
11. Centers for Medicare & Medicaid Services, Acute Care Demonstration, Fact Sheet, 2009. Also see: http://www.cms.gov/DemoProjectsEvalRpts/downloads/ACEFactSheet.pdf
12. Casale, A. et al. “ProvenCare: A Provider-Driven Pay-for-Performance Program for Acute Episodic Cardiac Surgical Care,” Annals of Surgery, Vol. 246, No. 4, 2007, pp. 613-623. 
13. Department of Health and Human Services, “Administration Implements Affordable Care Act Provision to Improve Care, Lower Costs,” News Release, April 29, 2011. Also see:
http://www.hhs.gov/news/press/2011pres/04/20110429a.html
14. Sloan, F.A., Morrisey, M.A., and J. Valvona. “Case Shifting and the Medicare Prospective Payment System,” American Journal of Public Health, Vol. 78, No. 5, 1988, p. 553-556.
15. Morrisey, M.S., Sloan, F.A., and J. Valvona. “Medicare Prospective Payment and Post-Hospital Transfers to Subacute Care,” Medical Care, Vol. 26, No. 7, 1988, p. 685. 
16. Hadley, J., Zuckerman, S., and Iezzoni, L. “Financial Pressure and Competition: Changes in Hospital Efficiecny and Cost-Shifting Behavior,” Medical Care, Vol. 34, No. 3, 1996, pp. 202-219.
17. Angelelli, J., et al. “Access to Postacute Nursing Home Care Before and After the BBA,” Health Affairs, Vol. 21, No. 5, 2002, pp. 254-264.
18. Centers for Medicare & Medicaid Services, Overview Hospital Outpatient PPS, 2011. 
19. http://www.hcfo.org/grants/impact-profitability-hospital-responses-financial-stress