Assessing Fatigue in Patients Receiving Kidney Replacement Therapy Using PROMIS Computer Adaptive Testing

Abstract

Fatigue is a debilitating symptom for many patients receiving kidney replacement therapy (KRT). Patient-reported outcome measures can help clinicians identify and manage fatigue efficiently. We assessed the measurement characteristics of the Patient Reported Outcome Measurement Information System (PROMIS)-Fatigue Computer Adaptive Test (PROMIS-F CAT) in patients receiving KRT using the previously validated Functional Assessment of Chronic Illness Therapy-Fatigue (FACIT-F) questionnaire. Cross-sectional study. 198 adults treated with dialysis or recipients of a kidney transplant in Toronto, Canada. Demographic data, FACIT-F scores, KRT type. Measurement properties of PROMIS-F CAT T scores. Reliability and test-retest reliability were assessed using standard errors ofmeasurement and intraclass correlation coefficient (ICC), respectively. Construct validity was assessed using correlation and comparisons across predefined groups expected to have different levels of fatigue. Receiver operating characteristic (ROC) curves were used to assess the discrimination of PROMIS-F CAT, with clinically relevant fatigue defined by a FACIT-F score of≤30. Of the 198 participants, 57% were male, the mean±SD age was 57±14 years; 65% had received a kidney transplant. Based onthe FACIT-F score, 47 patients (24%) hadclinically relevant fatigue. PROMIS-F CATandFACIT-F were strongly correlated (ρ=-0.80, P<0.001). PROMIS-F CAT had excellent reliability (>0.90 for 98% of sample), and goodtest-retest reliability (ICC=0.85). TheROC analysis demonstrated outstanding discrimination (area under ROC=0.93 [95%, CI 0.89-0.97]). APROMIS-F CAT cutoff score of≥59 accurately identified most patients with clinically relevant fatigue (sensitivity=0.83; specificity=0.91). A convenience sample of clinically stable patients. FACIT-F items are a part of the PROMIS-F item bank, although there was minimal overlap with only 4 FACIT-F items completed in PROMIS-F CAT. PROMIS-F CAT has robust measurement properties with low question burden to assess fatigue among patients with KRT.