The Quantified-Self (QS) movement is defined as “self-knowledge through numbers.” Participants analyze their daily activities, which includes anything from their meals and exercise or when they listen to music or feel happy. Goals range from simply sharing data to initiating data-driven behavior change and even self-experimentation.
The self-tracking trend of QS originated from LifeLogging, and those involved in the QS community use data collected through various devices to gain insight into the true outcomes of their behaviors. Although self-tracking is still the province of early adopters, this trend is edging into the mainstream and positioning itself as a growing feature of health care.
The trouble with the traditional patient:
Medical diagnosis is reached through subjective (history) and objective (physical exam, labs, and imaging) information, considered in light of prior probability. Such information processing is the foundation of medical practice, and, like all complex systems, it has challenges and limitations.
Diagnosis hinges on the medical history. Unfortunately, the subjective nature of the information can pose a challenge to obtaining reliable data. Human memory is more a typewriter than hard-drive. As patients reassemble memories rather than retrieve them from a stored format, their mental condition and emotional state skew their recall. Obtaining an accurate and reliable medical history is cumbersome, repetitive, and ineffective. Unlike lab work and imaging, the subjective data required for diagnosis is difficult to retrieve, inaccurate, and riddled with narrative fallacy; an illusionary correlation of constructing a story around facts.
Though more efficient and reliable, objective data poses challenges of reliability (varying sensitivity and specificity), as well as the difficulties of separating signal from noise; more information is not always better. Furthermore, most tests provide snapshots in time (a blood level or CT image) rather than a temporal panorama that better informs the context of such information.
Founded in 2007 by Kevin Kelly and Gary Wolf of Wired Magazine, QS is a global collaboration of users and toolmakers interested in tracking personal data. This allows individuals to explore how their bodies function and find the smallest possible changes that can produce the most effective results. The technology of self-tracking is rapidly progressing and becoming a relevant component of digital mobile devices, consumer medical devices, and social media. The Robert Wood Johnson Foundation has awarded QS a grant because “the movement represents a major transformation that has the potential to impact not only how we understand ourselves, but also how we relate to institutions and professional experts.” The implication is that well-informed patients will present to their physician with personalized data that can simplify diagnosis and treatment.
Mobile devices have become ingrained in our society, and thanks to these devices we are able to function in systems of greater complexity and interact with a continuous flow of information from the internet. Mobile tracking apps and devices will allow us to overcome limitations in the way we retain information, communicate, and collaborate.
Having constantly monitored patient health data may allow physicians the opportunity to integrate information from patient’s genome, proteome, and environment to promote health and prevent illness. Currently, QS users are aggregating in communities to better understand the specifics and implications of their personal data, as well developing collaborative solutions.
Many QS patients embed their tracking data in social media services like CureTogether and PatientsLikeMe. These simple and free websites allow individuals to track and share their treatments and symptoms, thus fostering patient-driven research and even hosting open-source clinical trials. A patient-driven health care model has been characterized by the patient empowerment, increased information flow, transparency, customization, collaboration, and patient responsibility. Although burdened by cognitive bias and systematic error, these communities are focused on taking the necessary precautions and learning the appropriate tools for their evaluations. Such crowd-sourcing may leverage large populations to find novel associations worthy of further and more rigorous investigation.
The current QS technological landscape:
The QS community has organized The Complete QS guide to Self-Tracking, which details the field of wireless devices, innovative tests, and mobile health care apps that will revolutionize the use of diagnostic information at the patient’s bedside. Technology is becoming more affordable and many consumer self-tracking products are already on the market. Consider four products currently in use:
FitBit is widely considered by many in the QS movement to be the best self-tracking device available. The device’s signature attraction is its simplicity. Designed as a smooth clip that can be unobtrusively attached to clothing or carried in a pocket, the FitBit counts steps and calculates calories burned, including a rough algorithm for calories consumed while inactive. This data is uploaded to a web profile and displayed in chart and graph format, to which one can add food consumed and receive an energy balance for trimming pounds. Additional functionality includes customizable goals, journal spaces for qualitative comments, and a social space where you can compare progress with your friends and family. The FitBit also serves as a sleep monitor that, when strapped to your wrist, can report on duration and quality of sleep based on how much you toss and turn.
FitBit uploads wirelessly to a USB hub whenever it is within a 15 foot range, and can maintain its charge and store data for over a week. It can be worn for long periods of time and costs about $100. There is no mandatory monthly fee for the web service. The primary drawback is the risk of losing the device as it can slip off clothing or fall out of its clip holder. Additionally, the FitBit cannot record activities like cycling or swimming. Although FitBit supports a mobile-friendly site where you can input data, it lacks the always-on functionality of a robust app.
BodyMedia Armband is similar to the FitBit in principle: it logs calories burned vs. calories consumed, monitors sleep, and features web displays. The difference is that, in addition to counting steps, it targets accuracy by utilizing galvanic skin response and temperature variation to capture the caloric burn of activities like folding clothes or doing yard work.
Another feature that sets BodyMedia apart is its design as a body area network that will operate as a hub for various future BodyMedia devices, such as heart monitors, that would sync to the user’s phone via Bluetooth. It also monitors sleep and allows for data input to calculate an energy balance. The future of a body area network looks bright, especially if this generic health data can be meshed with vital signs and cardiac data from other services like AirStrip Patient Monitoring.
The Zeo Personal Sleep Coach readjusts the focus on simplicity: it only monitors sleep. The user simply straps on a headband before bed, evaluates data in the morning (with a numeric sleep score), and is then free to view their data displayed on a web profile, including sleep coaching for better rest. Zeo actually measures brain activity, rather than simply gross body motion, to get a sleep analysis that was previously the special province of research labs.
It is crucial to note that the Zeo technology has been validated in scientifically controlled studies by comparing it to full in-lab polysosmnography. Sleep duration and quality are assessed with analysis of the different stages of sleep, providing the user with crucial feedback for optimizing their sleep hygiene. The Zeo appeals with its powerful analysis, trackable data that can motivate training for better sleep, and a free coaching service that emails tips.
23andMe allows the user to supply DNA by mailing a saliva sample, which is processed and, within weeks, displayed to the user as a web profile. The data is broken down into two main features. The first, “My Health” uses the growing library of markers for genetic predispositions to various diseases. A lot of behind-the-scenes statistics are condensed into reports on the user’s average disease risk as a percent compared to the average, and a confidence rating of this data. Additionally there is a display of carrier status for many genetic conditions, predisposed drug responses, and descriptions of traits.
The second feature “My Ancestry” offers a geographic distribution of the user’s heritage for both parents, a relative locator, the ancestry of individual chromosome segments, and quite frankly, far more genealogical data than one could have guessed existed.
The single largest problem facing widespread genomic mapping services such as 23andMe is a significant signal-to-noise problem as the wealth of information is poorly understood even by experts in the field, and there is almost no guidance for the kinds of behavior change or treatments that are appropriate for certain results. This is especially relevant to issues of treating the worried well. According to endocrinologist and author Dr. Clifton Meador, “The increase of well people seeking medical care lowers the prevalence of all diseases and increases the rates of false diagnoses.”
Stay tuned for Part 2 of this article where we discuss Health care 2.0, Physician 2.0, Med Student 2.0, and QS Skepticism.
Follow the authors on Twitter!
Igor Irvin Bussel (Rosalind Franklin University of Medicine and Science) – @irvinbussel
Aaron Stupple (SUNY Upstate Medical University) – @astupple