Psychometric properties of the German teen and parent versions of the Problem Areas in Diabetes Scale (PAID)

Abstract

The negative impact of psychosocial burden in connection with the treatment of Type 1 diabetes (T1D) indicates the need for regular screening of diabetes distress in adolescents with T1D and their parents. Psychometric properties of the German versions of Problem Areas in Diabetes scale-Teen (PAID-T) and Parent (P-PAID-T) are examined in order to provide a clinical screening tool. Linguistically translated questionnaires were used in a multicenter study with 459 families. Confirmatory factor analysis, validity, and reliability were examined. Teens (42.8% female) had a mean age of 14.7 years. Most parent-caregivers were mothers (74.4%) and were born in Germany (83.1%). Results corroborate the three-factor model for the PAID-T with acceptable model fit, and convergent and discriminant validity was observed. The four-factor model for parents was also supported but had inadequate discriminant validity in this study. Teen and parent scores showed excellent Cronbach's α = 0.91 and 0.93, respectively. The PAID-T and P-PAID-T scores were positively correlated with HbA1c (r_s = .343 and .252, respectively, p < .001) and negatively correlated with treatment satisfaction (Diabetes Treatment Satisfaction Questionnaire) and KIDSCREEN-10 index (teens: r_s = -.545 and -.575; parents: r_s = -.563 and -.489, respectively, all p < .001). The P-PAID-T correlated positively with depressive symptoms measured in Patient Health Questionnaire -9 (r_s = .537, p < .001). The German versions of PAID-T and P-PAID-T produced scores that demonstrated good reliability and validity. Like the original English versions, the German versions are useful to detect diabetes-specific distress in families and to tailor interventions for affected teenagers and their parents. (PsycInfo Database Record (c) 2023 APA, all rights reserved).