Differential item functioning of the functional independence measure in higher performing neurological patients

Abstract

When comparing outcomes of the Functional Independence Measure (FIM ) between patient groups, item characteristics of the FIM should be consistent across groups. The purpose of this study was to compare item difficulty of the FIM in 3 patient groups with neurological disorders. Patients with stroke (n=295), multiple sclerosis (n=150), and traumatic brain injury (n=88). FIM scores were administered in each group. The FIM consists of a motor domain (13 items) and a cognitive domain (5 items). Rasch rating scale analysis was performed to investigate differences in item difficulty (differential item functioning) between groups. Answering categories of the FIM items were reduced to 3 (from the original 7) because of disordered thresholds and low answering frequencies. Two items of the motor domain ("bladder" and "bowel") did not fit the Rasch model. For 7 out of the 11 fitting motor items, item difficulties were different between groups (i.e. showed differential item functioning). All cognitive items fitted the Rasch model, and 4 out of 5 cognitive items showed differential item functioning. Differential item functioning is present in several items of both the motor and cognitive domain of the FIM. Adjustments for differential item functioning may be required when FIMdata will be compared between groups or will be used in a pooled data analysis.