IMPACT OF STRUCTURED INSULIN ORDER SETS ON INPATIENT HYPOGLYCEMIA AND GLYCEMIC CONTROL

In hospitalized patients, glycemic excursions outside recommended glycemic targets have been associated with increased morbidity and mortality. Despite recommendations to avoid use of correctional insulin alone for managing hyperglycemia, this approach remains common. We performed a quality improvement project aimed at both reducing hypoglycemic events and promoting increased use of basal insulin by updating our insulin order sets to reflect clinical practice guideline recommendations. Brooke Army Medical Center correctional insulin order sets were modified to reflect higher treatment thresholds and targets, and a basal insulin order was added with a recommended weight-based starting dose. Pre- and postintervention analyses were performed. Patients were included if they were prescribed subcutaneous insulin during their hospital stay. The following outcomes were measured: (1) glucose levels, and (2) prescriptions for basal insulin. A significant reduction in hypoglycemia events was noted following the intervention (glucose <70 mg/dL: 9.2% pre-intervention vs. 8.8% postintervention; glucose <55 mg/dL: 4.2% pre-intervention vs. 2.2% postintervention). When excluding patients that were ordered correctional insulin alone but did not receive a dose, an increase in basal insulin use was seen (50% pre-intervention vs. 61% postintervention). Rates and severity of hyperglycemia (glucose >180 mg/dL) remained unchanged. The alteration in insulin order set parameters resulted in a significant reduction in hypoglycemia without significant increases in hyperglycemia. Although basal insulin use increased, optimal dosing recommendations were not often utilized. Further interventions are necessary to reduce hyperglycemia. Abbreviations: CPOE = computerized provider order entry; EMR = electronic medical record; HbA1c = hemoglobin A1c; LOS = length of stay; QI = quality improvement; SSI = sliding scale insulin.