Comparison of assessment scores for fatigue between multidimensional fatigue inventory (MFI-K) and modified chalder...

Because of the absence of biological parameters for fatigue, appropriate instruments for assessing the degree of fatigue are important in the diagnosis and management of people complaining of fatigue-like symptoms. This study statistically analyzed the fatigue scores from two typical questionnaire-based instruments: the Korean version of the Multidimensional Fatigue Inventory (MFI-K) and the modified Chalder Fatigue Scale (mKCFQ).

Seventy participants (males n  = 40, females n  = 30, median age 48 years old, range of 25-67) were grouped into three groups ('mild'  = 20, 'moderate'  = 42, and 'severe'  = 8) according to self-reported fatigue levels using a 7-point Likert scale. The similarities and differences between two instrument-derived scores were analyzed using correlations (r) and multidimensional scaling (MDS).

The total scores of the two assessments were significantly correlated (r  = 75%, p  < 0.001), as were the subscores ('Total Physical fatigue': r  = 76%, p  < 0.001, 'Total Mental fatigue': r  = 56%, p  < 0.001). Relative overestimation of the MFI-K (45.8 ± 11.3) compared to the mKCFQ (36.1 ± 16.2) was observed, which was especially prominent in the 'mild' group. The scores of the three groups were more easily distinguished by the mKCFQ than by the MFI-K. In terms of the five dimension scores, we found a higher correlation of the two assessments for 'general fatigue' (r  = 79%, p  < 0.001) and 'physical fatigue' (r  = 66%, p  < 0.001) than for the reductions in 'motivation' (r  = 41%, p  < 0.01) and 'activity' (r  = 26%, p  > 0.05).

Our results may indicate the usefulness of the two instruments, especially for the physical symptoms of fatigue ('general' and 'physical' fatigue). Furthermore, the MFI-K may be useful for conditions of moderate-to-severe fatigue, such as chronic fatigue syndrome, but the mKCFQ may be useful for all spectra of fatigue, including in subhealthy people.