Basal-bolus therapy with insulin detemir using the 303 algorithm in the US PREDICTIVE 303 trial

Introduction: The aim of this study was to compare a simplified patient-driven algorithm (303 Algorithm) to physician-driven adjustments in a subset of 193 patients with type 2 diabetes from the PREDICTIVE 303 study who were using basal-bolus insulin therapy. Methods: PREDICTIVE 303 was a 26-week, randomized, phase 4 study, in which subjects were either instructed to adjust their insulin detemir dose every 3 days by ±3 units if mean fasting plasma glucose (FPG) values were above 110 mg/dL or below 80 mg/dL (303 Algorithm), or had physicians adjust the insulin detemir dose according to usual practice (Standard-of-care). Results: Patients in both groups achieved similar reductions in glycated hemoglobin (-0.2% and -0.3% for 303 Algorithm and Standard-of care groups, respectively; between groups P=0.60). 303 Algorithm group patients achieved a greater reduction in FPG (-21.3 mg/dL vs. 0.2 mg/dL; between groups P=0.03). Both 303 Algorithm and Standard-of care groups experienced a similar rate of overall hypoglycemia, and similar weight reduction (-1.7 kg and -0.4 kg, respectively; between groups P=0.07). Over 82% of patients in both groups used insulin detemir once daily. Conclusions: Adjustments of a once-daily detemir dose by patients using the 303 Algorithm in a basalbolus setting is equally effective in improving glycemic control in patients with type 2 diabetes compared with physician-directed basal dose adjustments.