Psychometric testing of the Norwegian version of the Expanded Prostate Cancer Index Composite 26-item version (EPIC-26)

Abstract

Objectives: In patients with prostate cancer, evaluation of quality of care requires the inclusion of patient-reported outcomes measures assessed by validated and reliable instruments. Basic psychometric tests of the Norwegian version of the Expanded Prostate Cancer Index Composite with 26 items (EPIC-26) were performed in this study. Material and methods: Translation of the original questionnaire (University of California, Los Angeles Prostate Cancer Index Composite), field testing and retranslation were done according to published guidelines. The final EPIC-26 items were subsequently selected from the Norwegian version of the EPIC-50 with slight verbal adjustments to comply with the English version of the EPIC-26. Reliability and validity were tested among 471 patients who, between 2009 and 2010, had been included in a prospective Norwegian multicenter study assessing adverse effects after radical prostatectomy or prostatic radiotherapy, usually combined with (neo)adjuvant hormone treatment. All patients completed the EPIC-26 before treatment and 3 and 12 months afterwards. Results: Internal consistency was documented by Cronbach’s alpha coefficients ranging from 0.64 to 0.91 for the five domains/subdomains. Item-to-scale correlation coefficients ranged from 0.20 to 0.88, with the lowest value (0.20) for overall sexual problem. Criterion validity was proven by significant correlations between individual responses to the International Prostate Symptom Score and sum scores of the irritative/obstructive subdomain score of the EPIC-26. Satisfactory sensitivity and responsiveness reflected clinical utility for assessing between-group differences and treatment-related changes. Conclusions: Based on basic psychometric tests, the Norwegian version of the EPIC-26 showed acceptable reliability and validity for assessment of adverse effects after treatment of non-metastatic prostate cancer.