Translation and validation of the Chinese version of the Pancreatic Exocrine Insufficiency Questionnaire in patients after pancreatectomy

Vollständiger Abstract

Worum geht es in dieser Arbeit?

Pancreatic exocrine insufficiency (PEI) is common after pancreatic resection, but no Simplified Chinese PEI-specific patient-reported measure has undergone published cross-cultural evaluation. We translated and adapted the Pancreatic Exocrine Insufficiency Questionnaire (PEI-Q) and evaluated selected psychometric properties in patients after pancreatectomy. Following Beaton guidelines, the PEI-Q was translated and cross-culturally adapted. This single-center cross-sectional study included 163 patients. Item analysis, content validity assessment, principal component analysis, principal axis factoring sensitivity analysis, parallel analysis, and confirmatory factor analysis of the original 3-domain structure were performed. Internal consistency was evaluated; test–retest reliability, criterion validity, and responsiveness were not. Overall Cronbach alpha was 0.813 (95% confidence interval: 0.742–0.857); domain alphas were 0.667, 0.719, and 0.603, with 2 domains below 0.70. Total-scale McDonald omega was 0.820. Corrected item-total correlations ranged from −0.006 to 0.680; removing item 14 increased alpha to 0.843. Exploratory analyses retained 3 factors, but confirmatory factor analysis did not support the original structure (comparative fit index, 0.710; Tucker-Lewis index, 0.664; root mean square error of approximation, 0.130; standardized root mean square residual, 0.130). Principal axis factoring confirmed low communality for item 14 (0.013). The bowel movement symptoms domain showed a 25.8% floor effect; no ceiling effects occurred. The Chinese PEI-Q showed strong content validity and acceptable overall internal consistency. However, 2 domains had borderline consistency, and the original structure was not confirmed. It may support symptom monitoring and patient-clinician communication but should not be used for diagnosis or regarded as fully validated until further evaluation.