Evidence Hub

Structural Validity and Internal Consistency of the Brazilian Portuguese Diabetes Eating Problem Survey–revised in Adolescents With Type 1 Diabetes

Cunha MdCdSO, de Araújo MFM, Gonçalves MdS, Moreira TMM, Queiroz MVO

2026

175 participants

Abstract

Abstract Introduction: The Diabetes Eating Problem Survey–Revised (DEPS-R) is a diabetes-specific screening measure for disordered eating. A Brazilian Portuguese version was culturally adapted, but its internal structure had not been independently evaluated from item-level data. Methods The 16-item DEPS-R was completed by 175 adolescents with type 1 diabetes. Item distributions, corrected item–total correlations, Cronbach’s alpha, ordinal alpha, and McDonald’s omega were examined. Polychoric correlations informed tests of previously proposed one- and three-factor models, parallel analysis, Velicer’s minimum average partial test, and exploratory factor analysis. Associations with glycated hemoglobin and body mass index-for-age were prespecified. Results Internal consistency was good (alpha = .835; ordinal alpha = .892; omega = .893). The one-factor model showed inadequate fit (CFI = .941, TLI = .932, RMSEA = .093, SRMR = .112). A published three-factor model fit better, but two factors correlated .914 and SRMR remained .090. Parallel analysis and the minimum average partial test supported two correlated factors; several items cross-loaded. Scores correlated with glycated hemoglobin (rho = .379) and body mass index-for-age (rho = .265). Discussion The total score has adequate internal consistency and clinically coherent external associations, but dimensionality was not unequivocal. The total score may support screening, whereas Brazilian subscales should not be recommended…

Read the paper

Related studies

Go deeper in Evidence Hub

Evidence Hub adds a study data sheet (design, sample size, evidence level, key numbers), an AI summary and a chat grounded in the paper's text.

Open this study Sign up free