Psychometric properties of the German version of the Diabetes Eating Problem Survey-Revised: additional benefit of disease-specific screening in adolescents with Type 1 diabetes.

Abstract

To examine the psychometric properties of the German version of the abbreviated 16-item Diabetes Eating Problem Survey-Revised in a sample of young people with Type 1 diabetes. A total of 246 young people, aged 11-19 years, with Type 1 diabetes from six pediatric diabetes centres in Germany were assessed using the Diabetes Eating Problem Survey-Revised. In addition, they underwent screening with two generic tools as well as the WHO five-question well-being index. A clinician's report was also obtained. The Diabetes Eating Problem Survey-Revised was found to have good internal consistency (Cronbach's α = 0.84). The Diabetes Eating Problem Survey-Revised scores significantly correlated with those provided by the non-specific screening tools (r = 0.37, P ≤ 0.000 and r = 0.50, P ≤ 0.000 for boys and r = 0.62, P ≤ 0.000 and r = 0.79, P ≤ 0.000 for girls), indicating convergent validity. The mean (sd) total of the scores was 12.0 (9.6). Criterion validity was confirmed against HbA1c value, BMI standard deviation score and expert (clinician) report. Of the boys included in the study, 11 scored higher than the threshold score on the Diabetes Eating Problem Survey-Revised, of whom only three (27%) were classified as 'suspected to have a disordered eating behaviour' by their clinicians. The Diabetes Eating Problem Survey-Revised delivered more specific information than generic screening instruments and identified more young people with eating disorders than did clinician report, especially regarding the detection of boys at risk. The results of this study support the utility of the German version of the Diabetes Eating Problem Survey-Revised to identify eating disorders in young people with Type 1 diabetes at an early stage. (German Clinical Trials Registry no.: DRKS00004699).