Factors associated with glycaemic outcome of childhood diabetes care in Denmark.

AIMS

To study how structure and process of care is associated with outcome assessed by HbA(1c).

METHODS

Data for this cross-sectional study originated from the nationwide Danish Registry for Childhood Diabetes and two questionnaires. One questionnaire was sent to all children under 16 years of age with Type 1 diabetes in the year 2000 (N = 1087, response rate 80%). Another questionnaire was sent to the 19 centres in Denmark treating these children (response rate 100%). Simultaneously the children were asked to take a blood sample for central HbA(1c) analysis. Linear mixed models were used for analysis of associations between structure and process indicators and HbA(1c). Age, diabetes duration, sex, ethnicity, family structure and parents' occupational status were included as patient factors possibly affecting HbA(1c).

RESULTS

More visits to the outpatient clinic and higher insulin dosage were significantly associated with higher HbA(1c) (P = 0.002 and P = 0.0001, respectively). Increased frequency of blood glucose monitoring (BGM/week) and completed nephropathy screening were significantly associated with lower HbA(1c) value (estimates -0.008 and -0.49, P = 0.02, respectively). The structure indicators were not associated with HbA(1c), but telephone hot-line was positively associated with the process indicator BGM (estimate 4.02, P = 0.04). Children without Danish parents performed BGM significantly less frequently (-7.11, P = 0.0005) and had higher HbA(1c) (0.41, P = 0.06).

CONCLUSIONS

Most process indicators were significantly associated with HbA(1c), indicating relevant action of staff on glucose regulation. The structure indicators were not associated with outcome, necessitating more detailed studies on the influence of staffing resources, treatment strategies and targets in childhood diabetes management.