Objective: Inpatient protocols for glycemic control

Research Design and Methods: This retrospective observational study analyzed the adult admissions to a Teaching Hospital from October 2010 to February 2013. The association between the protocol using insulin Glargine/Lispro vs NPH/regular and rates of hypoglycemia (glucose ≤ 70 mg/dl), severe hypoglycemia (≤ 550 mg/dl), severe hyperglycemia (≥ 300 mg/dl), and average glucose per patient-day were evaluated. The effect on these outcomes of being hospitalized on the week/weekend and the personnel shift (7-3, 3-11, 11-7) were also analyzed.

Results: The rates of hypoglycemia (severe or not) were lower in the analog insulin protocol after multivariate logistic modeling. The rate of hyperglycemia was higher on the analog insulin protocol, but this association disappeared after multivariate modeling. The outcomes were not affected by the weekend, but the rate of hypoglycemia was lower in the 3-11 shift, and the rate of hyperglycemia was lower in the 11-7 shift.