Adherence to Guidelines for Inpatient Pharmacologic Management of Type 2 Diabetes in Adults and Glycemic Outcomes

Abstract

Objectives: Diabetes is often poorly managed in hospitals. This study assessed the level of adherence to current Canadian practice guidelines for inpatient pharmacologic management of type 2 diabetes and whether it affected the frequency of hyperglycemia or hypoglycemia. Methods: Retrospectively, we assessed the first 3 days of routine inpatient capillary blood glucose measurement (CBGM) records for hyperglycemia (>8 mmol/L fasting, >10 mmol/L nonfasting) and hypoglycemia (<4 mmol/L) in adults with drug-treated type 2 diabetes admitted to internal medicine without metabolic decompensation or nil per os (NPO) status at 2 hospitals during October through December 2014. Patients, divided according to their admission orders into guideline-adherent versus guideline-nonadherent groups were compared for frequency of hyperglycemia and hypoglycemia. Factors predicting guideline adherence were assessed. Results: Of 150 patients with diabetes who were admitted, 108 met entry criteria. A total of 89 patients received guideline-based care (82%), whereas 19 patients did not (18%). Charlson index and preadmission medications did not predict guideline-based care, but admitting physicians' seniority did (junior, senior resident, attending physician; p=0.05). In the adherent group, 43% of CBGMs were hyperglycemic, versus 64% in the nonadherent group (p=0.01). For hypoglycemia, proportions were 2% versus 1%, respectively (p=0.21). Conclusions: Adherence to guidelines for inpatient type 2 diabetes management is good and may be greater with more training. Hyperglycemia was more common in patients who did not receive guideline-based care. Hypoglycemia was uncommon and did not appear to be more common in the guideline-adherent group, although numbers were small. These results may alleviate physicians' fear that providing adequate insulin to hospitalized patients may cause hypoglycemia.