OBJECTIVE: To compare the effectiveness of 2 insulin protocols to treat glucocorticoid-induced hyperglycemia in the nonintensive care hospital setting. METHODS: A randomized, open-label, parallel-arm study was conducted comparing standard recommended care of complete insulin orders (CIO) (i.e., 3-part insulin regimen of long-acting basal [background], rapid-acting bolus [mealtime], and rapid-acting correction factor) to an experimental group following a regimen of Neutral Protamine Hagedorn (NPH) plus CIO (NPH-CIO). The primary outcome was mean blood glucose (BG), and the secondary outcome was percent of BG in target range of 70 to 180 mg/dL. Hypoglycemia was also evaluated. RESULTS: Sixty-one patients completed 2 to 5 consecutive inpatient days (31 CIO; 30 NPH-CIO). Baseline mean BG results were 237.2 +/- 50.2 and 221.9 +/- 35.8 mg/dL (P = .30) in the CIO and NPH-CIO groups, respectively. No significant difference in overall mean BG between the 2 groups was detected; however, a significant difference arose on day 3: mean BG 181.8 +/- 32.6 mg/dL (CIO) versus 157.2 +/- 6.1 mg/dL (NPH-CIO) (P = .03). Moreover, the total daily doses (TDDs) of insulin did not differ: 34.8 +/- 43.0 units (CIO) versus 35.8 +/- 25.0 units (NPH-CIO) (P = .13). Percent of BG in target was 54.6% (CIO) and 62% (NPH-CIO) (P = .24). Incidence of severe hypoglycemia (<50 mg/dL) was the same in both groups (0.1%). CONCLUSION: NPH added to 3-part insulin regimen (CIO) may be an effective way to a combat glucocorticoid-induced hyperglycemia, though further research is needed in a larger population. ABBREVIATIONS: A1C = hemoglobin A1C BG = blood glucose CIO = complete insulin orders DM = diabetes mellitus NPH = neutral protamine Hagedorn NPH-CIO = neutral protamine Hagedorn plus CIO TDD = total daily dose.