Select Page

Table 17.15 Characteristics of the Fatigue Severity Scale

 
Criterion Evidence
Reliability

Test-Retest: Test retest reliability of the scale indicated no significant differences on the FSS scores from time one to time two. Patients were tested at 2 time periods separated by 5 to 33 weeks (Krupp et al., 1989). When tested with a group of patients who had been diagnosed with Hep C, the ICC scores were 0.82 (Taylor et al., 2000). ICC values for a Turkish study were found to be 0.81 (Armutlu et al., 2007) (multiple sclerosis). The scale has been found to have good test-retest reliability (Dittner et al., 2004).

Internal Consistency: Cronbach alpha scores for the FSS were 0.81 for an MS population and .88 for a normal healthy population. (Krupp et al., 1989). Cronbach score for those with Hep C was 0.94 and the CC was 0.82 (Taylor et al., 2000). Paired t-tests were completed by looking at the scores from the screening test and the baseline tests, but no difference was found (mean difference -0.03, t=0.95, p=0.34) (Taylor et al., 2000). Armutlu et al. (2007) found Cronbach scores ranged from 0.8899 to 0.9401. (Ziino & Ponsford, 2005) found good internal consistency when the scale was used with a group of TBI patients (Cronbach score .90 with item total correlation ranging from 0.37 to 0.84). Overall the scale has been found to have high internal consistency (Dittner et al., 2004). It has been suggested that the scale could be shortened as there appears to be a high level of redundancy with in the scale ((Amtmann et al., 2012) (multiple sclerosis). In a study with polio patients, Cronbach score was greater than 0.95, with item to total correlation ranging from 0.68 to 0.88 (Burger et al., 2010) (polio). In a review by Tyson and Brown (2014), the authors rated the internal consistency of the FSS specifically used among ABI population as excellent.

Validity

Concurrent Validity: FSS scores have been found to be highly correlated with both VAS scores (r=-0.76) and the Medical Outcomes Survey Short Form (MOS SF-36 ; r=-0.76) (Taylor et al., 2000). Ziino and Ponsford (2005) have found the FSS, VAS-f subscales and COF subscales were all significantly correlated. Between the COF–ME and the COF-PE, a strong positive correlation was found (r=0.56 for each measure). For the VAS-F (Vigour and Fatigue), lower vigor scores were associated with higher fatigue scores (Ziino & Ponsford, 2005). The FSS has demonstrated weak concurrent validity for disability when used to assess an ABI population (Tyson & Brown, 2014).

Construct Validity: (LaChapelle and Finlayson (1998); ABI) noted negative correlations between time since injury and the FSS (r=-0.42, p<0.001), as well as between the impact of fatigue on cognitive and physical functioning (r=-0.41, p<0.001; r=-0.48, p<0.48 respectively). Amtmann et al. (2012) found a high correlation with both the subscales of the Modified Fatigue Impact Scale (MFIS) and the MFIS total score in a study that included only MS patients. The FSS had the highest correlation with the MFIS-physical subscale (p=0.77) and the lowest correlation with the MFIS Cognitive (p=0.55)

Predictive Validity: The scale has been shown to discriminate between fatigued and non-fatigued patients (Friedman et al., 2010; Krupp et al., 1989; LaChapelle & Finlayson, 1998; Taylor et al., 2000). Burger et al. (2010) found only a moderate correlation between the 3 VAS scores (daily life, self-care, and household and occupation) and the FSS scores, possible because the FSS measures only physical symptoms of fatigue compared to the VAS.

Responsiveness The FSS has been found to be sensitive to change with time and treatment (Dittner et al., 2004). When compared to the MFIS, the FSS had floor to ceiling responses ranging from 0.9 to 6.8, while the MFIS had a range of 1.1 to 0.7 (Amtmann et al., 2012).
Tested for ABI/TBI patients?* Yes
Other Formats Currently there are two more versions of the FSS: FSS-7, FSS-5. The scale has been translated into Australian English, Canadian English, French, Canadian French, German, Swill, New Zealand English, UK English, Mexican Spanish, Spanish, and Taiwanese ((Kleinman et al., 2000) (chronic hepatitis C).
Use by Proxy? No