Future of Diabetes Management – Recap of ATTD 2017

Written by Petteri Väisänen, Research Engineer at Quattro Folia.

As a biomedical engineer working on improving diabetes management, my favourite event is definitely the annual conference of Advanced Technologies & Treatments of Diabetes (ATTD). It was held from 15th to 18th of February, 2017 in Paris, France. Although, the scenery around downtown is full of history, the actual conference is really about the future! And in this case, the future of diabetes management. Next is my view of things that will be part of the future…

Since many of the presentations at ATTD started with the same slide, I find it only appropriate to start with the same figure. Thanks to T1D Exchange, there is now relevant data of the status of glycemic control in USA [1]. Figure 1 shows the average A1c among the type 1 diabetics and the figure is alarming. The findings show that there is an significant gap between the recommendations set by American Diabetes Association (ADA)  and the actual A1c values of type 1 diabetics. As an example, ADA recommends that diabetics in the age of 16-18 should have an A1c value of 7,5% (58 mmol/mol) [2]. However, the actual mean value is around 9% (75 mmol/mol). There is no question that these high values will have consequences, not only for the individuals themselves, but also for the society and economy. For instance, diabetes (and especially poorly managed) is shown to have impact on productivity, absenteeism and early retirements [3].


Figure 1: Solid line is showing the average hemoglobin A1c (the biomarker of glycemic control) by age among the type 1 diabetics in USA (n=16,057) [1]. The dashed line is showing American Diabetes Association (ADA) guidelines for adolescents and adults [2].

It is important to remember that diabetes management is not all about the A1c, but most of us have seen the figure with the exponential risks for secondary diseases that is related to poor glycemic control shown in DCCT study [4]. Furthermore, there is evidence showing that the prior glycemic control has sustained negative effect, even though the control is returned to normal. This is known as “metabolic memory” [5,6,7]

But enough of the gloomy part… Let’s open the door for brighter future and things that may improve individual’s diabetes management. Below is few highlights from ATTD.

Continuous glucose monitoring (CGM), including flash glucose monitoring (FGM)

Okay, CGM and FGM are nothing new, but they are here to stay. Hopefully in future, they will have a wider user base. There are plenty of studies that shows the benefits of CGM [8], and not only for T1D, but also for T2D [9]. Furthermore, CGM can be cost-effective option for even those T2D who do not use prandial insulin as a form of treatment [10]. In the case of CGM devices, there are two major positive things. Firstly, the manufacturers are eager to develop ways to improve the accuracy of meters, even with less calibration measurements. Increased accuracy will allow CGM to be used for treatment decisions also. Secondly, the increasing competition makes CGM accessible for more diabetics.

Insulin pumps and insulin types

Again, not a totally new thing, but there is some serious potential in these! Device manufacturers are trying to find ways to speed up the insulin absorption with pumps. They are developing solutions e.g. for dispersed injections, ways to tackle Tamborlane effect, intraperitoneal infusion, or even by warming the infusion-site. Insulin manufacturers offer wide variety of things to discuss, but let’s pick the most prominent: Ultrafast insulins! For instance, faster-acting aspart shows 25.7 minutes’ improvement on peak-time when compared to the insulin aspart on the market today and thus, better mimics the endogenous secretion of insulin [11].

Diabetes management software

This is where the magic happens! We have had the technology to connect pumps, meters and sensors to mobile devices for a long time. Now the devices are actually connected which enables applications to find ways to improve glycemic control. One that I have to mention is Artificial Pancreas (AP); a system that consists of an insulin pump, CGM, and a control device (including control algorithm). Many are aiming to have fully automated closed-loop control systems that would mimic the functions of human pancreas as much as possible. Although that might be far away, hybrid closed-loop algorithms are already here and some are already FDA approved [12]. You might only need to do as little as “announce a meal”, and the system does the rest… almost. There are cases where intensive exercise brings challenges to the controlling algorithms. And with the AP, the competition lies in the controlling algorithm. Who has the most efficient control algorithm which is also safe to use in all cases? Well, we still have to wait for the evidence from multiple studies that are ongoing, but I will say this: The competition is a promise of results!


Figure 2: An artificial pancreas system; the insulin pump is controlled with algorithm that can be in separate control device or in mobile phone. The algorithm has glucose sensor data as input but in future inputs can include factors such as insulin data, activity, and geolocation.

Before fully automated closed loop control is achieved, there are some tools that people with diabetes are shown to benefit from. For instance, the insulin bolus calculation is not that easy as one might think. Study shows that 63% of manually calculated doses were incorrect [13]. One way to solve this problem is to use a bolus insulin calculator that has been proven to provide correct values while taking into account carbs, exercise, and personal factors.

Second example is basal insulin titration for T2D that can be done though certified application. In result, the health care professional´s burden is reduced while the patients’ fear of hypoglycemia is also reduced [14,15]. Patients feel more engaged to the treatment because the application can educate as well as motivate the patient throughout the titration process.

Enabling continuous A1c measurement

At ATTD, we had the opportunity to present our latest results of individually adaptive Continuous A1c algorithm. It enables fast care feedback meaning that our algorithm accurately fills unmonitored gaps in glycemic control by using just routine monitoring data. By doing so, individuals with diabetes get timely feedback about their care quality since they can see how their individual A1c value changes. This is a feature that we are really excited about and plan to release it on the next version of Balansio Mobile. For more information about the Continuous A1c algorithm, click the link below to see our poster that we presented in Paris.

View the poster here

To conclude, data-driven and connected self-care is the future. Algorithms, such as Continuous A1c, are here stay and to improve diabetes management. Yet, the algorithms are only tools and diabetes management is a lot more than equations. This is why we are working hard to bring the interactive and educational support for all diabetics.


[1] Miller et al. Diabetes Care 2015; 38:971-978

[2] Chiang et al. Diabetes Care 2014; 37:2034-2054

[3] Breton et al. Diabetes Care 2013; 36:740-749

[4] Skyler JS. Endocrinol Metab Clin North Am 1996; 25:243-254

[5] Berezin A. Diabetes Metab Syndr 2016; 10:176-183

[6] Cerielle et al. J Clin Endocrinol Metab 2009; 94:410-415

[7] Aschner & Ruiz. Diabetes Technol Ther 2012; 14:68-74

[8] Liebl et al. J Diabetes Sci Technol 2013; 7:500-519

[9] Meade LT. Diabetes Technol Ther 2012; 14:190-195

[10] Fonda et al. J Diabetes Sci Technol 2016; 10:898-904

[11] Heise et al. Diabetes Obes Metab 2017; 19:208-215

[12] FDA approval of Medtronic Minimed 670G system 2016; P160017

[13] Sussman et al. J Diabetes Sci Technol 2012; 6:339-344

[14] Hsu et al. Diabetes Technol Ther 2016; 18:59-67

[15] Sieber et al. ADA 2016; 1219-P