Background: We previously reported an early intervention model of diabetes care, whereby an inpatient diabetes team performed glycemic surveillance and provided proactive management within 24 hours of admission. In the RAPIDS cluster randomized study, early intervention decreased hyperglycemia and hospital-acquired infection (HAI) compared to usual care [Diabetes Care. 2019;42(5):832-40]. We investigated if improvements in ward glucose control and patient outcomes were sustained following the completion of proactive care intervention.

Methods: We performed a 3-month extension observational study on eight medical and surgical wards immediately following the completion of the RAPIDS study, recruiting consecutive inpatients with pre-existing diabetes or new hyperglycemia. Glycemic and clinical outcomes during the three time periods: baseline (usual care), active (proactive care), and extension (usual care) were compared.

Results: We analyzed 1518 inpatients (mean age 71y, 87% type 2 diabetes, 25% insulin-treated) and observed 8610 patient-days. In the intervention wards, during baseline, active and extension periods, patient-day mean glucose were: mean 9.4 (SD 3.3), 9.0 (2.7), and 9.4 (3.1) mmol/L (p<0.001); and proportion of patient-days with mean glucose >15 mmol/L were 7.3%, 3.3%, and 5.7% (p<0.001). HAI rates paralleled hyperglycemia: 6.4% during baseline period, 2.4% active period (p=0.022 vs. baseline), and 5.4% extension period (p=0.056 vs. active). In the control wards there was no change in hyperglycemia or HAI rates.

Conclusion: Following withdrawal of proactive care with early intervention, hyperglycemia and HAI rates increased to baseline levels. Ongoing intervention with specialist diabetes care teams may be required to achieve sustained glycemic and clinical improvements in hospital.


M. Kyi: None. J. Wang: None. P.R. Wraight: None. P.G. Colman: None. S. Fourlanos: Advisory Panel; Self; Medtronic, Mylan, Sanofi. Speaker’s Bureau; Self; AstraZeneca, Novo Nordisk Inc.


Australian Diabetes Society; Sanofi; National Health and Medical Research Council of Australia (APP1093152); Royal Melbourne Hospital (GIA-030-2019, GIA-027-2016)

Readers may use this article as long as the work is properly cited, the use is educational and not for profit, and the work is not altered. More information is available at