Diabetes Patients Design Their Own 'Artificial Pancreas'

Aaron Neinstein, MD


April 21, 2016

The OpenAPS system seems to work quite well. It doesn't work perfectly in every situation—again, this is not a panacea nor an FDA-approved, commercially available artificial pancreas—but it almost always works better, Dana says, with reduced cognitive effort, compared with what people were using previously.

The results can be seen graphically in tweets posted by Howard Look, CEO of Tidepool, a nonprofit organization that facilitates the sharing of diabetes data. Below is an example showing two nights of typical type 1 diabetes management followed by a night using the "closed loop" OpenAPS.

Figure 4. Tweet from Howard Look shows glucose management with OpenAPS.
Photo courtesy of Aaron Neinstein

Real-World Experiences

I spoke with one of the users of the OpenAPS system, who told me that it has given him overnight control of his blood glucose levels. He has almost completely eliminated overnight hypoglycemia, which is a major fear of most people with type 1 diabetes, given the high lifetime prevalence of "dead in bed" syndrome.

This user noted that overnight lows were "always the most scary and depressing part of diabetes... waking up in sweat, shaking, and lying still until it's over. It doesn't happen anymore [with OpenAPS]." He now wakes up with a blood glucose level of 100-110 mg/dL almost every morning, starting each day "with a fresh chance." He also noted that because of this, he no longer has overnight alarms for low blood glucose values, so his "alarm fatigue" has disappeared.

In another example, a parent who set his child up with OpenAPS tweeted his own sleep data, showing that he gets more hours of sleep each night now that he is not having to manage his son's diabetes overnight.

Has OpenAPS completely eliminated the entire burden of treatment? No. Patients or their caregivers still have to deal with sometimes-unpredictable pump site changes, hormone-driven variations in insulin sensitivity, and dozens of other variables that make type 1 diabetes such a challenge to treat. And for safety reasons, the OpenAPS does not deliver mealtime boluses, so a user has to continue estimating meal carbohydrates, pre-bolusing before meals, and bolusing at mealtime.

But the ability to let the system handle things overnight, or for a few hours between meals, is an incredible relief to patients and their caregivers.

As with any do-it-yourself project, there are still numerous hiccups along the way as people build their own systems. But the smiles on users' faces and the light in their eyes as they talk about OpenAPS tells the story better than any numbers. They feel hope about something that once left them hopeless. They see endless possibilities where there once were few. They feel empowered to improve their lives where they once felt feeble. The OpenAPS project inspires, enables, and builds.

A 'Perfect Storm'

There is a sea-change underway in healthcare—most often invisible, yet rumbling just below the surface, ready to erupt—that explains the emergence of OpenAPS. Dr Eric Topol described several converging "C" factors in his book The Creative Destruction of Medicine: constant connectivity, collaboration and crowdsourcing, customized consumption, and cloud computing.

OpenAPS and the #WeAreNotWaiting movement are emblematic of healthcare that is patient-driven rather than physician-driven; technology-enabled, crowdsourced and social-media enabled, and run on cheap component technology rather than large enterprise technology, using learning systems and personalized algorithms rather than population-based treatments. This is truly the perfect storm of the modern world: readily accessible tools and technology, combined with patients who are desperate for help, combined with social media connecting them all together.

Innovations to "solve your own pain" have been slower to emerge in healthcare than in other industries. For example, Tinder, the mega-hit dating app, was created more than 3 years ago by students at the University of Southern California who were dissatisfied with online dating services. Snapchat was also created by college students who wanted to create social communications with less weight and permanence.

The story of OpenAPS has so many incredible facets: that adoption of such a life-or-death medical technology is spreading via the Web and social media; that the core algorithm powering something so seemingly complex is actually only 270 lines of software code; that it was essentially built as a hobby project by a couple with no experience working on medical devices, with a supporting social media community continuing to refine, tweak, and iterate it. It reminds me of the lesson from the 2007 Pixar movie Ratatouille: "Anyone can cook."

This is the true lesson of the #WeAreNotWaiting movement. Healthcare is not something that we, the medical establishment, are supposed to develop inside an ivory tower and then throw over the wall at our waiting patients. This is a team sport. The people with the illnesses are the ones we are here to serve, so we should engage them and their talents at every opportunity.

What does this mean for us as clinicians? For one, we must take note of these efforts and fight the urge to resist them or bury our heads in the sand. Better yet, we should participate in the conversations. We can continue our role as educators and ensure that the tools are designed properly. We can continue to coach our patients and help them feel positive and encouraged about managing their illness.

#WeAreNotWaiting is a positive and productive community effort that helps its members feel ownership and pride while avoiding helplessness and depression. People who are newly diagnosed with type 1 diabetes often feel scared, alone, and vulnerable. It has always been our job to help them feel empowered, educated, and engaged. This community is one tool that we can recommend toward achieving that goal.

Perhaps most exciting of all is that, alongside continuing research efforts, we are witnessing the inflection point in diabetes history where the artificial pancreas truly is just around the corner.


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.
Post as: