Home / Resources / Articles / Insulin Pump vs. Multiple Injection: Which Is The More Effective Treatment?

Insulin Pump vs. Multiple Injection: Which Is The More Effective Treatment?

Apr 22, 2017

Proper training key to best results in both groups.

Based on a study released February 8th by the S. Heller Department of Oncology and Metabolism at the University of Sheffield, insulin pump treatment may be slightly more beneficial than standard multiple daily injections when subjects are provided with similarly structured training. Previous research had shown subjects who had training prior to using a pump experienced benefits, but no known study had compared injections to pump treatments when subjects received comparable training.

The purpose of the Relative Effectiveness of Pumps Over MDI and Structured Education (REPOSE) study was to explore the relative effectiveness of both treatment methods on subjects who received equivalent levels of training. The randomized, parallel, open-label controlled trial took place at multiple locations — three secondary care centers in Scotland and 5 centers in England. In total, 317 subjects participated, all of whom had type 1 diabetes, and none of whom had an initial preference for either using pump or injection for treatment. Participants also had to be willing to self-monitor glucose levels, count carbohydrate intake, and self adjust their insulin treatment. Excluded patients included patients who were already on optimal therapy using multiple injections, had a strong desire for insulin pump treatment, had serious diabetes complications, or were unable to speak English. Each participant was assigned to a  “dose adjusted for normal eating” (DAFNE) course for three-pump or three-injection training that occurred for one week with a follow-up at week 6 and additional training as requested. The study focused on two primary outcomes: HbA1c change after two years when baseline was ≥7.5%, and the proportion of subjects achieving a HbA1c ≤7.5%. Secondary outcomes were measured at 6, 12, and 24 months. They include but aren’t limited to: both moderate and severe hypoglycemia, cholesterol levels, body weight, and insulin dose. Interventions included mealtime use of insulin aspart and twice daily background dosing of insulin detemir for multiple injection patients and insulin aspart for patients assigned to the insulin pump.

The study began with 317 participants. After factoring excluded and withdrawn participants, a total of 267 participants made baseline assessment, and after 24 months, 248 participants finished primary outcome data. Primary outcome results for those who lasted 24 months and had a baseline HbA1c of ≥7.5% showed the change decreased by 0.85% and 0.42% for the pump and injection groups respectively with an insignificant mean difference in change from baseline of -0.24% (95% confidence interval −0.53% to 0.05%, P=0.10) in favor of the insulin pump. Of participants who reached an HbA1c of <7.5%, 23.3% were in the multiple daily injection group and 25.0% were in the pump group (odds ratio 1.22, 95% confidence interval 0.62 to 2.39, P=0.57). The primary analysis was repeated at six- and 12-month follow-up visits over 24 months, with the largest change from baseline HbA1c occurring at the 6-month mark, showing a mean adjusted difference of -0.25% (95% confidence interval −0.52% to 0.02%). Both groups saw reductions in average hypoglycemic events in each patient per year from 0.17 events before baseline to 0.10 events during follow-up. While body weight remained consistent in both groups, an increase in HDL, decrease in LDL, and insulin dose decrease was seen in both the multiple injection and pump groups. Overall, the participants in the pump group had statistically significant improvement in treatment satisfaction (P=0.07 at six months, P<0.001 at 12 and 24 months) compared to patients assigned to the multiple injection group.

Although the groups showed no statistically significant difference in HbA1c control, it should be noted that those using a pump may experience more treatment satisfaction and overall quality of life. A large weakness in this study is the population being primarily white, and additional studies using a more diverse population could be beneficial. Researchers also believe that patients who were assigned to the pump group may have had some apprehension to the pump, despite more face-to-face contact time with professionals compared to the multiple injection group, which could account for the insignificant decreases seen in the pump group.

Practice Pearls:

  • When trained properly, reduction in HbA1c can be seen in users of an insulin pump or those who use multiple daily injections.
  • Using an insulin pump may improve satisfaction and quality of life in type 1 diabetes patients when compared to patients using multiple daily injections.
  • Proper insulin treatment adjustment and administration training is a key component in successful glucose control.


“Relative effectiveness of insulin pump treatment over multiple daily injections and structured education during flexible intensive insulin treatment for type 1 diabetes: cluster randomized trial (REPOSE).” BMJ 2017;356:j1285 http://dx.doi.org/10.1136/bmj.j1285

Priscilla Rettman, BS, PharmD Candidate 2017, Philadelphia College of Osteopathic Medicine- GA Campus