1. What are the characteristics of joint replacement patients (DRGs 209 & 210) served in IRFs and SNFs? How are they similar or different?

   We need to know if there are systematic differences among joint replacement patients served in both settings and take these into account when evaluating outcomes and cost effectiveness as outlined below. See also Question 6.

2. How are the interventions and processes of care for joint replacement patients different in IRFs and SNFs?

   How can we best characterize the differences in the care received in an IRF vs. a SNF? What makes the care in an IRF and a SNF different? What do they do similarly and what do they do differently? Consider all interventions, e.g., nursing services, physician interventions, physical and occupational therapies, nutritional support, medications, passive motion exercises, weight-bearing exercises. Also consider timing, intensity, frequency, and duration of therapies.

3. What specific interventions or combinations of interventions in IRFs and SNFs make the biggest difference in outcomes for joint replacement patients taking into account patient differences?

   The relevant outcomes for this study include: (1) the onset of complications during the course of treatment, (2) change in severity of illness from admission to discharge, (3) discharge destination (including unscheduled discharges to acute care), and (4) change in functional status from admission to discharge. We will consider (5) rehospitalizations or readmissions in the first 3-6 months following discharge from post-acute care.

4. Which joint replacement patients do better in an IRF and which do better in a SNF?

   How can we best characterize the differences between patients who do better in one setting or the other? How can these characterizations assist in developing post-acute placement criteria and in characterizing the patient mix that should be used in defining an IRF pursuant to the 75% rule?

5. What is the relative cost-effectiveness of IRF and SNF care for joint replacement patients?

   For which patients is it more cost-effective to be placed in an IRF and which patients in a SNF taking into account outcomes both at discharge and rehospitalizations during the first 3-6 months following discharge?

6. Are comorbidities among joint replacement patients an adequate indicator of additional medical need during the rehabilitation process? Can a severity-of-illness measure serve as a better indicator of medical need? Are patients with greater medical needs served better in an IRF or a SNF?

   Do comorbidities or a severity of illness indicator have better predictive validity in terms of service utilization, costs, and outcomes?

7. Can we design a more efficient course of rehabilitation interventions for joint replacement patients in IRFs and SNFs to reduce the length of stay and costs?

Inclusion Criteria:

• Are 21 years of age or older,
• Received a hip or knee replacement of any type for any reason, and
• Admitted from any source.

To be included in the study’s database, the above enrollment criteria need to be met. In addition, patients also must:

• Have a minimum rehabilitation length of stay (LOS) of 3 days, and
• Have complete admission and discharge FIM data.

Exclusion Criteria:

• Joint replacements such as shoulder and ankle replacements since the rehabilitation for these patients is quite different. For most study patients, the joint replacement will be their first, but with the aging of the population, revisions are becoming more common. According to the MedPAC-RAND report, 4.8% of IRF and 6.3% of SNF joint replacement patients had a revision.