Latent Class Factor Analysis of the Functional Independence Measure confirmed four distinct dimensions in patients undergoing neurological rehabilitation.

AIM: The FIM instrument is broadly applied in varying rehabilitation services for outcome assessment. Thus, it is important to examine its applicability for services which may differ from the situations and patients for which it was originally developed. The aims of the present study were to examine 1) whether the four FIM dimensions "activities of daily living", "sphincter management", "mobility", and "executive function" can be retrieved by Latent Class Factor Analysis (LCFA); and 2) whether the four dimensions show floor effects in patients with acquired brain injuries undergoing intensive post-acute rehabilitation.

METHODS: We analyzed the FIM data of 269 patients with acquired brain injuries undergoing intensive post-acute rehabilitation. To examine the dimensional structure of the FIM instrument we carried out LCFA. Cronbach's alpha was used to measure the internal consistency. We examined the distribution of the dimension scores to identify floor effects.

RESULTS: LCFA confirmed the postulated four dimensions. The explained variance of items assigned to the four dimensions ranged from 46% to 89%. Cronbach's alpha coefficients of the four subscales ranged from 0.94 to 0.96. The percentage of patients scoring the minimum possible score in each of the retrieved dimensions ranged from 22.3% to 47.9%.

CONCLUSION: When applying the FIM instrument to patients undergoing intensive neurological rehabilitation its dimensionality should be kept in mind. For some patients this outcome measure might not be discriminative enough due to floor effects.