Psychometric validation of the Spanish version of the Expanded Prostate Cancer Index Composite-26

Abstract

Abstract Purpose To assess the validity, reliability, and responsiveness of the Spanish version of the Expanded Prostate cancer Index Composite (EPIC) with 26 items. Methods Multicentric longitudinal study of patients diagnosed with localized or locally advanced prostate cancer (any T, any N, M0) treated with active surveillance, surgery, external radiotherapy, or brachytherapy. The EPIC-50 was administered initially to the cohort (n = 324 patients), until it was replaced in November 2019 by the EPIC-26 (n = 543), in both groups before treatment and 12 months after. We assessed confirmatory factor analysis (CFA), reliability with Cronbach’s alpha coefficient, criterion validity with the intraclass correlation coefficient (ICC), and responsiveness by testing a priori hypotheses on deterioration effect size (ES). Results The CFA confirmed the five-domain structure of the EPIC-26 proposed by the original instrument (comparative fit index = 0.95). The agreement between EPIC-50 (gold standard) and EPIC-26 domains was excellent (ICC > 0.90). Cronbach’s alpha was > 0.7 in almost all domains, and the floor effect was near zero, although ceiling effect was higher than 50% in urinary incontinence and bowel domains. Hypothesized changes between before and 12 months after treatment were confirmed: ES > 0.8 in both urinary incontinence and sexual domains among patients who underwent surgery; and ES ranging 0.44–0.48 for bowel and sexual domains in patients treated with external radiotherapy. Conclusion The Spanish version of the EPIC-26 has demonstrated adequate metric properties, similar to those of the original version, with acceptable goodness-of-fit indices, good criterion validity, reliability, and responsiveness to detect changes after radical prostatectomy or external radiotherapy.