Development and Psychometric Evaluation of a Fatigability Index for Full-Time Wheelchair Users with Spinal Cord Injury
Published in: Archives of Physical Medicine and Rehabilitation, Volume 99, Issue 9, pages 1827-1839 (September 2018). doi: 10.1016/j.apmr.2018.04.003
Posted on RAND.org on April 02, 2019
To develop and evaluate psychometrically a self-reported instrument assessing physical fatigability (PF) and mental fatigability (MF) in adults with spinal cord injury (SCI).
Peer-support groups at rehabilitation centers, online support groups.
Adults with SCI (NZ464) in the United States.
Main Outcome Measures
The dimensional structure was assessed by confirmatory factor analysis. The relationship between item responses and fatigability was measured with item response theory (graded response model). Reliability was measured with test information functions. Differential item functioning was evaluated with Wald chi-square tests and the weighted area between the curves. Construct validity was assessed using the known groups method.
An 82-item pool was developed from prior qualitative research and consultations with rehabilitation experts. A non-probability sample (NZ464) was used to evaluate the psychometric properties of the PF and MF scales. The item pool was reduced to 75 based on factor loadings and R2. Both scales are primarily unidimensional, despite moderate multidimensionality. There is good discrimination overall: 18 PF items and 26 MF items have high or very high discrimination power (slopes > 1.35). The measurement precision in the theta range -2.0 to 2.5 is the equivalent of 0.94 reliability for PF and 0.91 for MF. For both measures, F statistics P values were significant at P<.01, and means were higher for those with paraplegia vs quadriplegia, and for those with incomplete paraplegia.
The Fatigability Index is the first instrument designed to assess physical and mental fatigability in adults with SCI. The index highlights causes of fatigue and areas requiring immediate intervention. Development of short-forms and further research on representative samples are necessary.
Research conducted by
This report is part of the RAND Corporation External publication series. Many RAND studies are published in peer-reviewed scholarly journals, as chapters in commercial books, or as documents published by other organizations.
The RAND Corporation is a nonprofit institution that helps improve policy and decisionmaking through research and analysis. RAND's publications do not necessarily reflect the opinions of its research clients and sponsors.