Background

Point prevalence studies of diabetes were performed at Royal Melbourne Hospital (RMH) in 1996, 2000, and 2008. These documented a hospital population enriched for diabetes, with prevalence among inpatients of 15, 23, and 27% respectively.^1,2,3

Aims

To determine the current prevalence of diabetes among RMH inpatients as well as the admission indication, length of stay, and glycaemia in diabetes inpatients.

Methods

A point prevalence study of RMH inpatients (n=310) was conducted on May 2, 2012 (ED and ICU excluded). Demographic and anthropometric data were obtained by patient interview and measurement, admission diagnosis and classification of diabetes status from documentation in the clinical record, and capillary blood glucose level (BGL) data from the current observation chart.

Results

The mean age was 63.4 years (58% male). Diabetes was documented in 93 (30%) inpatients. This was significantly higher than previous RMH surveys, and double the prevalence reported in 1996 (p<0.001) (Figure 1). Mean BMI was 27.5 kg/m² in non-diabetes versus 30.0 kg/m² in diabetes inpatients with 83 (89%) having type 2 diabetes.  Of the diabetes patients, 36 (39%) were admitted for a problem directly related to diabetes (acute coronary syndrome, stroke, diabetic foot complication, or heart failure secondary to known/presumed ischaemic cardiomyopathy). Fifty (54%) diabetes inpatients were on insulin therapy and 16 (32%) of these had documented hypoglycaemia (BGL < 3.5mmol/L). Median length of stay was 9 days for patients without and 11 days for patients with diabetes, p=0.028 (15 days for insulin-treated and 7 days for non-insulin-treated diabetes).

Conclusion

One in three hospital inpatients had diabetes, with prevalence doubling in the last 15 years. Inpatients with insulin-requiring diabetes have greater length of stay and are at risk of hypoglycaemia. Recognition of the increasing frequency of diabetes in inpatients has major implications for hospital service and care planning, locally and nationally.