Diabetes phone apps and CGMs – do they result in improved patient outcomes?

By 21/07/2016 No Comments

Diabetes, has been described by the International Diabetes Federation (IDF) as one of the largest global health emergencies of the 21st century (1).  Every 6 seconds, 1 person dies from the complications of diabetes, with 5 million deaths Worldwide due to diabetes in 2015 (1). With the prevalence of diabetes expected to rise to 642 million by 2040, the burden of diabetes will continue to have a devastating impact on both patients and healthcare systems alike (1).

Why is diabetes so difficult to manage and keep under control? The fact is that diabetes differs from all other conditions. It is a silent beast, high blood glucose (hyperglycaemia) is largely invisible until complications develop. This is particularly true in type 2 diabetes where patients may be undiagnosed until complications develop at which point significant damage may have already occurred. In addition, and potentially more importantly, the majority of the onus of care is on the patient. The need to regularly monitor blood glucose levels and adjust food, insulin, additional medications and exercise accordingly requires a considerable amount of time, dedication and self-motivation.

FingerPoked Gareth cartoonSince the development of the first blood glucose meters in the 1970s, which were larger than house bricks and not easily portable, self-monitoring of blood glucose has come a long way. Today the average blood glucose meter easily fits in the palm of a hand, requires a very small sample of blood and provides an accurate blood glucose level in around 5 seconds.

Blood glucose levels are affected by many factors including the amount of carbohydrate, fat and protein consumed, levels of circulating insulin, type of insulin, over or under dosing of insulin, other medication, exercise, stress, illness, alcohol consumption and even extremes of weather. This may result in significant variability in blood glucose levels from one day to the next even if the same food is eaten and insulin levels are given. Therefore, in order to get a good understanding of overall diabetes control, it is important to track blood glucose levels to identify any trends occurring over a defined period of time.

Over the past five years there has been a large surge in the number of diabetes management apps available for either iPhone or Android systems (there are around 100 diabetes apps available for iOS devices alone). These apps vary slightly in capability but all allow the user to enter their blood glucose reading, the event associated with that reading (e.g. breakfast, exercise, felt low), most allow the user to enter carbohydrate units, units of insulin, other medications and some also track exercise, weight and blood pressure.

In my opinion these apps do add value but they still require the patient to input values into the app and that’s where their weakness lies. It’s akin to getting a new app to track calories, for example, in the few days it’s easy to enter calories but after a week or two the enthusiasm wanes, it becomes a chore and the app is no longer used. To have real value, in my opinion at least, an app needs to have an automated data entry eliminating the need to input data from the glucose meter to the device. To my knowledge there are 2 apps that employ data automation of blood glucose levels. Glooko works on iPhone and Android and is designed to sync with glucose monitoring devices and fitness trackers but requires an online subscription ($60 per year). iBGStar(R) Diabetes Manager app works together with the BGStar(R) blood glucose meter and the iBGStar(R) which connects directly to an iPhone. I envision a time when automated data transfer becomes a standard feature of these apps.

Measuring blood glucose levels 4-5 times per day provides a snapshot of blood glucose levels at that moment in time but may not provide a full picture of overall control.  Continuous glucose monitoring (CGM), however, provides real time tracking of blood glucose levels via a sensor placed just under the skin and can identify high and low blood glucose levels and trends that can be missed via fingerstick measurements alone (see Figure below). An analogy that has been used is that a fingerstick is a single photograph of an event while CGM is a video of the full event. The use of a CGM provides patients, caregivers and healthcare professionals a better understanding of a patient’s overall diabetes control over a 6 day period and facilitates adjustments in insulin dosing.

Fingerstick vs CGM

A number of companies market CGM systems including Medtronic, Abbott and Dexcom.  Abbott also market an alternative to a CGM which allows measurement of glucose in interstitial fluid independent of an insulin pump using a handheld reader which is ‘flashed’ over the sensor.  The main downside to these systems is the cost. Starter kits including transmitters cost in the region of £500 with sensors that generally last for 6-7 days costing in the region of £40-60 each. At the present time, despite the potential to substantially improve patients HbA1c levels and thus prevent costly diabetes related complications, CGM is not routinely funded by either the NHS in the UK or Medicare in the USA.

A group of parents of children with type 1 diabetes have developed Nightscout (GCM in the cloud) which allows real time access to CGM data via personal website, smartphones viewers or smartphone apps (see Figure below).  Nightscout is an example of parents and patients taking diabetes control into their own hands. When first implemented, Nightscout was a solution specifically for remote monitoring of Dexcom G4 CGM data. Today, there are Nightscout solutions available for Dexcom G4, Dexcom Share with Android, Dexcom Share with iOS, and Medtronic. The goal of the project is to allow remote monitoring of blood glucose levels using existing monitoring devices. It should be noted that Dexcom now market an off the shelf solution that facilitates smartphone viewing and Medtronic have recently received a CE mark for a similar system which is expected to launch in 2017.

Do it yourself medical monitoring for diabetes

So have these advances in self-monitoring technologies resulted in better outcomes for patients? In short, no benefits have been realised to date. Data from the National Diabetes Audit (2) shows that the proportion of people with type 1 and type 2 diabetes achieving the NICE HbA1c target of < 58mmol/mol (<7.5%) has been relatively stable over the last six years; <30% and <67% of patients with type 1 and type 2 diabetes met the target, respectively (see Figure below).

Percentage of people with diabetes in England and Wales achieving HbA1c targets by diabetes type and audit year

When there are published data to show that both glucose tracking via apps and the use of CGMs result in improved HbA1c levels (3, 4, 5, 6), why have HbA1c levels in England and Wales remained constant? It is possible that the reach of the technology has not penetrated far enough to make a difference. In 2013 it was estimated that only 1.2% of diabetes patients that have a smartphone use a diabetes app, with the number expected to increase to 7.8% by 2018 (7). Similarly, the proportion of patients with access to CGMs is low at approximately 10% in the USA (8) and likely to be even lower in the UK.  CGMs may be too cost prohibitive for many patients at the current time to gain widespread use. In my opinion, if adopted on a larger scale, these technologies could make a real difference to patients, both in terms of diabetes control, empowerment and quality of life. These benefits could also translate into cost savings for healthcare systems as fewer patients develop costly diabetes complications. With eighty percent of the £10 million to treat diabetes associated with the treatment of complications (9), should the NHS and other healthcare systems consider funding CGMs and using healthcare providers to educate and promote diabetes apps to put some of the power and motivation into the hands of the patient? After all it is their daily battle and beast to silence.


1) IDF Atlas 2015; 2) National Diabetes Audit 2013-2015; 3) Liang et al. 2011. Diabetes Med 28: 455-463; 4) Quinn et al. 2008. Diabetes Technology and Ther 10(3): 160-168; 5) Huang et al. 2010. Diabetes Care 33(6): 1269-1274; 6) Deiss et al. 2006. Diabetes Care 29(12): 2730-2732; 7) Research2guidance report 2013; 8) Wong et al for the T1D exchange clinic network. 2014. Diabetes Care 37(10): 2702-2709; 9) The cost of diabetes report V2 2014 Diabetes UK.