New Dawn in Diabetes: Apps Can Improve Compliance, With Caveats

October 06, 2017

"When I became a doctor, every new intervention for diabetes was a drug or a shot. Now what really gets the buzz is software."

Dr David C Klonoff

So announced David Klonoff, MD, an endocrinologist from Mills-Peninsula Medical Center,  San Mateo, California, during an International Diabetes Federation (IDF)-sponsored symposium at the recent European Association for the Study of Diabetes (EASD) 2017 Annual Meeting.

While most current interventions are "for treatment," there is also great potential for prevention of diabetes, he stressed.

Dr Klonoff, who specializes in diabetes technology and is also the founding editor of the Journal of Diabetes Science and Technology, obviously has a vested interest in promoting this view, but there is no doubt that he is spot on in identifying technology as the "new frontier" in diabetes care.

Every week, there are new press releases proclaiming a dizzying array of novel technologies, and at every diabetes conference there is a host of company stands describing such a myriad of products that it is almost impossible for doctors, other healthcare professionals, and diabetes patients and their carers to keep up with.

"It's a lot to digest," acknowledged Kyle Jacque Rose, a board member of IDF Europe and a chemical engineer who has type 1 diabetes and works as an advisor for the diabetes technology company MySugr.

This subject is "very intimidating," not least due to the "sheer number of products" out there. As of 2016, there were a dizzying 165,000 healthcare apps, Mr Rose noted, adding that this figure is "already out of date."

Due to this "labyrinth of products," it's understandable that many are baffled, he said, and this confusion extends to national insurers in many countries "that want to understand how they can start to cover some of these advances but don't know where to start. "

Also, "we have member associations [diabetes associations in different countries] asking us at IDF whether we have a guideline," he notes.

Patients Need Motivation and Inspiration: Goals Beyond Glycemic Control

"Diabetes patients," Mr Rose stressed, "have to make around 50 therapy decisions every day — completely on our own."

Diabetes patients have to make around 50 therapy decisions every day — completely on our own.

Anything that can aid this is welcomed. Apps can help people to gain confidence and become independent with their own decisions, helping with issues such as therapy compliance and related aspects such as food, exercise, and helping to remember appointments by providing "motivation and inspiration."

Many apps have been born as a result of diabetes patients or their carers having encountered especially stressful situations, giving them a notion of how to help others, he said.

Kyle Jacque Rose

As an example, Mr Rose described how a family were at a sporting event with their son who had diabetes. He was suffering a hypoglycemic event and they had run out of blood glucose testing strips. So they had the idea of putting a callout over the loudspeaker system and in this way identified someone with spare testing strips. From this experience came the idea behind the app HelpAround.

Diabetes technology encompasses hardware, the actual equipment that people with type 1 and type 2 diabetes use (such as insulin pens and pumps and continuous glucose monitors) as well as software.

Mobile apps are defined as software programs that run on smartphones and other communication devices, or they can attach to a smartphone or other device, or be a combination of accessories and software.

In an attempt to streamline understanding of the many products out there, Mr Rose explained that apps for diabetes can broadly be divided into five categories. These are:

  • Logbooks/trackers. Examples include products such as that mySugr, as well as MyNetDiary, Glucose Buddy. and Dbees.

  • Nutrition/exercise apps. These include Carbs & Cals, Calorie King, Figwee, Eat Out Well, Strava, Runtastic, FitBit, MyFitnessPal, GoMeals, Meal Memory, and Bolus Calculators.

  • Wellness/motivation/coaching. Diabeo, Diabetes Goal Tracker, Withings, ihealth, Livongo, WellDoc, Tactio, Omada. mySugr, OneDrop, and HelpAround are some examples.

  • Middleware and electronic health records (EHRs). These incorporate products from Diasend, Glooko, Tidepool, Biomedtrics Ditto, SiDiary, Epic, and Allscripts.

  • Device manufacturer software. These are available from Abbott, Dexcom, Johnson & Johnson, Medtronic, Roche, and Sanofi.

Ideal App Needs to Have "No Friction," No Increased Workload for Docs/Nurses

©Micha Klootwijk|Dreamstime

Dr Klonoff said that mobile apps achieve their goals through a variety of approaches, including "story-telling" and "gamification." For example, the MySugr app has an avatar called "diabetes monster."

They can provide increased access to timely information for patients and their healthcare providers, and they support remote monitoring and diagnosis with wearable devices and offer decision support or timely recommendations.

But it's imperative that they are usable and that they provide clinical benefit, he urged.

Regarding usability, there are certain questions that users should ask themselves about the features of an app, he said: how much time will it involve — is it quick to use? Does it provide useful information? Is it user-friendly? And does it provide feedback? People like to get feedback as soon as possible, he noted.

And uptake and adoption of apps is facilitated by easy log-in (whereby there is a balance between security and simplicity) and simplified data and alert displays.

There is also the issue of whether a phone will prioritize a medical app (ie, ensure that this above all others works if the device's battery is low); "they didn't used to," Dr Klonoff observed.

And key to the whole process: is it approved by a healthcare provider?

For this, it's vital that any new products are interoperable with other digital systems, compatible with EHRs, and accepted professional standards and that they are safe in a mobile platform environment.

"Systems need to be able to communicate with each other and be compatible with the EHR. I think the marketplace will force compatibility," stressed Dr Klonoff.

It is also imperative that the doctor or other healthcare professional shouldn't experience any increase in work, time spent, or cost as a result of a patient using an app and that the health outcomes should be improved.

If I hear about work, time, or cost increasing, I'm not interested.

If I hear about work, time, or cost increasing, I'm not interested," he said, adding that the ideal app has "no friction."

Mr Rose agreed, stressing that as well as helping patients manage the ups and downs of the diabetes journey in their lives, apps need to address the needs of doctors and others taking care of patients in terms of analyzing data and guiding/teaching their patients.

Patients and healthcare providers need to "work together to set goals for individual diabetes self-management," he observed.

Economic Benefit, Security and Privacy: No "One-Size-Fits-All" Approach


Both speakers also touched on other important aspects to consider: economic benefit, which is needed to satisfy payers, security to preserve safety and satisfy product regulators, and data privacy to satisfy legal requirements.

Does the cost of an app involve a onetime fee, for example, or is it a subscription model? said Mr Rose. Another issue is compatibility with different phones or devices — such as IOS vs Android.

And it's important to understand, Dr Klonoff stressed, that "security and privacy are different."

Currently, mobile apps that meet the definition of a medical device, are an accessory to a regulated medical device, or transform a mobile platform into a regulated medical device, come under the umbrella of devices for regulatory purposes, Mr Rose explained.

Some apps are currently deemed class I devices, the first level of regulatory clearance. Some are class II and in this case will often be connected to another established medical device.

However, the "large majority" of apps are currently "not registered with a regulatory authority," he conceded.

Of note, the UK Medicines and Healthcare Products Regulatory Agency (MHRA) issued guidance detailing five categories of apps in August 2016.

Mr Rose also emphasized that there are "different tools for different phases of disease." A "one-size all approach is not going to work; there are so many products."

Take the issue of food in diabetes as an example: different apps can help with a diverse range of subjects such as "carb counting" and working out the impact of specific whole foods on blood glucose.

"Meal memory," which enables users to take a photo of their food with the smartphone and then counts carbs, among other things, might be helpful at the beginning of a diabetes diagnosis, then "later on, you may graduate to something else," he noted.

This could be something like the CareLink deep data analysis that shows users how to handle "all foods."

And it's important to realize when assessing apps that while some products "are just student projects," others "have a whole company behind them," he added.

Dr Klonoff agreed, noting that the device manufacturers "are doing some interesting work around things that we wouldn't have thought possible."

He added that many large pharmaceutical companies and device manufacturers have "bought into" technology, not least because they believe it will enable them to "see patterns that will help inform future product development."

As examples he cited approvals this year that have included Lilly's insulin dose calculator app, Go Dose, and an app called Cornerstones4Care (C4C), comarketed by Novo Nordisk and Glooko, which allows people with diabetes to measure and track blood glucose, activities, and meals. Other collaborations include one between Sanofi and a data and analytics firm called Evidation Health.

But Much More Research Still Needed

Dr Klonoff said that inconsistencies among current apps — in terms of whether they are officially regulated as devices, whether they are cost-effective, and security and privacy issues — make it imperative that much more research is performed in digital health for diabetes.

Yet there are "some common flaws in current digital health clinical trials," he stressed.

First of all, is the study "a randomized clinical trial?" and then does it include enough subjects, with enough power to see a statistical difference? And does it run for long enough? "We need to make sure we are looking at good-quality data."

He cited the example of a study indicating improved medication adherence and frequency of blood glucose self-testing using a mobile health platform vs usual care among type 2 diabetes patients in India.

The paper, he noted, indicated that there are approximately 69 million people with type 2 diabetes in India, and yet the study tested the platform in just 91 patients (Telemed J E Health. 2017;23:733-740).

Meanwhile, one doctor conducting research isn't convinced yet that medicine can rely entirely on technology, at least not for diabetes.

John Grumitt, of Newcastle University, who is also vice president of Diabetes UK, explained the "educational and psychological" approach his institution has pioneered for type 2 diabetes that has just been launched in the United Kingdom and will soon be introduced into the Netherlands, detailed at

He and his colleagues feel that, for type 2 diabetes in particular, it's too early to take healthcare workers out of the equation.

"We converted an existing [type 2 diabetes] education product into a digital version, but we couldn't find any evidence to support automated behavior change," he explained.

So "we added a clinical psychologist [coach] to support the patient over an extended period of time to achieve sustained change in clinical outcomes."

"Digital can help, but you need to have real people support you to make change, and that gives you an economic benefit. All the digital tools you like won't support you every day of your life," he stressed.

But Mr Rose noted that some of the "wellness/motivation/coaching" apps "do have a real diabetes educator" — albeit they are interacting remotely with the patient — and most are also applicable to type 2 diabetes. "So I'm curious, does that digital intervention from that real diabetes educator [in the app] fulfill your criteria?" he asked Dr Grummit.

Dr Grumitt responded that some of the things covered by apps — the automation of food diaries, activity trackers. and the like — "can be really helpful, but there's a journey that people make and it's very personal. The big gain is around adapting people's behavior."

His team is working with the national data center in England recording every conversation between psychologist and patient and will have results on how the program is progressing within 2 years.

Dr Klonoff said: "I like the approach you are using; it makes a lot of sense." But, he said, cost-effectiveness is still key, and he doubted that involving a healthcare professional one-on-one would turn out to be economical.

Dr Grumitt said that due to the amount of time involved, the program is not expensive. "We provide 100 minutes per year [one-on-one with a healthcare provider]. The evidence says 80 minutes per year is enough, so we're overdoing it a bit."

Are Patients Scared of Technology?

Dr Klonoff believes the way to bring costs down and make such diabetes technology scalable — considering all the data involved — may well turn out to involve artificial intelligence (AI), and "many start-up companies [in Silicon Valley] are turning to using AI," he observed.

Chair of the session Joan Manuel Valente Nabais, from Universidade de Évora, Portugal, a doctor of chemistry who has type 1 diabetes himself, wondered whether many people with diabetes are simply "frightened" of technologies such as AI.

Dr Klonoff acknowledged that this is an understandable emotion but used the example of driverless cars or autopilot on an airplane to describe why he thinks doctors and patients should put their faith in new technologies.

"The pilot has a choice of manual or autopilot when he lands the plane, and my friends in the aviation industry tell me the manual landings are always bumpier," he mused.

Mr Rose, meanwhile, recounted his experience of participating in a  trial of the artificial pancreas in a study in France.

"When I was in the artificial-pancreas trial, I kept putting aside the device and doing my own [blood glucose] checks," he said. The doctor running the trial "knew, that with my type A, engineer, and cyclist personality, I wasn't going to relinquish control so easily."

But after a few days, his faith in the technology grew, and he was able to stop himself from performing his own checks.

Dr Klonoff said: "This seems weird, but it does work, and we will get used to it."

Should Social Media Play a Role?

©David Prahl|Dreamstime

Moving to discuss the role of social media among diabetes patients, Dr Klonoff said he is not entirely convinced of the benefits of such platforms — which include Facebook, Twitter, and Instagram, among others — for improving adherence to treatment.

He said there are a number of points for consideration: patients must be comfortable with social media, and diabetes educators can assist with this; the site in use must be trustworthy, with ratings systems available in many cases to educate users; and healthcare professionals must be convinced of the benefits.

He said he doesn't specifically deter patients from using social media — "by all means go for it" — but "I'm not going to push it, either, until I see research indicating benefit for improving adherence."

And he conceded that in some instances there can be a gap between older diabetes doctors and many of their very tech-savvy younger patients.

One diabetes advocate spoke from the floor to defend social media, although he didn’t specifically address Dr Klonoff’s main argument — that more evidence is needed to show that social media increases compliance with treatment.

Bastian Hauck, a type 1 diabetes patient active on Twitter (@bast-ha), said: ""We have lots of evidence to show that peer-to-peer support is helpful in diabetes. Social media isn't new, it's just a new method of communication."

Social media isn't new, it's just a new method of communication. We don't need clinical trials to prove that it works.

Mr Hauck stressed that if there is misinformation circulating online, the community of people with diabetes are very quick to call it out.

"We're getting so many counterpunches! We don't need clinical trials to prove that social media works," he stressed.

Dr Klonoff is a consultant for Ascensia, Lifecare, Onduo, and Voluntis. Mr Rose works as advisor for MySugr and with INSEAD Healthcare Initiative as an EiR Visiting Faculty in Fontainebleau, France and Delta Project Management as a consultant.

European Association for the Study of Diabetes (EASD) 2017 Annual Meeting; September 10, 2017; Lisbon,  Portugal.  IDF Europe Symposium.

Follow Lisa Nainggolan on Twitter: @lisanainggolan1 . For more diabetes and endocrinology news, follow us on Twitter and on Facebook .


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.