Use Apps for Diabetes, but With Care

Anne L. Peters, MD


October 11, 2019

This transcript has been edited for clarity.

Hi. I'm here at the European Association for the Study of Diabetes (EASD) meetings in Barcelona, where I'm presenting information about the use of digital apps in the management of diabetes. I'm on the American Diabetes Association (ADA)/EASD committee for evaluating devices, and our position statement on digital apps will be coming out soon, to be published in Diabetes Care and Diabetologia.

This is a huge topic. There are more than 1 billion people on the planet who use some sort of smartphone. There are more apps than I can count, and many people use these to help take care of their health. The biggest area is in diabetes.

Now, this may seem all well and good, and I am in favor of people using apps. What I'm not in favor of is people getting medical advice in ways that I can't control or at least validate. That's an issue because basically anyone can make an app and put it on the Internet for people to use.

What Types of Apps Are Available?

We need to think about how we look at apps. The first thing is to characterize them. What I call simple apps are those that help patients with nutrition, carb counting, and fitness. There are all sorts of apps that don't give much medical advice. Even with those, however, I have a bit of an issue.

For instance, in terms of the carb-counting apps, do you know who wrote those apps? Was it a registered dietitian or somebody who only thought they knew about carb counting? In comparing some of those apps, I found that the amount of carbohydrate recorded for common foods such as pasta or bread can differ substantially. As a clinician, I want to know who wrote the app and how much trust I can put in that app when I ask a patient to use it to help them with meal planning and carb counting.

Physical activity apps are simple. In my experience, they are great to use to count your steps, but at some point they become boring. I have a whole drawer full of Fitbits because I don't find that they add much to my exercise routine. I like Fitbits and I encourage them for patient use, but I want to figure out how to take that tool and create an app that will encourage physical activity long-term.

You can get glucose-monitoring apps with every kind of glucose meter and continuous glucose monitors, and many of them are helpful to patients because they can look at the data. But glucose-monitoring apps range from those that are really well done and helpful, to those that [falsely] promise the world. Some apps tell patients that if they follow the app, their diabetes will be cured.

Other apps provide health tips or help with setting targets. However, if you set a target that's too low for a patient who is at risk for hypoglycemia, you can increase their risk for lows. I have some patients who have very high A1c and glucose levels, and I don't want them to get too low too quickly. I want to be able to define the target range on any app that they're using, and I can't always do that.

I don't think apps that promise all these things are good for my patients. I want patients to realize that diabetes is a long, slow process that requires care and treatment over time, and that there's no one quick fix.

Again, I love apps that help patients see their glucose numbers, but I also want them to see those numbers translate into something clinically meaningful. To some degree, I like to choose the targets for my patients and to teach them what I think is important.

The FDA-regulated apps can be exceedingly helpful, in part because I know that they come from a source that is reputable. These apps are doing things like adjusting insulin, communicating with pumps, helping with the automated insulin delivery systems, or working with a smart pen. These apps are integral because many devices won't work without one.

Next Steps and Backup Plans

As we move forward, we have to think about interconnectivity. It's important for an app to actually work in the doctor's office and on the patient's smartphone. It's also important for an app for a pump to work with an app for a sensor and a blood glucose meter, so we can start pulling the data together to provide meaningful recommendations to patients.

We also have to think about what happens if the systems break down. I have had patients whose smartphones died, who lost their smartphones, drove over it with their car, or dropped it into water by mistake. All of those things can happen. If some piece of these integrated systems breaks down, gets lost, or is broken, there has to be a clear backup plan for a non–app-dependent way to treat our patients.

The committee's recommendation is to encourage the development of these apps, but we want people to use them wisely, in conjunction with their healthcare team who can help patients use and interpret them correctly.

Finally, a big issue with these apps is that people will use them for a while and then lose interest. We need to work with app developers to make them useful for patients over time and to find ways to really show benefit.

We definitely need outcomes data with apps, but outcomes data can't be obtained in the traditional way. We can't do RCTs, for instance, because these apps evolve relatively quickly over time, and data obtained at the beginning of the use of an app are going to be irrelevant 2 years later.

I think this is an exciting field. There is much promise, but I don't want to oversell it. We need to use apps wisely to help our patients improve their diabetes outcomes. Thank you.

Follow Medscape on Facebook, Twitter, Instagram, and YouTube


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.