Construct validity and test-retest reliability of the Fatigue Severity Scale in people with chronic neck pain.

Abstract

To investigate an appropriate scoring system and unidimensionality using Rasch analysis, discriminant validity, and reliability of the Fatigue Severity Scale (FSS) in people with chronic neck pain. Cross-sectional. Tertiary institution. Patients with chronic neck pain (n=100) and asymptomatic controls (n=40). Not applicable. The FSS. Twenty-six of the 100 participants with chronic neck pain agreed to complete the FSS again within 1 week after the first administration for the assessment of the test-retest reliability. Two items obviously threatening unidimensionality were eliminated, and the 7-item FSS was developed through Rasch analyses. The 7-item FSS demonstrated the appropriateness of its 7-point scale and adequate internal consistency (Rasch-generated reliability, .83-.91). The 7-item FSS had a negligible floor effect (1%) and ceiling effect (2%). The item-person map demonstrated limited distribution of item difficulty in comparison with the distribution of person ability. The chronic neck pain group demonstrated significantly (P<.001) higher scores in the 7-item FSS than the control group, indicating discriminant validity. The 7-item FSS also demonstrated adequate test-retest reliability with a mean interval of 4.1 days (n=26) for each item (quadratic-weighted κ=.83-.94), and as a whole (intraclass correlation coefficient=.95). A 0-to-100 scale table reflecting Rasch scores was developed, and the minimum detectable change was 9.5 in the 0 to 100 scale. The 7-item FSS appears unidimensional and reliable, and can be used quickly in clinical practice to gain a basic understanding of fatigue symptoms in people with chronic neck pain. Further, it is possible to modify the 7-item FSS to enhance discriminant ability within people with chronic neck pain by adding additional items, enlarging the distribution of item difficulty.