In PREDICTIVE 303 (Predictable Results and Experience in Diabetes through Intensification and Control to Target: an International Variability Evaluation), patient-directed insulin dose adjustments of once-daily insulin detemir appeared to be a safe and effective alternative to physician-directed dose adjustments in a primary care setting.

The primary objective of PREDICTIVE 303, stated lead investigator Luigi F Meneghini, MD, Diabetes Research Institute, University of Miami Miller School of Medicine, Florida, United States, was to show that patient self-adjusted dosing using insulin detemir and a simplified algorithm (Group 1) was non-inferior to physician-directed standard-of-care dosing (Group 2). Detemir was started in either group once daily at bedtime as an add-on therapy to any other glucose-lowering regimens, or as a replacement of previous basal insulin in patients with type 2 diabetes. The primary outcome measure was HbA1c reduction from baseline.

Randomization was done at the site level. Patients from sites assigned to Group 1 adjusted their detemir dose every 3 days based on mean fasting blood glucose (FBG) values using the following simplified algorithm: mean FBG <4.4 mmol/L, reduce dose by 3U; FBG between 4.4 and 6.1 mmol/L, no change; FBG >6.1 mmol/L, increase by 3U. Detemir dose for Group 2 patients was adjusted by physicians according to the standard of care.

Mean baseline HbA1c was 8.5%. At 26 weeks, mean HbA1c was 7.9% for patients in Group 1 and 8.0% in patients in Group 2 group (p=0.01 between groups; p<0.0001 vs baseline for both groups). FBG, which was 9.7 mmol/L (178 mg/dL) at baseline, dropped to 7.8 mmol/L (143 mg/dL) in patients in Group 1 and to 8.4 mmol/L (154 mg/dL) in Group 2 patients (p<0.0001 between groups; p<0.0001 versus baseline for both groups). As expected, the reductions in HbA1c observed in the insulin-naïve subjects in Group 1 and Group 2 were substantially greater, with no significant differences between groups (−1.1% vs −1.0%, respectively; p=0.09 between groups; p<0.0001 vs baseline for both groups)

At 26 weeks most patients (88%) remained on once-daily insulin detemir (91% in Group 1, 85% in Group 2). The mean daily insulin detemir dose at 26 weeks was 0.7 and 0.5 U/kg, in Groups 1 and 2, respectively. Among insulin-naïve patients, rates of once-daily insulin detemir dosing were higher (95% in Group 1, 92% in Group 2).

At study end, the overall rates of daytime and major hypoglycemia (event/patient/year) were significantly reduced in both groups versus baseline (p<0.05). Daytime, nocturnal and overall hypoglycemia were significantly lower Group 2 versus Group 1 (p<0.0001). Weight remained constant in Group 1 but dropped from 98.2 kg to 97.9 kg in Group 2.

In summary, Prof. Meneghini said that basal insulin titration was successfully carried out in primary care practices. “Compared with standard-of-care, the 303 Algorithm resulted in better or equal improvement in glycemia with slightly greater incidence of non-major hypoglycemia, and no significant weight gain.”