The internal consistency of PRO fatigue instruments in sarcoidosis: superiority of the PFI over the FAS.

INTRODUCTION: The Fatigue Assessment Scale (FAS) is a 10-item patient reported outcome (PRO) questionnaire that is used to measure fatigue in sarcoidosis. After several months of use, we began to question the reliability of the FAS in our clinic population. Therefore, we administered an additional fatigue PRO, the Patient Reported Outcomes Measurement Information Systems (PROMIS) Fatigue Instrument (PFI). Our hypothesis was that the internal consistency/reliability (Cronbach's alpha) of the PFI would be superior to the FAS in sarcoidosis patients because two of the ten FAS items (items #4 and #10) required reverse scoring (these items were scaled in the opposite direction to the other 8 items).
METHODS: The FAS and PFI were administered during the same clinic visit to consecutive patients in our sarcoidosis clinic. We calculated a) the Cronbach's alpha for a) the FAS; b) the FAS without items #4 and #10; and c) the PFI.
RESULTS: 107 consecutive sarcoidosis patients underwent FAS and PFI testing. The Cronbach's alpha was 0.740, 0.911, and 0.963 for the FAS, FAS with items #4 and #10 removed, and the PFI respectively. In female patients, the Cronbach's alpha of the FAS was 0.663, which is considered as "questionable" in terms of internal consistency.
CONCLUSION: We found that the PFI had "excellent" consistency in our sarcoidosis clinic. The FAS did not demonstrate the same degree of internal consistency. The Cronbach's estimate of the FAS with items #4 and #10 removed was vastly superior to the FAS. These data support our contention that FAS items #4 and #10 detract from the internal consistency of this PRO. They also suggest that the PFI is superior to the FAS in terms of reliability.