The Fatigue Symptom Inventory: a systematic review of its psychometric properties

Abstract

Purpose: Fatigue is one of the most common symptoms experienced by persons with chronic illness, including cancer. The effective management of fatigue hinges in part on the accuracy and reliability of its measurement. The purpose of this study was to review and characterize the use of the 14-item Fatigue Symptom Inventory (FSI) in published studies and to evaluate the available evidence for its psychometric properties.

Methods: A systematic review of the literature identified 55 studies reporting results for the FSI. Data were analyzed to characterize internal consistency reliability of multi-item FSI scales and test-retest reliability. Correlation coefficients were summarized to characterize concurrent, convergent, and divergent validity. Standardized effect sizes were calculated to characterize the discriminative validity of the FSI and its sensitivity to change.

Results: Sample sizes across studies ranged from 9-1,756. Approximately half of the samples consisted exclusively of females. Alpha coefficients for multi-item scales ranged from 0.84-0.96. Most items demonstrated low to moderate test-retest correlations. Correlations with other fatigue measures ranged from 0.41-0.86. Correlations with depression and anxiety measures were positive (range = 0.23-0.76). Correlations with measures of vitality and vigor were negative (range = -0.28 to -0.77). Effect sizes for discriminative validity and sensitivity to change ranged from small to medium and from small to large, respectively.

Conclusions: Findings provide good empirical evidence of the usefulness of the FSI and strongly support its use in future studies.