Dimensionality, psychometric properties, and population-based norms of the Turkish version of the Chalder Fatigue Scale among adults

Abstract

Abstract Background Fatigue is emerging as a major public health problem that is highly associated with poor health-related quality of life and disability. Among adults, fatigue has become increasingly common because of workload or lifestyle changes. This study aimed to cross-culturally adapt the Chalder Fatigue Scale (CFS) into Turkish, to investigate its psychometric properties, and to establish normative data in healthy adults by age and gender. Methods The validity of the CFS was tested with a total sample of 476 healthy adults aged 20–40 years (264 males and 212 females) and test–retest/measurement error analyses were performed with 161 participants (94 males and 67 females). The test–retest reliability was examined using the intraclass correlation coefficient (ICC), and internal consistency was determined using Cronbach's α-coefficient. Predictive validity was assessed using the Receiver Operating Characteristic to validate the cut-off value of the CFS for non-fatigued and fatigued participants. Factor analyses and hypothesis testing were conducted to assess construct validity. Hypothesis testing examined convergent and known-group validity by testing 14 predefined hypotheses. Results The mean (SD) and median (25–75%) CFS scores were 10.7 (4.9) and 11 (7–14) for the total sample (n = 476). The cut-off point for CFS was set at ≥ 12 with a sensitivity of 65.8% and a specificity of 85.9%. The CFS provided evidence of excellent fit of the two-factor structure (CFI = 0.963, RMSEA = 0.06, SRMR = 0.02). There was evidence of strong internal consistency demonstrated by Cronbach's α = 0.863 and good test–retest reliability by ICC = 0.76. Thirteen out of 14 hypotheses (92.9%) were confirmed and the scale showed low to moderate correlation with other measurement instruments (r = 0.31–0.51). Conclusions The CFS has been shown to be a reliable and valid instrument that can be used in various populations for the assessment of fatigue. Level of evidence Level II.