The 2018 Quantified Self Symposium: The Future of Self-Collected Data and Single-Subject Research
I recently participated in the 2018 Quantified Self Symposium on Cardiovascular Health at the University of California, San Diego. I hope this summary will capture the inspiring, exciting and pioneering nature of this event (I have rarely used these three adjectives combined to describe an event!)
As a quick reminder, the mission of the Quantified Self is to support people in making personal discoveries using everyday science; it has been shown over the last few years that individuals can make significant discoveries about their own health using self-collected, high temporal resolution data from open tools.
The goal of this annual event is to celebrate the community involved in these efforts, as well as explore how the work of these pioneers can eventually become mainstream.
How do we know that individuals can make consequential discoveries about their own cardiovascular health using self-collected data?
Larry Smarr, a physicist, avid self-tracker and leader in scientific computing at the University of California San Diego, kicked off the event with a beautiful reminder of why it important for us to collect data about our own health:
It might seem obvious that the human body is a set of multi-component non-linear systems developing in time. However, what is still not that obvious (yet) to most people is that the most logical way to understand this system is to measure it over time with multiple variables.
The good news is that the underlying technology that is enabling us to read our bodies is getting exponentially cheaper. This obviously goes far beyond devices like Fitbit, and includes practically everything from the microbiome to MRI data.
As a result, more and more people are (and will be) able to collect data on their bodies, helping build a longitudinal, comprehensive view of their health. This shift will eventually help us move to a more preventative, pro-active healthcare system.
Susannah fox, formerly the CTO of the U.S. Department of Health and Human Services, reminded us that five years ago, a study showed that 70% of Americans engaged in some sort of self-tracking. However, "about half of them were actually doing it in their heads." It would be interesting to see how these numbers might have changed today, especially that data collection technology has (and will) become more and more available to everyone.
Why has N-of-1 research never had any major impact on medical practice?
Randomized Control Trials, which aim at telling us whether a specific treatment works on average, are widely used in medical practice. The problem is that if a treatment is not working on a patient we have in front of us, it does not really matter that it works best on average, said Reza Mirza, a resident doctor at McMaster University.
Reza gave a couple of examples of how these conventional trials have been used since the sixteenth and seventeenth centuries (!) More recently though, a few pioneer physicians have started to rely on N-of-1 experiments to inform their decision making process. However, these remain isolated cases. Reza explained that one reason might be that it is a lot harder to organize an N-of-1 study than it is to do a conventional trial.
Hugo Campos: 10 Years with an Implantable Cardiac Device
Hugo Campos has been at the forefront of the battle to give patients with implanted cardiac devices access to their medical data. In July 2015, he was honored by the White House as a Champion of Change for Precision Medicine for his data liberation advocacy.
Hugo explained how he correctly diagnosed his Atrial Fibrillation (AFib) through self-collected data (i.e. with his Alivecor device). He went to the emergency room and informed the ER doctor that he needed a cardioversion (i.e. an electroshock). He couldn't get the data/alert from his implanted defibrillator, but he eventually received the treatment.
Science has to “meet patients where they are.” - Hugo Campos.
Lessons from a participatory, high-frequency blood testing project
(a more detailed description of this project is available here).
A few months ago, a group of 21 Quantified Self enthusiasts (including myself) started a new collaborative project called “Blood Testers.”
The reasoning behind this project was that, in most Quantified Self projects, one person does almost all of the work, perhaps with a bit of advice from friends and online feedback. But what if we could work in a group of people with varied skills to explore questions we developed through conversation and collaboration?
Our immediate goal with this experience was to learn more about ourselves through high-frequency self-testing of blood lipids (i.e., cholesterol and triglycerides). Our long-term goal was to help advance self-directed research by better understanding what makes these types of projects succeed or fail.
A detailed summary of my experience with the project is available here.
A few participants have presented their findings at the Symposium. Perhaps one of the most interesting lessons was that a single measurement, taken at the doctor's office at a single time of the day, month or year might not be telling the whole story.
We all learned that our cholesterol (including fasting measurements) varied a lot based on different factors that include the food we've had before (Jana beck presented her experience with low-carb and high-carb diets), the physical activity we've done, not to mention our circadian rhythm and (if applicable) menstrual cycle.
Whitney Boesel, a writer, researcher, sociologist and DIY medicine enthusiast, presented her experience with high-frequency measurements of morning fasting cholesterol throughout the month, including nine, 10 and 14 months postpartum.
What's Next for Self-Collected Data and Single-Subject Research
The rest of the event focused a lot on the barriers that stand in the way of using personal and public data for understanding and improving individual cardiovascular health. One of the points that came up a lot is the lack of consensus about the "legitimacy" of self-initiated research and self-collected data.
As a person who collects data as a hobby (and spends the rest of the time helping people collect their data), I believe that the rise of self-collected data among consumers and patients will help overcome this perception. As Marcel van der Kuil, Sports & Health Tech entrepreneur, pointed out, as more and more people become aware of the importance of self-collected data, their combined force will drive change.