Objective: To investigate the psychometric properties of the Multidimensional Assessment of Fatigue (MAF) scale in a traumatic brain injury (TBI) sample.
Design: Prospective survey study.
Setting: Community.
Participants: One hundred sixty-seven individuals with TBI admitted for inpatient rehabilitation, enrolled into the TBI Model Systems national database, and followed up at either the first or second year postinjury.
Interventions: Not applicable.
Main outcome measure(s): Multidimensional Assessment of Fatigue.
Results: The initial analysis, using items 1 to 14, which are based on a 10-point rating scale, found that only 1 item ("walking") misfit the overall construct of fatigue in this TBI population. However, this 10-point rating scale was found to have disordered thresholds. When ratings were collapsed into 4 response categories, all MAF items used to calculate the Global Fatigue Index formed a unidimensional scale.
Conclusion(s): Findings generally support the unidimensionality of the MAF when used in a TBI population but call into question the use of a 10-point rating scale for items 1 to 14. Further study is needed to investigate the use of a 4-category rating scale across all items and the fit of the "walking" item for a measure of fatigue among individuals with TBI.