Validation study of the Parkinson's Fatigue Scale in advanced Parkinson's disease

Purpose: To validate the Parkinson's Fatigue Scale (PFS-16) in advanced Parkinson Disease (APD) patients using the scale's Spanish version. Patients and methods: In a clinical study for Levodopa-Carbidopa Intestinal Gel (LCIG), 59 patients were assessed over six months using the PFS-16 and other instruments. The psychometric properties of the PFS-16 were then analyzed. Results: Patients (60.7% men) were aged 68.02±7.43 years. PD duration was 12.57±5.97 years. Median Hoehn and Yahr (HY) stage of patients in "on" was 2 (range: 1-4). There were excellent data quality and acceptability for the PFS-16 as a whole, except for moderate-to-high ceiling effects in its items. Two factors explained 67% of the variance, yet parallel analysis demonstrated the unidimensional nature of the PFS-16, whose internal consistency was satisfactory (Cronbach's alpha=0.93; item homogeneity coefficient=0.19, and item total-corrected correlations=0.50-0.84). PFS-16 total score showed moderate-to-high correlations with fatigue-specific questions within clinical tools, namely item 20 of the Beck Depression Inventory (r_S=0.65) and item 4 of the Non-Motor Symptoms Scale (r_S=0.33). Weak-to-moderate correlations were observed between the PFS-16 and measures of anxiety, depression, apathy, and quality of life. There were no significant differences in PFS-16 total scores when grouped by age, sex, time from diagnosis, HY, and CGI-S. After treatment with LCIG, the relative change in PFS-16 total score was -17.6% and the effect size (Cohen's d) was 0.92. Moderate correlations between changes in the PFS-16 and several other clinical tools were also found. Conclusion: In APD patients, the PFS-16 showed satisfactory acceptability, internal consistency, construct validity, and responsiveness.