Background: Diabetes has become a major health problem worldwide, as well as in South Africa. This, coupled with the chronicity of the disease, relate to an increasing burden on health care facilities and an increasing number of hospital admissions of patients suffering from diabetes. Admissions are mostly related to diabetes itself, but the frequency of admissions for problems not related to diabetes is increasing as the prevalence of diabetes increases in the population. Proper inpatient glycaemic management is important for improving patient outcome and for reducing the risk of inpatient complications.

Objectives: The objective of this study was to evaluate current practices in the care of diabetic inpatients as well as to assess the glycaemic control that is achieved during hospitalisation.

Methods: An audit was done of clinical hospital records of adult diabetic patients admitted to Kalafong Hospital, a large secondary hospital in South Africa. All patients admitted who had type 1 or type 2 diabetes before admission, or who were newly diagnosed on admission or in hospital were included, irrespective of the discipline to which the patient was admitted. All patient admissions in the eight-month period preceding the initiation of the audit were included.

Results: The hospital records of 164 diabetic patients were audited. With regard to glucose monitoring, 60.8% of patients had irregular and erratic glucose monitoring, 37.2% had regular (either four- or six-hourly) monitoring and only 2% were monitored in relation to meals. Of the 164 patients, 160 were not fasting, 27 were treated with an insulin sliding scale at some stage during their admission, and in 14 (52%) of the patients who were on sliding scales the scale was used inappropriately. Most hospital inpatients with diabetes, i.e. 48 (30.4%), were treated with oral agents only; 29 (18.4%) were treated with oral agents plus a daily dose of NPH insulin and 17 (10.8%) with mixed insulin twice daily. Only three patients (1.9%) received insulin supplemental to their regimen. The glycaemic control treatment schedule was appropriate in only 19.5% of cases.

Conclusions: Based on our findings, the monitoring and management of blood glucose in patients with diabetes during hospitalisation in a large secondary hospital in South Africa is currently inadequate. This calls for an educational intervention for doctors and nurses working with diabetic inpatients as well as the introduction of a blood glucose management protocol.