Modifications of the FIM instrument under the inpatient rehabilitation facility prospective payment system.

OBJECTIVES: To describe the modifications made to the FIM instrument when it was incorporated into the Inpatient Rehabilitation Facility-Patient Assessment Instrument (IRF-PAI), and to compare FIM data collected before and after the IRF prospective payment system (IRF-PPS) was implemented in 2002 for patients with stroke.
DESIGN: Year-by-year comparison of data of Medicare patients with stroke discharged in 1998-2003 from 411 IRFs that submitted data to the Uniform Data System for Medical Rehabilitation for each of those years.
RESULTS: In the pre-PPS period, admission motor FIM ratings decreased slightly, and discharge motor, admission cognitive, and discharge cognitive ratings remained stable. Between 2001 and 2003, all four ratings decreased: admission motor by 1.8 FIM units, discharge motor by 3.3 FIM units, and admission and discharge cognitive each by 1.0 FIM unit. The lower admission FIM ratings led to an increase in the mean case-mix index from 1.39 to 1.49.
CONCLUSIONS: The decrease in FIM ratings in the IRF-PAI/PPS years may reflect alterations in coding practices as a result of changed rules for rating the FIM instrument, "downcoding" leading to assignment into higher-paying categories, changes in the IRF patient population, and/or changes in IRF patient outcomes. Coding changes should be considered when comparing pre-PPS and PPS FIM data.