The Accidental Innovator

SEVERAL YEARS ABO, Dana Lewis had a nightmare. She dreamed that she had awakened in the middle of the night in a pitch-black room, unable to move, as though she were tied down. In the dream, Lewis—who has type 1 diabetes—knew her blood sugar had plummeted, triggering the paralysis, but she couldn’t even roll over to look at the glucose monitor on her nightstand. Alone in her apartment, unable to help herself, she knew she was going to die. No one would even know.

It was just a dream, but not an unrealistic one. Many people with type 1 diabetes experience episodes of nocturnal hypoglycemia, as they’re known in medical parlance—and such an episode involving a seizure or a coma can, in turn, lead to what’s less esoterically called dead-in-bed syndrome. Ever since her freshman year in high school, when Lewis was diagnosed, she had gone to sleep knowing that she might not wake up.

To prevent this, Lewis relied on continuous glucose monitoring, or CGM. Data from a tiny sensor embedded in her skin was relayed to a nearby monitor, triggering alarms if her glucose fell too low. It was state of the art. It also didn’t work very well. Sometimes the alarm sounded when her glucose level was fine, and even when the device worked perfectly, she often slept through it.

Lewis, who had moved to Seattle in 2010 and for the first time in her life was living alone, wished the monitor could alert someone else, too. Could she somehow improve the device on her own? It turned out that other users had asked the same question. A group of parents of children with type 1 diabetes were frustrated by how hard it was to monitor their kids’ glucose levels while they slept or played outside. The parents had gotten together and, working in their spare time, started to reverse-engineer the monitor’s software, eventually forming an organization called CGM in the Cloud.

One of the group’s founders, a programmer named John Costik, shared the software code with Lewis. She and Scott Leibrand, a former engineer at Twitter (he was then her boyfriend, and now is her husband), wrote a program to play extra-loud alarms through her laptop when her nighttime blood sugar dropped. If Lewis missed an alarm, the program alerted Leibrand, who could call and drive straight over if she didn’t answer. That adjustment let Lewis sleep a lot easier.

These days CGM in the Cloud is called Nightscout—an open-source, do-it-yourself, volunteer-run project that finds new ways to use and transmit glucose data from personal monitoring devices. The effort now boasts more than three dozen software developers, and their remote-monitoring software is used by as many as 10,000 people on their computers, smartphones, tablets and watches.

Most breakthroughs in health and medicine are, unsurprisingly, the products of academic labs or the research and development departments of big corporations. But plenty of useful, inventive and therapeutic ideas come from outside those walls, from patients and their clinicians, often when they’re wrestling with a pressing lifestyle problem that has no other solution. And while an inventor in her garage probably won’t be the person to cure cancer, physicians and researchers are beginning to take note of the volume of innovation happening on the front lines, and how useful these micro-breakthroughs can be.

Harnessed systematically, with the right safeguards in place, this body of everyday discoveries could counter what sometimes seems to be a flattening trajectory of medical innovation, with its incremental improvements, me-too drugs and real but often extraordinarily expensive advances. A growing number of researchers are building bridges to these innovators, and connecting them both with other patients who might benefit and with pipelines of medical discovery.

“The patient innovation process is no different from open-source software,” says Harry DeMonaco, former director of pharmacy and the Innovation Support Center at Massachusetts General Hospital. “Patients innovate because what’s available to them from standard approaches is inadequate. Their needs are not being met—and so they meet those needs.” If patients are expected to take medical advice, says DeMonaco, they should be able to give it back to the medical community.

IN THE MID-2000s, DeMonaco spent two years as a visiting scholar at the Massachusetts Institute of Technology, where he worked with innovation specialist Eric von Hippel, the author of Democratizing Innovation. Von Hippel looks at how users, of everything from semiconductors to windsurfing boards and computerized payroll services, often have been the original developers of important products and services and have improved existing ones in ways that commercial designers hadn’t envisioned—and in some cases couldn’t envision.

That squares with a history of medicine in which many common treatments, ranging from Band-Aids to light therapy for seasonal affective disorder, have come from patients grappling with problems that top research labs never addressed. Sometimes innovations arose from communities affected by rare diseases that didn’t command much commercial attention. Others were spurred by quality-of-life issues that were more important to patients than to their doctors. “One thing we know about innovation is that people tend to solve their own needs, not other people’s needs,” says von Hippel. “It’s important to empower patients.”

In a 2015 study, von Hippel and innovation scholar Pedro Oliveira of Portugal’s Católica Lisbon School of Business & Economics surveyed 500 rare-disease patients and their caregivers and found that more than a third found their own solutions for problems caused by their disease. Most of these fixes weren’t actually new—they were known to the medical establishment but hadn’t made their way to patients. But in 40 cases, patients had developed something completely novel.

One solution caught Oliveira’s eye. It involved Joaquina Teixeira, a vice president at Portuguese-rare disease patient organization Raríssimas, whose six-year-old son, Gonçalo, had Angelman syndrome. Caused by deletion or mutation of a gene crucial to neurological and motor development, the disorder can make movement difficult.

Gonçalo ordinarily struggled to stand. But his mother observed how he became transfixed by floating helium-filled balloons at a party and tried to reach them. Later that day she purchased balloons to put inside her own home. Gonçalo didn’t just stand to reach them; he tried to walk and even jump. The balloons so enthralled him that he pushed through the difficulty of his disease. “What she did was very simple. It cost her just a few dollars,” says Oliveira. He and Teixeira started telling the story to other patients—and the balloon trick worked for them, too.

What was unusual was that the story was shared at all. In the survey by Oliveira and von Hippel’s group, only a few innovators said they’d told others of their work. “An estimated 6% to 8% of the world’s population—hundreds of millions of people—are afflicted by rare diseases,” they wrote in an article in the Orphanet Journal of Rare Diseases. “Our finding suggests that a massive, noncommercial source of medical innovations exists.”

To help get the word out, Oliveira and Helena Canhão, a rheumatologist and professor at the University of Lisbon School of Medicine, founded Patient Innovation, whose website contains hundreds of patient-submitted accounts. The group has also partnered with more than 20 patient associations around the world to solicit new ideas.

Eventually, says Oliveira, they hope to form a business incubator to help patients who want to commercialize their ideas. “As doctors, we’re much more worried about the pathophysiology, the drugs, trying to combat the causes of disease,” Canhão says. “Sometimes we’re not so worried about quality of life. The patients’ solutions are often directed toward this. They can improve life beyond the drugs we’re using. It’s not competing with the conventional. It’s complementary.”

THERE ARE OTHER ways to tap patient insights—for instance, by scanning online forums of patient communities. One of Harry DeMonaco’s favorites is Crazymeds, founded in 2003 by Jerod Poore of Montana, a self-proclaimed recluse and “Citizen Medical Expert.” The site lets people with psychiatric and neurological conditions share knowledge and questions about their treatments.

In 2004, after the Food and Drug Administration approved aripiprazole, an antipsychotic for treating bipolar disorder which is sold under the brand name Abilify, it took less than a month for discussions on Crazymeds to reveal that night sweats—a side effect buried deep in the fine print of the drug’s safety information—were a frequent problem. And even as doctors told them that Abilify wasn’t to blame, these patients came up with a solution—reducing their dose initially and then gradually increasing it back to the prescribed amount.

Few doctors would advise their patients to tinker with their medications, and for good reason. But the insights were valuable to DeMonaco, who also kept an eye on physician websites, waiting for doctors to mention the night sweats problem. They never did. DeMonaco used another online community, Ask a Patient, for a research project on the side effects of osteoporosis drugs. Following the advice of an MGH patient who had experienced incapacitating joint pain after taking one of a common class of drugs called bisphosphonates, DeMonaco found dozens of such cases on Ask a Patient. “I asked these folks, ‘When you went to a physician, did they think this drug could cause that disorder?’” recalls DeMonaco. “To a person, they said no. But now there’s labeling on the drugs for this effect.”

THIS, TOO, IS small-scale innovation—guidance from patients as drugs move beyond the relatively narrow testing of clinical trials to regular use by millions of people.

A related strategy to harness insights comes from “positive deviance,” an approach pioneered in the 1990s by nutritionist Monique Sternin and her late husband, Jerry Sternin. Working in Vietnam with Save the Children, they met with the residents of four villages to identify families who were extremely poor but whose children were nonetheless well nourished: deviants from the norm, but in a good way. That led to investigations of the behaviors that might underlie the deviations. In every case, families fed their kids shellfish and washed their hands before they ate, among other practices—bucking conventional village practice.

That kind of approach has since been refined and applied to many conditions, including heart failure, Chagas disease and childhood obesity. Cole Zanetti, a family physician and preventive medicine resident at New Hampshire’s Concord Hospital, for example, searched the hospital’s patient records to identify 90 people with “complex” type 2 diabetes who had not only the disease but also a host of complicating factors that included depression, anxiety and poverty. Among that 90 were four who had nevertheless managed to stay healthy.

With a marketer’s touch, Zanetti dubbed the four “bright spots” rather than deviants, and last year he enlisted their help in hosting monthly meetings with other patients who shared their complicated health profiles. Attendees commiserated and swapped small tricks: freezing fruit to make it impossible to eat fast enough to send blood sugar spiking; avoiding supermarket checkout lines that sell candy; recognizing that boredom often leads to unhealthy behavior. Zanetti himself merely listened, taking notes and letting the patients tell one another what worked for them. They weren’t the sort of people who would be asked to author journal articles or to serve on institutional advisory committees, but there was no better authority on how to live with their disease. By ignoring the contributions of the bright spots all these years, Zanetti says, “we weren’t asking the right people what’s important.”

DOCTORS AS WELL AS patients can represent an untapped reservoir of innovation. In one research project, DeMonaco and von Hippel concentrated on clinicians who had found previously unknown uses for pharmaceutical treatments. This happens frequently: A majority of so-called off-label applications, in which a drug is used to treat conditions other than those for which it has been federally approved, are discovered by physicians on the front lines. Like patients, though, doctors tend to keep their insights to themselves. That’s not because going off-label is verboten; while the FDA forbids the marketing of drugs for off-label purposes, doctors can suggest new uses. But it’s a hassle to publish those insights in journals, with few financial incentives.

One answer to making these breakthroughs better known, suggests DeMonaco, could be Doximity or SERMO, social networking platforms that let physicians discuss their ideas with other doctor members. There could also be more systematic approaches. Cures Within Reach, a nonprofit that searches for ways to “repurpose” drugs and funds clinical trials for those it discovers, has found that some two-thirds of researchers and one-quarter of clinicians already have an off-label idea ready for testing, according to Cures Within Reach president Bruce Bloom.

The organization originally supported research on new drugs. But a 2009 review discovered that none of some 190 new-drug research projects during the previous decade had yet helped a patient. In contrast, four of 10 repurposing projects had already reached clinical practice. “There are so many more opportunities to repurpose things we already have and drive them to patients quickly and affordably,” Bloom says.

Recent Cures Within Reach projects used the BCG vaccine against type 1 diabetes; treated autoimmune lymphoproliferative syndrome, a childhood immune system disease that had been considered untreatable, with rapamycin, a cancer drug; and attacked recurrent pediatric brain cancer with the anti-worm medication mebendazole. Each of those began with a doctor’s observation.

Normally, physicians have neither the time nor resources to pursue such ideas, and drug companies tend to be focused on developing new, patent-protected blockbuster drugs. So another Cures Within Reach project involves funding models that reward companies for money saved by putting generic drugs to new purposes. The first of these should go operational in the United Kingdom this year.

Bloom would love to see the National Institutes of Health put a chunk of money toward combing through journal articles and medical records in search of one-off treatments improvised by doctors and informally reported therapies. Then, says Bloom, “I’d have a huge drive among the clinical community. I’d say to physicians, nurses, nurse practitioners and lower-level workers, ‘You have a tremendous amount of knowledge about what works—and until now we haven’t given you a venue for testing your ideas.’”

THERE’S NO QUESTION that a paradigm of bottom-up discovery comes with a number of logistical, ethical and regulatory hurdles. In the early days of Patient Innovation, says Oliveira, some people resisted the project; they feared potential legal liabilities posed by innovations developed by amateurs, outside the institutional safeguards of professional oversight. But reputable doctors like Canhão and her team provided crucial credibility and support for the project.

Ownership of ideas and data is another issue. The DIY community that created Nightscout, the glucose-monitoring software, is in a long-standing dialogue with the device manufacturers whose data they use. The data come from their own bodies, but they didn’t invent the monitoring devices that gather this data.

The patient community scored a victory last October, though, when the U.S. Library of Congress, which oversees copyright issues, approved regulatory tweaks that protected the ability of patients to work with their own medical device data. That had technically been illegal, says Ben West, a prominent Nightscout programmer and petitioner in the Library of Congress decision. While companies hadn’t prosecuted patients for customizing their devices, West believes the possibility that they could sue was discouraging commercial development that might support even more sophisticated, user-friendly innovations.

Meanwhile, seeing the energy and sophistication of Nightscout, diabetes device companies have slowly come to respect the contributions of these amateur developers. Dana Lewis expects them to eventually collaborate with Nightscout, a step that could hasten innovation. “We’re at an exciting point now,” she says. “We’re able to talk with the companies about what they’re focusing on, and we’re basically a beta-testing group for this new technology. We would celebrate if our features got incorporated in future devices.”

After upgrading her CGM alarms, Lewis and her husband went on to design what they call the Do-It-Yourself Pancreas System—a combination of glucose monitor, insulin pump and software that automatically adjusts her insulin dose in response to food intake and physical activity. That’s normally an extremely difficult task for people with diabetes, who must make dozens of fundamentally high-risk calculations every day.

The system works so well, says Lewis, who two years ago ran her first marathon, that she recently ran a Ragnar Relay—an exceptionally grueling race that’s spread over two days and a night, and is even more demanding than a marathon. “With the exception of being an astronaut,” she says, “diabetes has not stopped me from doing anything I wanted to in my life.”