Introduction: Advances in the last decade in Artificial Pancreas system (APS) have been promising and accelerating with the introduction of Hybrid CGMS/Pump infusion technology as a major step toward long-term Artificial durable pancreas. Also, there exist virtual system termed the Virtual Pancreas System (VPS) that makes treatment effective, and cut cost to achieve the triple aim of optimum HgA1c/glucose variability, averting hypoglycemia or weight gain. The system also adjusts insulin pump setting and recently used the data from CGMS to direct therapy. However, what is the cost-effectiveness of such system compared to Hybrid Artificial Pancreas?

Methods: We compared the features of both systems from the literature review. We wanted to determine how much money ($) is spent on improving HgA1c by 1% in three months, reaching glucose within target (70-170) range, and hypoglycemia rate < 70 for both.

Results: 1.) VPS/CGMS lowered HgA1c by the absolute value of 2.4 in more than 90%. APS reported improvement in HgA1C of 1% recently.The Longest study is three months.

2.) VPS needs 14 glucose tests per month to get the optimum control or may use CGMS. APS needs CGMS.

3.) The diabetes care value index was: $236 per 1% HgA1C drop for VPS/CGMS, APS Value index still pending long-term study but short term it is $4000.

4.) VPS gets 90% glucose within target, APS 76% at this time, hypoglycemia 0 per patient in the VPS (or <0.2%), 1.1 per patient in APS (or <2%).

5.) VPS/CGMS costs $ 600/ year. APS as technology is still expensive. Our estimate it will be $12000/year for the next ten years

6.) VPS technology is available now. APS is still in early stage II or III clinical testing and will need ten years of clinical practice.

Conclusion: Virtual Pancreas System has superior Diabetes Care Value Index compared to any form of artificial pancreas system including the recently approved hybrid model intensive care. VPS will save more than $12000 yearly per patient, and $ 6 Billion nationally when compared to APS if applied only to <10% of people with diabetes.


S. Sakkal: None.

Readers may use this article as long as the work is properly cited, the use is educational and not for profit, and the work is not altered. More information is available at