Cancer-related fatigue (CRF) is the most common symptom experienced by patients with cancer. Clinically important improvement in the intensity of fatigue in palliative care patients has not been well established. We reviewed the data from 3 clinical trials of fatigue in 194 patients receiving palliative care treatment. Patients completed the Functional Assessment for Chronic Illness Therapy Fatigue (FACIT-F) and Edmonton Symptom Assessment System (ESAS) at baseline and day 8 and their global perception of fatigue improvement (Global benefit score [GBS]: 1 = not beneficial, 7 = greatly important] during day 8. A GBS of 4 or more (moderate improvement, consistently beneficial) was considered a clinically significant improvement. Change scores in the ESAS and FACIT-F from baseline to day 8 were compared to the GBS greater than 4. Receiver-operating characteristic curves were also derived for ESAS and FACIT-F change scores for a GBS greater than 4, greater than 5, and greater than 6. Results showed the mean patient age was 56 (+/-12) years, and 37% were men. A reduction of approximately 10 points in FACIT-F (sensitivity = 73%, specificity = 78%, area under the curve = 0.82) and 4 points in ESAS fatigue (sensitivity = 66%, specificity = 72%, area under the curve = 0.78) score was best able to predict a clinically important improvement (GBS >/= 4). We were able to characterize the relationship between FACIT-F and ESAS scores and patients' global perception of improvement but further studies are needed to validate our findings.