Research Plan

Joint replacements are on the rise. During the 10-year period, 1994 to 2003, the number of joint replacement patients discharged from acute care hospitals increased 51%, from 241,410 to 364,824 patients. Many of these patients took advantage of one or more types of post-acute, rehab services. These services, most typically, were provided at Inpatient Rehabilitation Facilities (IRFs), Skilled Nursing Facilities (SNF), at home with home health care providers or through outpatient therapy visits. The majority of joint replacement patients received at least some of their rehab care in IRFs and SNFs. For example, in 2003, 121,528 joint replacement patients received rehab services from IRFs while another 106,981 had their rehab needs provided at SNFs. IRFs and SNFs together served approximately 63% of all patients receiving joint replacements in 2003.

Admitting joint replacement patients to IRFs has been controversial for reasons related both to the "medical necessity" for the intensive therapy IRFs provide and to IRFs' compliance with the 75% rule-the principal criterion that defines an IRF. To qualify as an IRF, as distinct from an acute care hospital and therefore exempt from the acute hospital DRG (Diagnostic-Related Group) payment system, 75% of an inpatient rehab facility's patients must map to 13 impairment groups.

At issue here is Group 13 that limits the types of joint replacement patients that can be counted toward an IRF's qualifying 75%. Patients with joint replacements can help an IRF meet the 75% rule only if they (1) "underwent bilateral hip or knee replacement surgery during the acute hospital admission immediately prior to their admission to an IRF," (2) suffer from extreme obesity, or (3) are over the age of 85. In FY 2003, 87% of all IRF joint replacement patients did not meet these criteria, mainly because they had unilateral rather than bilateral replacements.

In developing and refining the 75% rule, CMS (Centers for Medicare and Medicaid Services) appears to assume that certain joint replacement patients may be served better in post-acute settings less-intensive and less expensive than IRFs-in SNFs, at home with home health care, or in outpatient settings. Of these various settings, SNFs are most often considered the best alternative to more intensive IRF-level care. At present, health systems data demonstrate that SNFs provide less intensive and less expensive care for longer lengths of stay. What is not known, however, is whether SNFs and IRFs serve a different mix of joint replacement patients or which rehab alternative produces better outcomes for which patients.

In response to a recent call from CMS to the rehab community to find answers to these questions, researchers from NRH and the Institute for Clinical Outcomes Research (ICOR) in Salt Lake City, UT have launched a 24-month study to investigate the post-acute management of joint replacement Adopting the descriptive moniker of JOINTS (Joint replacement Outcomes in IRFs and Nursing Treatment Sites) the study team focuses on patients who have had hip and knee joints replaced and receive rehabilitation services in IRFs and SNFs.

Fundamental Questions the JOINTS Study Seeks to Answer

Are IRF and SNF Patients Different?

Before embarking on an exploration of differences between IRFs and SNFs, the JOINTS team needs to know whether there are differences in the lower-extremity joint replacement patients served in each of these two post-acute settings. Accurately gauging the severity of illness that characterizes patients receiving services in IRFs and SNFs is essential to ultimately evaluating differences in outcomes and costs. The JOINTS Study will use a sophisticated severity-of-illness measure to enable researchers to characterize severity beyond traditional methods.

How Does Care Provided in IRFs and SNFs Differ?

In examining the differences in outcomes and costs between IRFs and SNFs, researchers need to understand the differences in the care received in these two settings. All interventions and processes of care from basic medical support to individual therapies need to be taken into account and characterized in terms of timing, intensity, frequency, and duration. Without these characterizations, both settings remain black boxes and prudent purchasers, both government and health plans, cannot fully know what it is that they are purchasing. Purchasers and providers alike need to know which clinical activities and interventions make the biggest difference for which patients and in what settings these activities and interventions are most likely to be found. It is not enough to say one setting is more effective than another without stating what it is about that setting that accounts for difference.

Which Patients Do Better in an IRF and Which Patients Do Better in a SNF?

Asking whether joint replacement patients do better in IRFs versus SNFs is the wrong question. The correct question is "Which patients do better in an IRF and which do better in a SNF?" The JOINTS Study starts with the presumption that neither setting has an exclusive franchise in this area. Research may find that one setting or the other does have consistently superior outcomes across all joint replacement patient subgroups but this remains an unknown at this time.

How Do Outcomes Weigh Against Costs to Achieve Them?

Outcomes eventually have to be evaluated relative to costs. SNFs have the a priori advantage here given their lower cost structure. In those instances where IRFs provide superior outcomes, one still has to ask whether the additional costs of IRF-level care are worth the outcome. In many ways, that is a societal valuation: How willing is society as a whole to spend additional dollars to achieve a superior outcome?

The answer to this question may vary with the outcome in question. Certain outcomes have an intrinsic economic component, for example, post-discharge placement in a community versus an institutional setting, or a reduction in post-discharge rehospitalizations. Because of their large economic consequences, the relative cost-effectiveness of these outcomes may be quite different than the relative cost effectiveness of outcomes such as functional status at discharge.

A 20-member Policy Advisory Panel, consisting of payers, consumers, providers, trade groups, professional associations, government agencies, and other researchers, will advise the JOINTS Study research team. The panel was designed to ensure that study findings will ultimately assist policy makers and providers in optimizing post-acute placement and service delivery decisions. The Panel held its first meeting on September 27, 2005 in Washington, DC.

The NRH-ICOR study team is taking a "best-practices" approach that identifies practice patterns associated with best outcomes across different types of patients and facilities. The JOINTS Study has been designed to provide important 'practice-based evidence' needed in shaping post-acute policy and practice related to the rehabilitation care of patients with joint replacements in both SNFs and IRFs. Early study results are scheduled for release in the fall of 2006, followed by more detailed findings in the months thereafter.