Characteristics of the Functional Independence Measure in traumatic spinal cord injury.

OBJECTIVES: The characteristics of the Functional Independence Measure (FIM) were examined for spinal cord injury (SCI) in regard to norms over time by level and completeness of injury, differential benefit of motor and cognition subscales, and "ceiling effect" after rehabilitation discharge.

DESIGN: Descriptive study of raw FIM data collected prospectively at admission and discharge from acute inpatient rehabilitation, and at 1, 2, and 5 years after injury.

SETTING: National Database of the 18 Spinal Cord Injury Model Systems.

SUBJECTS: Persons with SCI, age 16 and over, with functionally complete injuries at inpatient rehabilitation admission (ASIA grades A, B, or C), admitted to Model System an average of 8 days after injury (standard deviation = 13, median = 1 day). Maximum sample sizes for which data were available were: at rehabilitation admission, 3,971 cases; at discharge, 4,033; at year 1 postinjury, 903; 2 years, 712; and 5 years, 570.

OUTCOME MEASURES: The FIM motor and cognition subscales.

RESULTS: There is a substantial ceiling effect of the FIM cognition items even by inpatient rehabilitation discharge, ie, 80% to 90% of the cases average 6 to 7 (independent or modified independence) across the 5 FIM cognition items. At 1 year 89% to 97% of cases were rated independent. FIM motor items were consistent with level of injury and neurologic status. Motor items (excluding locomotion items) were highly intercorrelated (correlations range from .58 to .92 for self care, sphincter control, and mobility items). Trends over years 1, 2, and 5 were stable for both motor and cognition subscales. FIM motor gains were greatest between admission and discharge and gains continued through 1 year after injury, but at a much-decreased rate.

CONCLUSIONS: The cognition items are not informative for detecting changes over time in SCI; at best, these items could serve as a crude cognition screening assessment. Motor items, in contrast, appear to reflect well the functional status of individuals. High correlations among several of the motor items suggest item redundancy. FIM motor scores illustrated the improvements in neurologic and ASIA scores in appropriate cases. Individuals with ASIA impairment grades of B or C at admission make the most gains in FIM motor scores.