A comprehensive study of psychometric properties of the Edmonton Symptom Assessment System (ESAS) in Spanish advanced cancer patients

Background: The Edmonton Symptom Assessment System (ESAS) is developed for daily symptom assessment. Validation studies tested a variety of languages and patients. The purpose was to carry out a comprehensive examination of the psychometric properties of the ESAS through validation of the version in Spanish advanced cancer patients. Method: A reverse translation method was used to translate the ESAS. Previous studies find appropriate Spanish terms to explore, with verbal scales, fatigue, depression and anxiety. Psychometric aspects evaluated were reliability, validity, responsiveness and utility. Results: 171 advanced cancer patients participated. Internal consistency with Cronbach's Alpha was 0.75. In test-retest (0-6 h), Spearma's correlation was between 0.65 and 0.94. Factor analysis found 3 central domains: 'soft' and 'hard physical' and 'emotional'. Concurrent validity with the Rotterdam Symptom Check List (RSCL) found good correlation in physical symptoms (Kappa until 0.66) but weak correlation in emotional symptoms (Kappa 0.35). Discriminant validity (Spearman) found significant differences (p < 0.001) classifying by Karnofsky. ESAS discriminate between inpatients and outpatients (Mann-Whitney, p < 0.001). Responsiveness was tested with ESAS at 0-48 h (Wilcoxon test, p < 0.05). Average time to complete the instrument was 5.5 min. Conclusion: ESAS is a valid, reliable, responsive and feasible instrument with adequate psychometric properties when tested on Spanish advanced cancer patients. © 2011 Elsevier Ltd. All rights reserved