Multi-informant path models of the influence of psychosocial and treatment-related variables on adherence and metabolic control in adolescents with type 1 diabetes mellitus.

BACKGROUND: We assessed the associations between metabolic control and adherence and a broad range of adolescent and family characteristics (e.g., gender, family structure), treatment-related variables (e.g., disease duration, treatment modality), and psychosocial factors (e.g., symptoms of depression and anxiety, parental support, self-efficacy) in a nationwide study of Danish adolescents (age 12-17 years) with type 1 diabetes mellitus (T1DM). METHODS: Sixty-four percent of invited families participated by completing a survey and providing a blood sample. Two path models of associations between generic and diabetes-related family factors, adolescent self-efficacy, emotional difficulties, and metabolic control and adherence were tested, one for adolescents and one for caregivers. Demographic variables were included as covariates. RESULTS: Both path models demonstrated a satisfying model fit. In both models, metabolic control was associated with adherence, age, and T1DM duration. In the adolescent model, metabolic control was also related to treatment modality, single-parent household, caregiver non-support, and anxiety, whereas in the caregiver model metabolic control was associated with family conflict and caregiver support. In both models, adherence was related to age, duration, treatment modality, family conflict, caregiver support, family functioning, and emotional difficulties of the adolescent. In the adolescent model, adherence was also related to adolescent self-efficacy, whereas in the caregiver model adherence was associated with adolescent gender and caregiver non-support and support. Adolescent self-efficacy, emotional well-being, and difficulties related to adolescent/caregiver interaction appeared to be particularly important, as indicated by their stronger association with adherence and/or metabolic control. CONCLUSION: The results highlight the value of applying a multi-informant approach to address the psychosocial well-being of adolescents with diabetes in a large national sample. Self-efficacy, emotional, and family-related difficulties are important aspects to address in both clinical care and future research regarding adolescents with T1DM.

A Danish version of self-efficacy in diabetes self-management: A valid and reliable questionnaire affected by age and sex.

BACKGROUND: Managing the chronic illness type 1 diabetes (T1D) is extremely demanding, especially during adolescence. Self-efficacy is belief in one's own capabilities and this is crucial for diabetes management. Having a valid method for measuring self-efficacy is important. OBJECTIVE: Our aims were to psychometrically validate a Danish version of the self-efficacy in diabetes management (SEDM) questionnaire, and to examine the relationship between background variables and self-efficacy. METHODS: All Danish adolescents with T1D (n = 1075) were invited to participate in our study. In total, 689 agreed to participate and 602 completed the study. Data were collected using a web-based survey. All participants were asked to provide a blood sample for HbA1c measurement. Graphical log-linear Rasch modeling (GLLRM) was used to validate the questionnaire and its reliability was assessed using Monte Carlo simulation. RESULTS: We found the questionnaire to be valid and reliable, but it had a dual structure that suggested a need for 2 separate subscales. One subscale related to practical (SEDM1) and the other to emotional (SEDM2) aspects of diabetes management. Both subscales were targeted toward adolescents with lower self-efficacy and were associated with HbA1c. SEDM1 was influenced by treatment modality and age. In SEDM2 we found an interaction between age and sex. CONCLUSION: The Danish version of the SEDM questionnaire should be divided into two parts, each with a valid and reliable subscale for self-efficacy measurement. The relationship between self-efficacy and age seems to differ between boys and girls.