How Patient-Led Innovation Rewrites the Rules of Data Access with Anna McCollister

 Melanie: What becomes possible in healthcare when the patient isn't just along for the ride, but the one who's at the wheel? Thousands of people worldwide now manage their type 1 diabetes using an automated insulin delivery system that no company built. Instead, it was built by patients and parents and shared for free on GitHub.  

And this is not a hypothetical scenario. It's real life, and it's the direct result of a movement our guest today helped start. Anna McCollister has lived with type 1 diabetes for decades, and she spent much of her career turning her own frustration with inaccessible health data into change in healthcare.  

She was integral to #WeAreNotWaiting movement, which pushed device makers and regulators to open access for data for patients on their timeline, not the industry's. Today, Anna is the founder of Four Lights Advisors, a member of the ONC Health Information Technology Advisory Committee and a board member of the Sequoia Project.  

Her work at these places can be summed up in one word: trust. As we'll hear today, trust with patients isn't something you talk your way into. It's something you build one action at a time, and when you do, the whole system moves in the direction the patient has been steering all along. 

Anna, you're the perfect guest for helping us imagine and explore the possibilities when we give patients access to their own health data, and for exploring what it takes to earn the kind of trust to make that work.  

So let's open with a little of your background. What makes you so passionate about this work?   

Anna: I'm passionate about this work because I live this work. In fact, it was my life with type 1 diabetes that drove me in the direction of this work. I began my career first in journalism, then I did economic policy and foreign policy.  

And foreign policy didn't work out very well with my type 1 diabetes. So, I decided to switch careers, switch directions and get into healthcare. And over time did healthcare public affairs, global communications and ran the healthcare practice for two of the big global communications agencies. Over time I became more and more frustrated with research and stagnation in research.  

As all of the electronic health record data began to be generated, like 15 years ago, a friend and I decided to start a company to make it easier to explore incredibly complex EHR data in real time through a visual interface. That company really launched me into the world of health technology and health IT policy because I live in DC and have a background in public affairs.  

But I realized that many of the problems that I was experiencing as a patient and as a consumer were actually fixable. It's just that nobody had bothered to fix them. Whether that was with the lack of ability to access data from my diabetes devices and my medical devices, or whether that was getting access to my lab data when I needed it to really understand what was happening with my blood work and my kidney disease.  

It was all of these things that I suddenly realized these points of frustration were fixable, but nobody bothered to fix them.   

Melanie: Can you take us back to a moment in your own care, you have lots of stories, when you didn't get the data you needed? What did that reveal about the system and the problems that you say are fixable but nobody's fixed?  

Anna: Well, I'll take you back to two moments. One was with medical devices. I have type 1 diabetes. I use a continuous glucose monitor. I use blood glucose meters and an insulin pump. And this was probably about a year into my experience as a co-founder of a big data analytics company. I was having some difficult time with my diabetes, needed to go to the doctor and I was having an impossible time actually accessing the data and trying to download the data.  

The continuous glucose monitor that I used—I use an Apple computer—their data download process only worked with Microsoft computers. So I had to purchase parallels so that I would be able to download the data and view it. The software that both of these companies used looked like it was built in the era of the DOS prompt.  

It was horrible. I was infuriated that I had to purchase parallels and run Windows—to buy all new software just to be able to get access to my data. And again, at that point in my professional experience, I knew that all of this stuff was fixable. It was just that nobody was bothering to fix it.  

I started looking into the regulations: what are the company policies, reaching out to the company, going to different meetings, speaking up at different meetings with the FDA, getting involved in the FDA—both advisory committees and working groups around medical device data access—and speaking at different conferences of diabetes patients. I tried to use humor and empathy but also cajole the companies into actually doing the right thing.  

It took us a while to get to a point where we could really make change. But that's one of the things that really drove my interest in medical device data policy and what would ultimately become my contribution to the We Are Not Waiting movement, which we can get into in a moment.  

The second issue as it relates to electronic medical record data and particularly lab data is I have had type 1 diabetes now for several decades.  

I have all the complications from diabetes. I have eye disease, kidney disease, nerve disease and I try to keep on top of it. I had a nephrologist, this was many years ago at this point, who was an incredibly good nephrologist—very caring, spent a lot of time with me—but he was horrible about giving me access to my lab data.  

Absolutely awful. And, you know, I watch this stuff closely, and there are several key measures and biomarkers that you look at very closely in chronic kidney disease that tell you what the status is. I wanted to know what they were, and I could never get that data. 

It was absurd. It's my data. This is my health. I'm trying to do the right thing. The doctor won't return my phone calls. I can't access it directly through any lab system or through the hospital system. I just have to sit and wait for the doctor to call me back. And it just infuriated me that was the status quo and that it was completely acceptable at that point in time.  

Melanie: Wow. And so many people I'm sure, may not be in the exact situation, but a similar situation of not being able to access the data when they need to.  

You were among the founders of We Are Not Waiting. You just mentioned that hashtag, We Are Not Waiting is a patient-led grassroots organization that shaped access to health data—originally data from glucose monitors for diabetes management, right?  

Anna: The We Are Not Waiting movement—and I say this with all humility and knowing the small part that I have played in the beginning—I truly think it's one of the most impactful and important things to emerge in digital health in the time that I've been following it. And my role was very early. It was really focused on the policy side of it.  

It's a group of patients and parents of type 1 diabetes patients who became incredibly frustrated with the fact that we were not able to access our data in usable formats. It really began with two different devices that the data streams were not readily accessible through any mechanism that the manufacturer provided.  

A couple of the parents of children with diabetes and one patient were computer programmers who worked in software. They figured out how to hack into the receiver for the continuous glucose monitor. At that point, you couldn't download the information directly or access the data stream.  

But they hacked into the receiver and figured out how to connect it to an Android phone, send it to the cloud and pull it down on a different Android phone or a Pebble watch. And it was a truly life-changing experience for that parent who developed that system, a guy named John Costick, and others who adopted it quickly thereafter.  

For the first time ever, they could let their child go to school or soccer practice or sleepovers and follow their glucose from remote without having to be there hovering over the child. But they would know if something went wrong and they could call the soccer coach or the parent who was hosting the sleepover and say, "this needs to happen." All without having to be there constantly monitoring and standing over their child. 

The other device was an insulin pump that had a remote control so you could dose it without having to actually access the pump. It used radio frequency to do that. And another patient, named Ben West, figured out how to hack into that communications protocol so that you could access that data stream.  

Over time, rumors started circulating on Twitter and other places that these things had happened. I spoke at a meeting hosted by my friend Amy Tenderich, called the Diabetes Mine Innovation Summit. And I was the first to really talk about the need for data access, standard data formats and accessible APIs.  

These are all very fixable issues, it's just nobody bothered to fix them or point out different cases of consumer-grade products that were very easily able to send data and receive data. The FDA was at that meeting and medical device companies were at that meeting—that’s really when it began.  

Since then, a couple different versions of a closed-loop artificial pancreas system emerged. I'm using one that's called Loop, that is crowdsourced. It's all developed by the community, not the medical device manufacturer. The code is all coded by the community—so basically all parents and patients.  

The code is freely available on GitHub, and you can download the system onto your iPhone and set up the app yourself. It's all free. Last I checked, there are more than 30,000 people using Loop globally. It automates insulin dosing. If your glucose is dropping too much or too quickly, it'll turn it off. 

If it's rising, it'll dose insulin based off of your continuous glucose monitors. And it's all done through my iPhone—which serves as the brain and the connector between my continuous glucose monitor and my insulin pump.   

Melanie: That's amazing. And all patient developed?   

Anna: All patient developed—patients and parents.  

If there's an issue, it gets referenced and fixed immediately. And through social media, everybody gets an alert about what needs to be fixed and what they need to watch out for.   

Melanie: Where do you see the biggest gap today for what patients are entitled to and what they can practically access and use for data?  

Anna: Well, we still have a remarkable amount of challenges accessing all of the data from the clinical care system and the electronic health information that's generated in the clinical setting. The industry has sort of landed on this solution of patient portals.  

Patient portals are helpful—that was a major step forward and a really big battle that ONC took on, on behalf of patients that has been incredibly beneficial. But not all electronic health information is available in the portals, and that information is absolutely critical. A year and a half ago, I developed some vision issues that were urgent—a cluster of micro aneurysms in my right eye that were caused by long COVID and micro clotting from long COVID.  

I lost most of the vision in my right eye and couldn't see well at all. I was told that I needed urgent eye surgery by my ophthalmologist. I managed, somewhat miraculously, to get into the Wilmer Eye Institute at Johns Hopkins, which is one of the top eye institutes in the world.  

I got an appointment with the surgeon but wasn’t able to get the imaging from my eyes that I'd had taken every month for the prior 18 months. I wasn't able to get any of that imaging. They were telling me that I needed to contact the doctor. They sent me to medical records and they told me that I needed to write a check—a paper check—and send it to them in the mail. And when that check cleared, they then had 30 days to get me my imaging data. My surgery appointment was in four days. I was completely dumbfounded   

Melanie: Well, let's dive a little deeper into what it means for the patient to be in the driver's seat as you have been. The conventional view of healthcare innovation is that it's done for patients.  

You have pushed back on that. The better model for data, access, governance, AI—all of it—is to design with patients, not just for them. What does participatory design and governance look like in practice when the patient is the one who's driving innovation?   

Anna: There are a variety of different ways that you can include patients in co-designing different things.  

I'll give you an example. I work as an independent consultant, and I worked for 4 ½ years with a genetic testing company that had a remarkably good set of basic principles around data access and data sharing. They had four core founding principles, two of which were related to data: patients own and control their data, and that data is more valuable when it's shared.  

The company had been growing quite a bit. Those principles were something that they definitely supported, but there were aspects of it that hadn’t been operationalized. As they were moving towards broad scale commercialization of de-identified genetic data, they really wanted to do it in a way that didn't lose the trust of patients and providers.  

They came to me and said, "How can we do this in a way that we don't lose trust?" And I said, "Let's do better than that. Let's do it in a way where we actually earn trust." I created a framework for them around how patients can be a part of the data governance process—a framework I called my trust stack.  

This framework was sort of a multi-layered approach towards embedding patient interest, concerns and desires about what should happen with their data in all aspects of data governance. In that case, we created different mechanisms and tools for engaging patients directly in data use policies. And were very committed to transparency around data use.  

One of the things we did that I'd been trying to get companies to do for quite some time and was finally able to convince this company to do it, was to create an annual report on data use. Basically, it was an annual report on all of the ways that the company used de-identified genetic data that prior year.  

It showed a remarkable amount of incredible research, and how that data had been used to benefit patients by creating research into genetic variants and the impacts of specific variants on different forms of disease. And people were a little concerned about it. There was a lot of support for it because it was a really good company, but people were concerned. But when we released it, patients absolutely loved it.  

I'd had some discussions with patients before, and they were a little hesitant and were wondering, “How are they going to commercialize data? How are they going to do that ethically?” There were some serious concerns. But when I released the annual report that listed out all of the ways that the data is used, people were thrilled because their data had been used to benefit research and other patients as well as patients like them.  

Melanie: Sure. Well, that work has led you to work with HITAC. It's led you to a board seat with the Sequoia Project and to helping lead the Sequoia Project's consumer engagement work group. These organizations are helping to shape what's possible for patients with policy and governance and so forth. What does it take to make the patient's voice matter in those rooms?  

Anna: It's very challenging. Medical devices was a relatively straightforward and simple use case compared to electronic health records. It was something that we could just do and access, and the devices that we were hacking were our own devices. We weren't hacking into anybody's systems, we were hacking into the devices.  

All of your electronic health information is far more complex. The other thing is there’s a lack of a sense of urgency. Without a sense of urgency, there is no change. For the work that I've been leading with Sequoia, it's really around getting hospitals to rethink how patients access their data through the lens of user experience design and patient experience.  

Our objective was to get hospital systems and providers and the companies that support them to do it because it's the right thing, and to rethink the processes. We're not asking for major systemic changes. We're just saying rethink the way you present this stuff on your website. Make it possible to pay for data access the way you pay for your clinical services.  

Melanie: As we look ahead, what conventional thinking about patients and their data do you believe most needs to change?   

Anna: One is that portals are enough—they aren't. They're helpful in some ways, but they are not enough. They're certainly not sufficient, and they do not contain all of the electronic health information to which we are legally entitled. You can get reports, but that's not sufficient or particularly helpful. So that's one. Portals are not sufficient.  

Two, is that patients aren't accessing their data. They're not trying to get it as there are certain requirements for data being able to be downloaded to structured formats.  

Nobody is going around life every day when things are going great thinking about their health data. Nobody wants to think about their health data or accessing their health data, including me. You need it when you need it. And when you need it, you want it to be accessible.  

Third, is that the data can be and is confusing, but we can figure it out.  

And with the advent of artificial intelligence and LLMs, it's really easy to upload that data into the LLM and get really good, detailed and nuanced analysis of the data. I has a somewhat mysterious thing happening with my health earlier this year. I got access to the data. I uploaded it to three different LLMs and got different nuanced perspectives.  

Then I put it into a fourth, a specifically trained model for long COVID, and got an even more nuanced perspective—all of which were completely outside of the scope and capacity of my really good and caring physician.   

Melanie: If patient data access were truly solved tomorrow, what's the first change a patient or caregiver would actually notice?  

Anna: One of the things that's proved to be a somewhat intractable issue thus far that will probably require some policy changes, is the ability to access your data through mobile apps and personal health records. There's a variety of different complicated incentives and structures that have made it impossible for patients to be able to get any meaningful, regular or predictable access to their health data on their mobile phones.  

There are a variety of reasons that are cited as to why that data isn't flowing. They claim that it's patient protection—that it's potential violation of privacy. I don't necessarily think that's the case. At least I can't get access to my data through PHRs. So, there are reasons that are cited and then there are the reasons that are keeping the data from being accessible.  

If that data could be accessed, I think it would change a lot in terms of patient frustration. Also, it would make it easier for patients to be able to go to the physician with all of their data. I currently have to manually download all of my data and include bullet-pointed summaries for each of my doctor's appointments.  

You know, I have 19 different doctors. Last year alone, I had a 170 different doctor appointments, and not all of those doctors need that amount of data at every appointment. It's absurd at this point that we're still at that point in data access. Just having enough information to be able to have a reasonable discussion, an informed discussion during a doctor's appointment would be tremendous.  

Melanie: Well, as we bring this episode to a close, I have one last question for you. Considering the namesake of our podcast, what's one change you'd single out as the better way for improving how patients access and use their own data?   

Anna: One change. That's a hard one. I think what I was just talking about: make it possible to get access to your health data on your phone.  

We need to fix this. It's slowing progress. It's making life more difficult for patients. It's stopping innovative companies from being able to innovate and provide helpful solutions for patients and providers. And again, it's very easy to fix. We just must have a sense of urgency and a commitment to fix it.  

Melanie: Absolutely. Well, Anna, it was great to have you on the show. Thank you so much for sharing your personal stories, your professional perspective and the better way that we all expect for patients.   

Anna: Truly my pleasure. Thank you.  

Melanie: One thing came up again and again in my conversation with Anna. What's broken isn't always beyond fixing. Sometimes it's just that no one bothered to fix it. That's the same instinct behind #WeAreNotWaiting. Patients who hacked their own devices because they weren't going to wait for a manufacturer or a regulator to move first.  

So, patients climbed into the driver's seat and fixed it themselves. And today, an entire open source insulin delivery system exists because patients decided not to wait for permission. That's what happens when patients stop waiting, and the same principle holds on the other side of the table.  

Organizations that want to build patient trust can't just assume they have it. They have to earn it. When Anna advised a genetic testing company to publish an annual report that showed exactly how patient data had been used, the patients who had been the most skeptical ultimately became strong supporters, and that's because these patients could see for themselves how their trust had been earned and not taken for granted.  

That's the thread running through everything Anna does. Don't wait for permission to fix what's broken, and don't expect trust you haven't earned. Anna proves that with the patient at the wheel and the industry right alongside them, healthcare can move in the right direction  

Thank you for listening in today. If you've enjoyed this podcast, please rate, subscribe and review There's a Better Way: Smart Talk on Healthcare and Technology. With your help, we'll be able to continue to bring great conversations to the fore and to the wider listening public. Thank you.