Hospital blood glucose (BG) management is a challenge. We implemented an electronic health record (EHR)-based system at our 1541-bed academic medical center (AMC), alerting when a patient (pt) experiences significant hyperglycemia (hyper; 2 BGs ≥300 mg/dL over 48 hours) or hypoglycemia (hypo; <70 mg/dL). These Excursion Event Alerts (EEAs) fired to the clinician entering orders and provided links to insulin order sets and consult requests. The EEAs were launched on 14 non-ICU wards in mid-2019. We compared BG control during the 3 months prior to launch vs. the most recent 3 months. The control group consisted of 12 non-EEA wards. Our 1° outcome was the Quality Hyperglycemia Score-2 (QHS2), previously developed to assess/compare the overall BG quality on individual wards. QHS2 (scale, 0-100) incorporates 5 elements of BG management, crediting overall control as well as the minimization of significant fluctuations. After the intervention, there was a substantial 13.8% improvement in mean QHS2 (+9.36, p=0.029), comprised of more pt-days in target (+4.3%, p=0.024) and a trend for less pt-days with severe hyper (-2%, p=0.09), with no change in hypoglycemia. In contrast, there was no change in QHS2 on control wards. We conclude that a glycemic alert system was associated with improved BG control at our large AMC, suggesting that a simple EHR-based intervention can favorably impact hospital diabetes management.


L.B. Bak: None. T.M. Fandel: None. J.E. Bozzo: None. M.L. DeWitt: None. P. Chen: None. R.A. Pando: None. L. Ferro: None. R. Kaplan: None. K.C. Hendrickson: None. S.B. Amport: None. L. Sussman: None. S.E. Inzucchi: Advisory Panel; Self; AstraZeneca, Boehringer Ingelheim International GmbH, Lexicon Pharmaceuticals, Inc., Novo Nordisk A/S, Sanofi. Consultant; Self; Abbott, Merck & Co., Inc., vTv Therapeutics.

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