Archives of Physical Medicine and Rehabilitation
Volume 88, Issue 11 , Pages 1522-1525, November 2007

The State-of-the-Science: Challenges in Designing Postacute Care Payment Policy

  • Leighton Chan, MD, MPH

      Affiliations

    • Corresponding Author InformationReprint requests to Leighton Chan, MD, MPH, Bldg 10, CRC, Room 1-1469, 10 Center Dr, MSC 1604, Bethesda, MD 20892-1604

Rehabilitation Medicine Department, National Institutes of Health, Bethesda, MD.

Article Outline

Abstract 

Chan L. The state-of-the-science: challenges in designing postacute care payment policy.

This report describes Medicare’s postacute care (PAC) payment systems and their incentives, as well as global changes in capacity, quality, and utilization over time. It assesses the payment systems’ impact on PAC services, referencing relevant works in progress. Suggestions are made for future research.

Key Words: Health policy, Outcome and process assessment (health care), Outcomes research, Rehabilitation

 

POSTACUTE CARE (PAC) IS treatment provided to patients to improve their functional status after they leave an acute care hospital. Postacute care providers include inpatient rehabilitation facilities (IRFs), skilled nursing facilities (SNFs), long-term care hospitals (LTCHs), and home health agencies (HHAs). PAC costs represent 13% of the entire Medicare budget, or about $42 billion in 2005.1

PAC, as is true for health care in general, has been driven not only by the demographics of the elderly but also by the organization of its payment systems. Because Medicare pays the costs for approximately 70% of all PAC patients, its payment systems have been the main influencing force in the development of the PAC environment. In addition, private insurers often tend to follow Medicare’s lead.2

It has been well documented that a payment system’s incentives are among the most powerful forces that determine what type of care is delivered. For instance, when Medicare enacted a prospective payment system (PPS) for acute care in the 1980s, there were profound changes in the way patients were treated. In a landmark study by the RAND Corporation,3 investigators determined that the PPS for acute care hospitals resulted in a dramatic reduction in hospital length of stay (LOS). While these changes were not associated with an increase in mortality, as had been feared, there was a higher likelihood of patients being discharged home in an unstable condition, or discharged to another facility.3 It is likely that this new policy for acute care hospitals contributed to the growth of PAC facilities.

Just as with acute care hospitals, PAC facilities respond to the incentives and disincentives of payment systems. For instance, IRFs were initially paid under regulations defined by the Tax Equity and Fiscal Responsibility Act. This system contained incentives that encouraged facilities to alter LOS and discharge destinations to maximize profits.4, 5 Similar situations are applicable for nearly all care providers.

In the past decade, the payment methods for PAC facilities have undergone a radical transformation. All 4 provider types are now paid prospectively, as mandated by the U.S. Congress.6 Unfortunately, unlike the development of the diagnosis-related group based PPS for acute care hospitals, these new payment systems did not have adequate time for development, refinement, and testing before they were fully implemented.7 Therefore, it is critical that we examine, on an ongoing basis, the impact of these PPSs so that we can identify and correct any inappropriate consequences.

This report will expand on Kaplan’s8 article, which describes the individual payment systems, their incentives—both intended and unintended—as well as global changes in capacity, quality, and utilization over time. I will discuss what we know about the impact of Medicare’s payment systems on PAC services, and I will refer to some relevant works in progress. Finally, I will suggest areas for future research.

Back to Article Outline

State-of-the-Science 

Because all 4 PAC payment systems are new, there is not a great deal of evidence about their impact. Furthermore, because most of the changes are recent, any assessment must be considered preliminary. Kaplan’s8 article, however, provides an excellent synopsis of some of our current knowledge. The full impact of PPS for PAC is only just now becoming evident, however, and no published report will reflect some of the swift changes in the industry. For instance, anecdotal reports suggest that there has been a dramatic slowdown in the growth of LTCHs. In 2005, the number of these facilities increased by 28. This was followed, however, by a decrease of 2 such hospitals in 2006. Something similar has occurred with IRFs. Since 2004, there has been about an 8% to 10% drop in the number of IRF beds across the United States in both units and freestanding hospitals (American Medical Rehabilitation Providers Association, unpublished data, January 2007). Thus, the situation is very much in flux. One concern regarding the Medicare Payment Advisory Commission’s (MedPAC) current analysis is the fact that it may miss important findings on a smaller level. For instance, the PPS may affect some categories of patients more than others, and what is the impact of PPS on particular impairment categories? MedPAC does not address these types of questions.

In addition to Kaplan’s8 report, several other studies have been published in this area that are worth noting. In anticipation of the PPS, some investigators analyzed how facilities would fare under the new system. In 2004, Dobrez et al9 compared the expected PPS payments to a single IRF with the actual costs incurred by nearly 800 stroke patients between 1994 and 1998. They found that inflation-adjusted payments under PPS were more than $10,000 less than the costs per patient. In a larger study, Hoffman et al10 examined 1807 patients enrolled in the Traumatic Brain Injury Model Systems from 1998 to 2001. They found that the median cost of inpatient rehabilitation exceeded expected median PPS payments for all classes of patients by 16%. Only 3 of the 14 hospitals in the study received reimbursement under PPS that exceeded the costs for their patients with traumatic brain injury (TBI). This finding is of particular concern given the evidence that suggests that multidisciplinary care is the most effective way to treat TBI.11 There are 2 reasons why these results differ from MedPAC’s report, which suggests healthy IRF profit margins under PPS. First, the studies by Dobrez9 and Hoffman10 and colleagues do not reflect changes that the facilities made to improve their efficiency after PPS was initiated. In addition, MedPAC’s analysis of the health of the inpatient rehabilitation sector is global in nature and its findings may not reflect the actual results for either individual or small groups of providers, or for isolated diagnostic categories.

Speech and language pathologists have also assessed the impact of PPS on IRFs. Frymark and Mullen12 examined that impact on 2631 patients in 96 freestanding IRFs. Using data from the American Speech Language Hearing Association’s national outcomes measurement system, they found a significant reduction in LOS for speech- and language-related admissions after the PPS was enacted. In addition, at discharge, patients were more likely to be at a lower functional level in terms of swallowing, motor speech, and memory despite the fact that there was a significant increase in the intensity of speech-language services.

In a work in progress, Hoffman13 analyzed more than 150,000 TBI patients admitted to trauma centers between 1995 and 2004 to see whether implementation of the PPS had an impact on referrals for inpatient rehabilitation. Using data from the National Trauma Databank, and controlling for demographic characteristics such as race, age, and severity of injury, she found that there was a significant drop in the percentage of patients discharged to IRFs and a concomitant increase in the percentage of patients admitted to SNFs. This amounted to a dramatic decrease in the odds of being admitted to an IRF after the PPS was enacted. The effect of these changes on cost and quality care was unknown.

DeJong et al examined the impact of PPS on stroke rehabilitation in 3 IRFs between 2001 and 2003. This study of 539 patients suggested that the PPS did not “materially reshape stroke rehabilitation case mix, utilization, and outcomes.…”14(pS93) There was a shift, however, in therapy resources, from more severely involved stroke patients to moderately involved patients, which suggests that more severely affected patients may not receive the same intensity of services under the PPS. As DeJong pointed out, the study’s principal limitation was that their data were collected soon after the new payment system started and therefore their results may not reflect long-term consequences.

McCue and Thompson15 examined more than 100 IRFs to determine the impact of PPS. Using data from Medicare’s Healthcare Cost Report Information System, they examined the performance of 146 facilities—120 in PPS and 26 in a non-PPS comparison group. They determined that there was little change in access to IRFs. The overall number of discharges remained stable as facilities adopted the PPS. Although PPS facilities decreased their LOS to maintain a profit margin, they did not dramatically change the number of admissions or their patient case mix.

Yip et al16 examined the impact of PPS on a small group of SNFs in California. They found a shift in the case mix such that stroke patients and orthopedic patients became the predominant types. Of more concern was the fact that after the PPS was enacted, physical therapy services dropped by 30%, or 20 minutes a day. These results were consistent through all payer types, although patients who were in health maintenance organizations (HMOs) were slightly less affected.

Liu et al17 examined the impact of Medicare’s new payment system on HHAs. Using data from the Medicare Current Beneficiary Survey, they found that the system greatly reduced overall utilization. The likelihood of having a home health visit was reduced by approximately 20% to 30%. In addition, for those who were seen, there was a 30% reduction in the number of visits after a hospitalization and a 40% reduction in the number of people enrolling from the community. Liu tested a series of interactive models trying to identify whether certain subgroups were disproportionately affected by these reductions. Liu found little evidence of geographic variations but did find an increase in home health visits among stroke patients.

Schlenker et al18 also examined home health outcomes under prospective payment in 2005, using data from Medicare’s Outcome and Assessment Information Set. They did a regression analysis on a national random sample of 164,810 beneficiaries. As did Liu, Schlenker found a decrease in home health visits, although at a slower rate (16%). Unlike Liu, Schlenker was also able to assess patient outcomes and found that some outcomes improved after PPS, including community discharge, rehospitalization, and emergent care rates. Some outcomes worsened, however, including wound care, incontinence, and other important outcomes.

Few investigators have examined the impact of the so-called “transfer rule,” which attempts to limit the “churning” of patients from 1 type of facility to another. Work by Schoenman and Mueller19 suggests that the transfer rule had little effect on profitability of acute care hospitals. The effect on patient access to PAC and other outcomes was unknown.

Implications for Future Research 

As many have stated, the goal in PAC should be to provide the right care to the right patient at the right time and in the right place.20 Ideally, a payment system should provide incentives to accomplish this goal. Before such a system can be designed, however, data are needed on the optimal treatment settings, timing, intensities, etc. Without this data, the system will be designed based on empiric assessments and political or economic concerns, rather than on what has been proven to help patients. Thus, efficacy studies of the individual components of PAC are absolutely essential. Without these data, all we have is our historical behaviors and biases, many of which are rooted in outdated policies. One example is the so-called “3-hour rule” that defines inpatient rehabilitation treatment, despite the fact that in at least 1 study of 934 patients,21 the rule added more than $400,000 in charges, without resulting in discernable benefits.

To a certain extent, the current structure of the 4 PAC providers is a significant and inhibiting barrier to further understanding how the payment systems should be structured.20 There are 4 different classes of providers with 4 separate payment systems and 4 separate measurement tools. This makes it difficult, if not impossible, to accurately compare patients and outcomes from 1 setting to the next. Add to this the fact that there are significant financial interests for whom research in this area creates vulnerabilities and you have the recipe for the current situation: we lack confidence in our payment systems because we do not really know where to place our incentives. We do not have enough information on efficacy, much less on effectiveness, to identify the treatments that we should be encouraging and for which types of patients. Thus, we must find ways to break down the barriers between provider types to enhance research in this area.

Given that the Medicare program has not produced the data needed to perfect its own payment systems, perhaps we can learn from the experience of other health care systems that are not as balkanized. Several international studies should be noted, including those from Italy and New Zealand.22, 23, 24 These studies are of interest because rehabilitation is considerably different in other countries, with patients often having a much longer LOS; the studies can provide us with data that are not readily obtainable in the United States. Most other countries, however, have such radically different sociodemographics, acute health care, and payment systems as to make comparisons with the United States difficult.

Closer to home, there are some organizations from whose experience we can learn. For instance, there may be much to be learned from the experience of a large HMO such as Kaiser Permanente of California.25, 26, 27 Kaiser has a much different stroke care model than does the rest of the United States. It relies much more heavily on home health care and SNFs than inpatient rehabilitation or long-term care hospitalization. Although some research has examined the processes by which Kaiser does its PAC planning, Kaiser currently does not collect functional outcomes information throughout the PAC continuum. These data would be quite instructive inasmuch as it might be possible to compare Kaiser’s results to those experienced by Medicare beneficiaries in other parts of the country.

The Veterans Health Administration (VHA) medical system is another very large provider of PAC. Each year, thousands of veterans receive PAC services in several VHA facilities. Hoenig28 did a systematic review of the VHA rehabilitation literature regarding “cerebro-vascular accident, hip fracture, amputation, and geriatrics.” She found significant variations in PAC treatments and outcomes among VHA facilities that may be related to the PAC structure.

Back to Article Outline

Conclusions 

Although we have some general idea about the impact of the PPSs on PAC services, what is more remarkable is how much we do not know. We do not know, for example, how deep the recent decreases in the numbers of IRFs and LTCHs will be. We do not know with any specificity what the impact of the payment systems will be on access and quality. For instance, are patients with TBI or spinal cord injury getting better or worse treatment under PPS? Are there racial, ethnic, or sociodemographic barriers to receiving adequate PAC? If so, how do they affect outcomes? What is clear is that we must conduct randomized trials that test the individual components of PAC to determine the optimal intensity, duration, and frequency of interventions. Without this information, we lack the knowledge with which to align the incentives of our payment systems in an appropriate fashion.

Back to Article Outline

References 

  1. Medicare Payment Advisory Commission. A data book: healthcare spending and the Medicare program. Washington (DC): MedPAC; 2006;June §9. Available at: http://www.medpac.gov/documents/Jun06DataBook_Entire_report.pdf. Accessed June 20, 2007.
  2. Thompson JM, McCue MJ. Organizational and market factors associated with Medicare dependence in inpatient rehabilitation hospitals. Health Serv Manage Res. 2004;17(1):13–23
  3. Rogers WH, Draper D, Kahn KL, et al. Quality of care before and after implementation of the DRG-based prospective payment system (A summary of effects). JAMA. 1990;264:1989–1994
  4. Chan L, Koepsell TD, Deyo RA, et al. The effect of Medicare’s payment system for rehabilitation hospitals on length of stay, charges, and total payments. N Engl J Med. 1997;337:978–985
  5. Chan L, Ciol M. Medicare’s payment system: its effect on discharges to skilled nursing facilities from rehabilitation hospitals. Arch Phys Med Rehabil. 2000;81:715–719
  6. Gage B. Impact of the BBA on post-acute utilization. Health Care Financ Rev. 1999;20(4):103–126
  7. Newhouse J. Medicare’s challenges in paying providers. Health Care Financ Rev 2005-. 2006;27(2):35–44
  8. Kaplan SJ. Growth and payment adequacy of Medicare postacute care rehabilitation. Arch Phys Med Rehabil. 2007;88:1494–1499
  9. Dobrez DG, Lo Sasso AT, Heinemann AW. The effect of prospective payment on rehabilitative care. Arch Phys Med Rehabil. 2004;85:1909–1914
  10. Hoffman JM, Doctor JN, Chan L, Whyte J, Jha A, Dikmen S. Potential impact of the new Medicare prospective payment system on reimbursement for traumatic brain injury inpatient rehabilitation. Arch Phys Med Rehabil. 2003;84:1165–1172
  11. Salazar AM, Warden DL, Schwab K, et al. Cognitive rehabilitation for traumatic brain injury: a randomized trial. Defense and Veterans Head Injury Program (DVHIP) Study Group. JAMA. 2000;283:3075–3081
  12. Frymark TB, Mullen RC. Influence of the prospective payment system on speech-language pathology services. Am J Phys Med Rehabil. 2005;84:12–21
  13. Hoffman J. Change in inpatient rehabilitation admissions for individuals with TBI after implementation of the Medicare IRF-PPS. 2007;Presented to: State of the Science Symposium on Post-Acute Rehabilitation: Setting a Research Agenda and Developing an Evidence Base for Practice and Public Policy; Feb 12; Crystal City (VA).
  14. DeJong G, Horn SD, Smout RJ, Ryser DK. The early impact of the inpatient rehabilitation facility prospective payment system on stroke rehabilitation case mix, practice patterns, and outcomes. Arch Phys Med Rehabil. 2005;86(12 Suppl 2):S93–S100
  15. McCue MJ, Thompson JM. Early effects of the prospective payment system on inpatient rehabilitation hospital performance. Arch Phys Med Rehabil. 2006;87:198–202
  16. Yip JY, Wilber KH, Myrtle RC. The impact of the 1997 Balanced Budget Amendment’s prospective payment system on patient case mix and rehabilitation utilization in skilled nursing. Gerontologist. 2002;42:653–660
  17. Liu K, Long SK, Dowling K. Medicare interim payment system’s impact on Medicare home health utilization. Health Care Financ Rev. 2003;25(1):81–97
  18. Schlenker RE, Powell MC, Goodrich GK. Initial home health outcomes under prospective payment. Health Serv Res. 2005;40(1):177–193
  19. Schoenman JA, Mueller CD. Rural implications of Medicare’s post-acute-care transfer payment policy. J Rural Health. 2005;21:122–130
  20. Weinrich M, Stuart M, Hoyer T. Rules for rehabilitation: an agenda for research. Neurorehabil Neural Repair. 2005;19:72–83
  21. Johnston MV, Miller LS. Cost-effectiveness of the Medicare three-hour regulation. Arch Phys Med Rehabil. 1986;67:581–585
  22. Saitto C, Marino C, Fusco D, Arcà M, Perucci CA Prospective Payment in Rehabilitation Collaborative Group. Toward a new payment system for inpatient rehabilitation. Part I: Predicting resource consumption. Med Care. 2005;43:844–855
  23. Saitto C, Marino C, Fusco D, Arcà M, Perucci CA Prospective Payment in Rehabilitation Collaborative Group. Toward a new payment system for inpatient rehabilitation. Part II: Reimbursing providers. Med Care. 2005;43:856–864
  24. McNaughton H, DeJong G, Smout RJ, Melvin JL, Brandstater M. A comparison of stroke rehabilitation practice and outcomes between New Zealand and United States facilities. Arch Phys Med Rehabil. 2005;86(12 Suppl 2):S115–S120
  25. Sidney S, Quesenberry C, Schlegel J, Bernstein A, Cho J, Klingman J. The cost of stroke care in a prepaid medical care program and its relationship to functional status. Oakland: Kaiser Permanete, Division of Research; 2000;
  26. Brown D, McCarthy A, Whitney M, Sandel E. FIM study: standardizing functional independence measures for rehabilitation across the continuum. Oakland: Kaiser Permanente; 2000;
  27. Sandel ME, Wang H, Chan L, Terdiman J. Disparities in post-acute stroke rehabilitation: result of an integrated health system study in North California. 2007;Presented to: State of the Science Symposium on Post-Acute Rehabilitation: Setting a Research Agenda and Developing an Evidence Base for Practice and Public Policy; Feb 12; Crystal City (VA).
  28. Hoenig H. Service structure, process, effects on outcomes. 2007;Presented to: State of the Science Symposium on Post-Acute Rehabilitation: Setting a Research Agenda and Developing an Evidence Base for Practice and Public Policy; Feb 12; Crystal City (VA).

 No commercial party having a direct financial interest in the results of the research supporting this article has or will confer a benefit upon the author(s) or upon any organization with which the author(s) is/are associated.

PII: S0003-9993(07)00445-5

doi:10.1016/j.apmr.2007.05.032

Archives of Physical Medicine and Rehabilitation
Volume 88, Issue 11 , Pages 1522-1525, November 2007

Article Tools