One of my mentors in diabetes, Dr Umesh Masharani at the University of California, San Francisco, trained me to teach people with diabetes to "become your own scientist." He tells his patients, "You are the player. I am the coach."
Clinical studies have shown that measuring random fingerstick blood glucose (BG) levels without any purpose has no impact on diabetes control. But when people measure BG levels to learn something concrete—to test the effect of 5 units of rapid-acting insulin when eating scrambled eggs and a baguette, for example, or to test the effect of a 30-minute walk around the park—only then do the BG measurements matter. In other words, people are best served by following the scientific method of inquisition into action and response, cause and effect. They can then review these data to develop a diabetes self-management toolkit full of learnings from their personal responses and patterns, augmented by interactions with their diabetes healthcare team.
One significant challenge is discerning how to help people draw their own meanings and conclusions from their data. If they passively hand their data over to their clinicians and receive direct instructions on what to do, the interaction is less useful, and they are less prepared for the real world of daily self-management decisions.
The growing availability of digital data has made "data visualization" a prominent tool. Turn on the TV news for a discussion of an election poll, or comparison shop the Web to buy a car, and you will see colorful graphs and charts trying to communicate information.
Over the years, the display of diabetes data has progressed from hand-written paper logbooks to computer spreadsheets, vendor-developed software applications, and more recently third party-developed software programs. All share the same basic goals: to help patients track their diabetes data, understand previous diabetes management decisions, and improve these decisions in the future. These tasks have become more and more difficult as devices such as insulin pumps and continuous glucose monitors have begun to generate more and more data points.
Holding Data in the Palm of Your Hand
Recently, I was given a first look at a fresh take on BG monitoring, created by artist and technologist Justus Harris. Justus is an artist and technologist based in Oakland, California, and Chicago who was diagnosed with type 1 diabetes at age 14 years. He is blazing a new trail by bringing together personalized health data and the tactile world, creating 3-dimensional (3D) data visualization sculptures.
Often the most exciting innovations happen at the intersection of disciplines, such as the intersection of art, technology, and medicine. Sitting at this crossroads, Justus created an object that can be viewed through many different lenses. As an art gallery piece, it is a form of self-expression for someone struggling with a chronic disease, using art to humanize the BG numbers that he lives with on a daily basis. As a medical education tool, it is a very creative method that connects a patient with his BG data in a 3D, tactile, and visual fashion. This approach could even become a way for someone with low health literacy to rapidly understand whether his diabetes is in good or poor control, beyond what glycated hemoglobin conveys.
Justus was underwhelmed by the current data visualization tools for diabetes, seeing them as simply digitization of legacy analog tools (like the BG logbook). He looked instead for a paradigm shift. While the new generation of data visualizations may be more elegant, he believes that in the end, they still represent numbers and drawings on a page. Justus felt that he was lacking a more personal and tactile connection to understanding the vast amount of data that he was generating every day through fingerstick BGs, his continuous glucose monitor, and his insulin pump.
Gateway to Many Possibilities
As an informatics professional and endocrinologist, I had an almost visceral reaction to the sculpture when I first held it. It gave me a new perspective on how someone might relate to their health data using visual and tactile senses in a way I had never considered. This sculpture has allowed Justus to directly interact with his BG data in a way not previously possible. By scrolling his fingers around the sculpture, he can feel the patterns of his BGs (in a vast variety of ways, even comparing every Tuesday of the past month, through touch alone). He can look at a particular week's data and immediately get a sense of whether or not he met his personal goals.
Justus and I discussed how a data sculpture might be used in a real clinical practice. My first reaction was that it may not provide a physician with the density and format of data needed to help patients with type 1 diabetes tweak their insulin dose regimens. However, he believes that this sculpture is only the start, and I agree that there are many future possibilities.
In current diabetes practice, our pattern recognition abilities remain far from perfect, so we should only welcome new ideas and modalities. I immediately imagined the next iteration of the sculpture having two data points per day rather than one, a high and low BG of the day as opposed to the daily average.
But even if future iterations of data sculptures do not find their way into every clinician's office, they may still prove to be useful clinical tools. People with type 1 diabetes unfortunately must live and breathe the challenges of "big data" every day. They collect, track, and relay large amounts of data to their providers on a regular basis and are asked to adapt their therapies based on what they find. It is a 24/7/365 job, and we are far from finding the answer about how to best approach this process.
Challenges and Aspirations for Empowering Patients
In addition to the data challenges, people with diabetes are on a personal journey of how the disease weaves into their day-to-day lives. Current data visualization methods may not be able to capture that in a way that a 3D sculpture might. If this new medium enables a larger population of people to interact with their own data, that is something special and critically important. Empowered patients interact with their doctors at a higher level, raising the bar of care.
Although Justus knows how to use mathematical modeling, 3D rendering, and 3D printing, not everyone has these technical skills. Therefore, he hopes to create a set of design tools for the average person, allowing them to create their own health abstractions and tactile interactions. His goal is to "empower people to become experts about themselves," which is precisely what is needed to improve diabetes self-management.
Justus is taking an important first step toward helping people create artifacts that they can connect with on a human, emotional level; at the same time, he hopes to provide scientific and analytical purpose. Justus's work is enabled by a new philosophy of personal empowerment and open data in healthcare. Until just a few years ago, his sculpture would have required him to manually keep a BG log (or otherwise do some computer hacking) in an Excel spreadsheet in order to have access to his own BG data in a flexible format. Devices could be downloaded but only into proprietary, inflexible formats.
Many efforts like Tidepool, an open source, nonprofit "steward" of data apps, are helping to lead the push toward increased health data accessibility by consumers, moving away from the "vendor owns the data" model. Justus's work is an example of the endless creative possibilities when you give people access to their own data.
"Design thinking" has captured the attention of Silicon Valley as well as the digital health community. As we look ahead to the future of data visualization for diseases that have a large data component, top-notch programming skills may be necessary but no longer sufficient. We need to start thinking differently about design, technology, and data visualization for our patients and providers. This may require us to branch out from comfortable spaces. With a small yellow piece of 3D printed plastic, born from this design thinking mold, Justus is challenging current paradigms.