Translation, cross-cultural adaptation, and reliability of a depression Scale in Danish adolescents

Abstract Background The Center for Epidemiological Studies Depression Scale for Children (CES-DC) is a widely used 20-item questionnaire for children and adolescents measuring depressive symptoms. However, it has not been formally translated and culturally adapted to Danish. We aimed to translate and examine construct validity in terms of cross-cultural adaptation, reliability in terms of the internal consistency, test-retest reliability in terms of agreement between tests and measurement error, and, finally, to compare the full 20-item version with the abbreviated 4-item-version. Methods We back- and forward translated from the original English version to Danish with pilot tests of the translated version in 14 children and adolescents. In a random sample of 168 adolescents from all over Denmark (15-16-year-old), we conducted a test-retest of the translated CES-DC, and compared the 20-item scale with an abbreviated 4-item scale. Results The translation process revealed minor linguistic problems, while the pilot test revealed difficulties in understanding the wording, especially among the youngest participants. CES-DC showed acceptable test-retest reliability (interclass correlation coefficient (ICC=0.83(95%CI: 0.67;0.91)), but with large standard error of measurement (SEM) 4.93(95% CI: 3.86;6.30), a minimal detectable change of 13.66(95%CI: 10.70;17.46), and a systematic decrease from test to re-test. The 20-item and the four-item versions were highly correlated (r = 0.93). Conclusions Based on the findings in this study, we do not recommend using CES-DC, especially not in children under 13 years, neither for screening nor for evaluative purposes. We recommend testing a rephrasing of the answer categories before a new examination of the psychometric properties. Key messages • Even widely used scales may turn out problematic in new settings, especially among children and adolescents. • Thorough examination of validity and reliability in the target population is crucial.