Comparison of SF-36 vitality scale and Fatigue Symptom Inventory in assessing cancer-related fatigue

Purpose Cancer-related fatigue (CRF) is an important symptom in clinical practice and research. The best way to measure it, however, remains unsettled. The SF-36 vitality scale, a general measure of energy/fatigue, is a frequently cited measure. With only four items, however, its ability to adequately represent multiple CRF facets has been questioned. The 13-item Fatigue Symptom Inventory (FSI) was developed to assess multidimensional aspects of CRF. Our objectives were to assess the convergent validity and to compare the sensitivity to change of the two scales. Methods We administered both scales at 1 month (n=68) and 6 months (n=96) to a subset of heterogeneous patients receiving treatment in 16 cancer centers who were enrolled in a clinical trial of pain and depression. Distributions of standardized response means (SRMs) were compared to assess sensitivity to change. Results of both scales were compared to scores on a single fatigue item from the Patient Health Questionnaire (PHQ). Results Mean scores for both the FSI and the vitality scale demonstrated clinically significant fatigue in the sample. The vitality scale was strongly correlated with all three FSI scales (r=-0.68 to -0.77). The vitality and FSI scales also correlated strongly with the PHQ fatigue item. Moreover, distributions of SRMs for both scales were approximately normal. Conclusions Both the FSI and the vitality scale are supported as valid measures of CRF. Both demonstrated sensitivity to change across a range of effect sizes. The vitality scale may be an excellent choice when brevity is paramount; the FSI may be more appropriate when tapping specific dimensions is warranted.