Psychometric evaluation of the Korean version of the Diabetes Symptom Checklist-Revised (DSC-R) for patients with type 2 diabetes

This study was to elucidate the psychometric properties of the Korean version of the Diabetes Symptom Checklist-Revised (K-DSC-R), which is a patient-reported outcome measure of diabetes symptom burden. A sample of 432 Korean patients with diabetes was recruited from university hospitals. The data were analyzed using exploratory factor analysis (EFA), confirmatory factor analysis (CFA), multitrait/multi-item correlation, Pearson's correlation, t-test, ANOVA, and Cronbach's alpha for construct, item-convergent/discriminant, concurrent, and known-groups validity, and internal consistency reliability. EFA extracted a total of 29 items clustered into 7 subscales from the K-DSC-R. The construct of the seven-subscales was supported by CFA. The scaling success rates of item-convergent validity were 100% for all subscales, and those of item-discriminant validity ranged from 83.3% to 100%. Patients in more-depressed groups and in the HbA1c-uncontrolled group had higher K-DSC-R scores, satisfying the known-groups validity. The subscales of the K-DSC-R were moderately correlated with health-related quality of life, indicative of the established concurrent validity. The Cronbach's alpha of the K-DSC-R was 0.92. The psychometric properties of the K-DSC-R have been established. It is thus appropriate for use with respect to reliability and validity in practice and clinical trials for Korean patients with type 2 diabetes