AbstractPurposeTo evaluate the effectiveness and safety of a pharmacist‐managed protocol for transitioning critically ill patients from intravenous (iv) to subcutaneous insulin compared with a provider‐managed process.MethodsThis single‐center, retrospective, observational study included patients admitted to the medical or surgical/trauma intensive care unit who received a continuous infusion of iv insulin from January 2019 to April 2021. Patients were excluded if they were less than 18 years of age, pregnant, incarcerated, or received iv insulin for the diagnosis of diabetic ketoacidosis, hyperglycemic hyperosmolar state, calcium channel blocker or beta blocker overdose, or hypertriglyceridemia. The primary outcome was the percentage of blood glucose (BG) concentrations within the target range of 70–150 mg/dL from 0 to 48 h following transition to subcutaneous insulin. Secondary outcomes included percentage of BG concentrations within goal range following transition at 0–12 h and 12–24 h, incidence of hypo‐ and hyperglycemia, and percentage of patients requiring dose adjustments after initial transition.ResultsA total of 110 unique patients were included with 70 patients in the provider‐managed group and 40 patients in the pharmacist‐managed group. On average, pharmacists transitioned patients to 63% basal insulin based on their 24‐h total day dose of insulin. The pharmacist‐managed group achieved glycemic control in 53% of transitions at 12 h, 40% at 24 h, and 47% from 0 to 48 h, while the provider group achieved glycemic control in 25% of transitions at 12 h, 12% at 24 h, and 18% from 0 to 48 h (p < 0.001 for all time points). As for safety end points, the pharmacist‐managed group demonstrated lower rates of hypoglycemia (p = 0.001), severe hypoglycemia (p = 0.332), hyperglycemia (p < 0.001), and severe hyperglycemia (p < 0.001) compared with the provider‐managed group.ConclusionsPharmacists can effectively and safely transition critically ill patients from iv to subcutaneous insulin utilizing a standardized protocol.