A decade of diabetes: keeping children out of hospital.

OBJECTIVES: To document the number of children aged less than 15 years who developed diabetes and were managed within one large health district, and to evaluate the outcome of those children managed without hospital admission at diagnosis. DESIGN: A retrospective study over 1979-88, when a paediatrician and a physician with special interests in childhood diabetes initiated joint clinics. Data collected from the district diabetes register and files of consultants and health visitors specialising in diabetes. SETTING: Referral of children to consultants in Leicestershire (total population 863,000). MAIN OUTCOME MEASURES: The proportion of children managed without hospital admission, comparison of readmission rates and glycated haemoglobin concentrations between children admitted and those not admitted. RESULTS: Over 10 years 236 children aged 10-14 years developed diabetes (annual incidence rate 12.8/100,000 child population (95% confidence interval 11.3 to 14.7)). In total 138 were not admitted to hospital but received supervised management based at home. Admitted children were younger or acidotic or their family doctors did not contact the diabetes team. Duration of admission declined from seven days in 1979-80 to three days in 1987-8. Ninety two were not admitted to hospital during the 10 years for any reason. Significantly fewer children who received management at home were readmitted for reasons related to diabetes than the group treated in hospital (30 (22%) v 40 (41%); p = 0.004). Concentrations of glycated haemoglobin were no different between the two groups. CONCLUSIONS: Children with newly diagnosed diabetes may be safely and effectively managed out of hospital. Domiciliary or community based management depends on the commitment of consultants specialising in diabetes working in close cooperation with general practitioners, specialist nurses in diabetes, and dietitians.