Psychometric properties, factor structure, and German population norms of the multidimensional fatigue inventory (MFI-20)
Adrian Westenberger, Mariel Nöhre, Elmar Brähler, Matthias Morfeld, Martina de Zwaan
Abstract
Objective: The Multidimensional Fatigue Inventory (MFI-20) is commonly used, but its factor structure remains unclear. The MFI-20 consists of five subscales (general fatigue, physical fatigue, reduced activity, reduced motivation, and mental fatigue). This study investigates the psychometric properties, including the factor structure, of a general German population sample and tests group hypotheses on gender and age. Another objective is to provide normative data by gender and age groups. Methods: Using data from a representative German sample (n=2,509), reliability and convergent validity measures, group hypothesis testing, and confirmatory/exploratory factor analyses were conducted. Results: The MFI-20 demonstrated satisfactory internal consistency and showed adequate convergent validity with the SF-36. All subscales of the MFI-20 were significantly correlated (0.71-0.85). Physical fatigue exhibited the highest (0.42) and mental fatigue had the lowest (0.19) correlation with age. Fatigue scores were significantly higher for women and significantly increased with age. A five-factor structure showed poor model fit; using an exploratory factor analysis, a two-factor structure emerged (a general factor and a mental/motivational factor). Conclusion: The MFI-20 is a reliable and valid instrument for measuring fatigue in the general population, but the five-factor structure is not supported. The subscale general fatigue or the MFI-20 total score might measure fatigue sufficiently. The provided norms can be used for further research and individual assessment.