One night, 18-month-old Hazel Lumpkin woke up with her diaper completely soaked with urine. Her parents, Matt and Melody Lumpkin of Pasadena, Calif., rediapered her in a larger size, hoping that would fix the issue. But Hazel continued to saturate diaper after diaper.
So she called the pediatrician, “who brushed me off because it’s a really uncommon diagnosis for a kid that young,” Melody said.
Eventually, they took their daughter to urgent care where doctors discovered that little Hazel had a blood sugar level of 500 mg/dL when a reading of 200 mg/dL constitutes grounds for a diabetes diagnosis. She was suffering from the early stages of diabetic ketoacidosis, a life-threatening complication that involves a dangerously high buildup of ketones, a chemical that causes the blood to become acidic.
The Lumpkins were sent home with some basic education on how to keep Hazel’s blood sugar in check, and felt overwhelmed and frightened by their new round-the-clock duties.
“It feels a bit like someone has said, ‘Okay, you have to play a board game and you have to play it about every 20 minutes, 24/7,’ and it involves math, and if you get it wrong, a person that you love could die,” Matt said.
His daughter’s clunky and cumbersome diabetes management devices frustrated him.
A designer of apps and software, Matt prided himself on making the user’s experience of technology frictionless and efficient — but Hazel’s devices felt like sheer torture to use. Pressing a single button hundreds of times to input a blood glucose value, or having an obnoxiously loud beeping notification go off during his work meetings, made it feel like the tools available on the market weren’t built by people actually living with the disease.
He went searching for better solutions — and discovered a worldwide community of engineers, software developers and designers who didn’t want to wait for the big medical device corporations to release another mediocre diabetes product.
Instead, they upheld a do-it-yourself approach to diabetes management, which might mean taking apart an insulin pump so it can communicate with your smartphone or reverse-engineering a sensor to remotely access your blood glucose data.
Members of the #WeAreNotWaiting movement have diabetes, or they have family members with the ailment. They experience the day-to-day grind of dealing with it and aren’t afraid to tweak the technology to meet their needs.
“People are able to connect online and share innovations in both software and hardware, as well as new ways of using ordinary equipment. That is a really powerful thing,” said Eirik Arsand, a professor at the Norwegian Centre for E-health Research. “The DIY movement is a positive for patients, absolutely. On social media, you can see how many people who have said it changed their lives and made things easier.”
In Type 1 diabetes, the body’s immune system attacks insulin-producing cells in the pancreas. Without insulin to help glucose get into cells, the sugar builds up unchecked in the bloodstream, which can damage the heart, eyes, kidneys and other organs. As a result, people with diabetes must stick to a target range of blood glucose. But food, alcohol, physical activity and stress can all affect blood sugar levels, so staying within that range can be challenging.
The three major DIY projects for diabetes — OpenAPS, AndroidAPS and Loop — are free, open-source algorithms that attempt to reduce the burden of managing diabetes. The software automatically adjusts an existing insulin pump’s baseline or basal delivery based on readings from a continuous glucose monitor to keep patients in the safe range between meals and during sleep. Medical device companies have historically kept insulin pumps isolated from continuous glucose monitors, but these DIY systems managed to “closed the loop” between the two tools.
For this reason, they are known as hybrid closed-loop systems.
Matt and Melody chose Loop, which works with an iPhone using info from Hazel’s glucose monitor to automatically adjust her insulin every five minutes. As her blood sugar goes up, Loop sends a signal to her pump to raise her insulin levels. When her blood sugar goes back down, it shuts off the pump.
“The job that we had been doing every two or three hours, all throughout the day, is now being done every five minutes by a robot that doesn’t get tired and doesn’t get bored with it,” Matt said. “We sat there crying because we could finally breathe. It’s hard to overestimate how much this changed our lives.”
Arsand, who was diagnosed with Type 1 diabetes in 1997, saw the early possibilities of using mobile technology to improve diabetes management.
In 2005, he and his colleagues tested a prototype system that could automatically transfer readings from a child’s blood glucose monitor to their parent’s mobile phone.
In a recent study on the DIY movement, Arsand found that DIY activities are highly visible on social media and the Internet in general, but there are hardly any results or discussions in the scientific literature — something he hopes to change with his own work.
“We are ‘doing-it-yourself’ within research. We are fascinated by the DIY community and even implement some of the tools ourselves,” said Arsand, who for his own care regularly uses Nightscout, a DIY project that allows real-time access to blood glucose data via smartphone, smartwatch or a website.
“We watch what they are doing and try to inform other researchers, health-care personnel and authorities about what is going on,” he said.
Patients must also consider the inherent risks of DIY devices and services, which have not been approved by the Food and Drug Administration.
In May, the FDA published a warning against the use of unauthorized diabetes management devices that “could result in inaccurate glucose level readings or unsafe insulin dosing, which can lead to injury requiring medical intervention or death.” The FDA warning was prompted by a case reported to the agency involving a patient with a DIY system who suffered an insulin overdose as a result of incorrect glucose readings.
Although the patient recovered, the incident raised broader questions about the safety of DIY medical devices.
“The FDA has a rigorous process to ensure that cleared devices are safe and effective. A patient who modifies a cleared device to function in a way that is not consistent with the manufacturer’s indications on the product label is taking a risk,” said David Klonoff, medical director of the Dorothy L. and James E. Frank Diabetes Research Institute of Mills-Peninsula Medical Center in San Mateo, Calif. “I agree with the warnings by FDA against the use of unauthorized devices.”
Klonoff also stressed that many physicians, for liability reasons, will not work with devices that haven’t been cleared by the FDA, so DIY patients may receive less medical supervision than those using authorized devices.
But even FDA-approved medical devices can sometimes fail, as seen in the recent service outage of Dexcom G6 continuous glucose monitors that sent diabetes parents into a panic. While the Dexcom Follow feature normally alerts parents when their child’s blood sugar goes too high or too low, it went down without warning in the middle of the night. The company said the problem has been identified and it is working to make sure it doesn’t happen again.
Instead of siloing DIY and industry, Klonoff believes that a collaboration among different stakeholders could result in technology that optimizes safety, usability and innovation.
“The best product will derive from ideas about DIY, medical device companies and researchers,” Klonoff said. “Patients have some great ideas about what works for them, manufacturers have methods for building great products, and researchers have ideas for how to design products with new features.”
Matt Lumpkin now works as a product designer for Tidepool, a software nonprofit firm that aims for this kind of hybrid approach. The organization is developing Tidepool Loop, based on the original DIY Loop software, and plans to seek FDA approval for it. It will maintain the free, open-source spirit of the DIY movement while going through the proper regulatory channels and closely collaborating with medical device companies.
“The majority of people are using DIY solutions because it’s all there is to get the features they’re looking for and a user experience that’s focused on improving their quality of life,” Lumpkin said. “I think this patient-led movement and #WeAreNotWaiting has pushed companies to take the concerns that are articulated in these DIY projects more seriously and move in those directions.”
As for Hazel, she is now 4 years old and recently celebrated what her parents call her second “Loopiversary.”
The family hasn’t had any issues with the DIY system. Melody Lumpkin said that every piece of technology that contributes to Hazel’s care — an insulin pump to replace the injections, a continuous glucose monitor instead of finger sticks and Loop — has helped the family’s feelings of empowerment and control over the diabetes.
“When we got a Dexcom [continuous glucose monitor] and put it on her, it was like the first time I could breathe since she was diagnosed. When we started using Loop, we were finally able to start sleeping again,” Melody said. “It seems like every tool we’ve added, I felt like I’ve gotten a little bit of my life back.”
