Did the Medicare inpatient rehabilitation facility prospective payment system result in changes in relative patient severity and relative resource use?

BACKGROUND: The Centers for Medicare and Medicaid Services implemented a prospective payment system (PPS) in 2002 for care provided by inpatient rehabilitation facilities (IRFs) to Medicare beneficiaries.
OBJECTIVE: We sought to examine changes in the composition of Medicare beneficiaries in IRFs by examining the percentages of patients having worse functional or health status than the average for their payment groups (relative severity) and of patients having greater cost or longer length of stay than the average for their payment groups (relative resource use) before versus after IRF PPS; to examine whether observed changes in relative resource use were expected given predicted changes; and to explore whether these effects varied by IRF Medicare volume.
METHODS: In an observational study of indicators of Medicare beneficiary relative severity and relative resource use, we studied cases paid for by Medicare during 1999 and 2002 having an acute care stay preceding their IRF stay (n = 363,542 in 1999 and 446,002 in 2002).
RESULTS: Similar percentages of cases had longer than expected lengths of stay, greater-than-expected costs per case, and worse-than-expected functional status pre- versus post-IRF PPS. Cases under the IRF PPS had lower predicted probabilities of death 150 days after admission. Although predicted relative resource use remained steady, observed relative resource use decreased after IRF PPS.
CONCLUSIONS: IRF patient composition has not changed meaningfully for Medicare beneficiaries, but patients within payment groups are being provided less care, which could be attributable to the IRF PPS, existing trends in decreasing length of stay, or both.