Recent UK surveys report that the inpatient prevalence of diabetes is 5.5-31.1%1. Uncontrolled hyperglycaemia in hospitalised patients is associated with prolonged length of stay and increased risk of adverse outcomes2. Insulin is one of the top 5 high-risk medications used in hospitals3, accounting for more medication errors than any other drug4. The first aim of our survey was to determine the prevalence of diabetes in a Sydney Teaching Hospital. The second aim was to determine the frequency and type of insulin prescribing and administration errors.
The study was a single-day bedside survey of patient clinical records. An 18-item Inpatient Diabetes Survey Form was completed for all inpatients admitted to St Vincent’s Hospital Sydney on 14th November 2013. Twelve surveying teams, comprising 2 members per team, completed the survey.
We surveyed 394 inpatient records. After exclusion of 8 patients in intensive care with transient hyperglycaemia almost one quarter (23%, n=89) of patients had diabetes (70.8% male, mean age 66 years, 76.4% type 2 diabetes, 10.1% steroid-induced diabetes, 29.2% admitted on insulin). Fourteen (15.7%) had diabetes diagnosed during the current admission. The prevalence of diabetes on the cardiothoracic ward was 45.2% (n=14) and on the haematology/oncology ward was 39.4% (n=13). Of the patients with diabetes, 44.9% (n=40) were treated with insulin during their admission. Of those, 85% had insulin orders written clearly, 95% had the insulin type written correctly, 77.5% had documentation of receiving insulin and 92.5% had insulin prescribed at the correct time.
Nearly one quarter of inpatients had previously-diagnosed or newly-diagnosed diabetes, with almost half of these requiring insulin during admission. This data provides a basis for workforce planning and resource allocation for future inpatient diabetes management. The inpatient hospital admission is a critical time for detecting and treating undiagnosed diabetes. Future surveys will determine whether the prevalence of diabetes amongst the inpatient population is increasing and whether initiatives instituted to improve insulin prescribing and administration are effective.