This document discusses the quantified self movement and mobile health applications in patient-centered care. It describes how wearable devices and sensors can track various health metrics like activity, sleep, mood and biological indicators. The data collected can help patients better understand and manage their health conditions. However, challenges remain regarding validating technology accuracy, establishing standards, integrating data with electronic health records, and addressing barriers to adoption. The quantified self has potential to empower patients but more research is still needed.
1. Patient-Centered Care
Activated Patients
Lecture b: The Quantified Self & mHealth
This material (Comp 25 Unit 2) was developed by the University of Alabama at Birmingham, funded by the
Department of Health and Human Services, Office of the National Coordinator for Health Information
Technology under Award Number 90WT0007.
This work is licensed under the Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International
License. To view a copy of this license, visit http://creativecommons.org.
2. Activated Patients
Learning Objectives
• Bring a global perspective to Do-It-Yourself (DIY)
medicine and describe the factors influencing its
expansion
• Discuss the impact of DIY medicine on both
clinical practice and clinical research
• Discuss the potential promise and peril of the
changes that DIY will bring to healthcare
• Discuss the role of the Quantified Self and
mobile health applications in patient-centered
care
2
3. Quantified Self
• A movement to incorporate technology into data
acquisition on aspects of a person’s life in terms
of inputs, states, and performance.
– Self monitoring
– Gamification
3
4. Wearable Sensors and Computing
• The combination of wearable sensors and
wearable computing has been known by many
names over time
– Life logging
– Self-tracking
– Auto-analytics
– Body hacking
– Self-quantifying
– Self-surveillance
– Personal informatics 4
5. Patient Uses
• Self-quantification may be used for DIY medicine
for personal purposes
– Monitoring chronic illness
– Understanding what affects daily function and
quality of life
• Self-quantification may also be used in
partnership with a clinician
– Challenges remain here including evolving
standardized/best practices
5
6. Research Applications
• Data analyzed via traditional techniques to
establish correlations among the variables
• Quantitative methods from business and science
applied to health data
• Data Visualization to generate hypotheses
6
7. M-Health
• 60% of US adults track at least one health metric
• Mobile health app usage growing
• Patients willing to share data to aid in
diagnosing and treating themselves or to aid
others
• Empowered by availability of devices and
change in attitudes toward tracking and sharing
data
7
8. Available Tools and Biosensors
• Mood
• Activity
• Sleep
• Biological
• Diet
• Other
8
9. Quantified Self - Mood
• Multiple apps to track mood
• Spectrum of users
• “Journaling” functions
• Social media linkages
• Collect aggregate data across users
• Examples
– Track Your Happiness, MoodPanda,
Moodscope, Moodjam, Optimism
9
13. Quantified Self - Other
• Microbiome analysis—uBiome
• Prompts to adjust posture—LumoLift
• Eating speed—HapiFork
• Prompts for activity--Belty
13
14. Using Data for Patient-Centered Care
• Self-quantification opens doors
– Individualized home monitoring
– Closer follow-up
• Potential issues
– How do we incentivize uptake
– Alert fatigue
– How do clinicians bill for such monitoring
– Insurance repercussions?
14
15. Quantified Self – Challenges
• Accuracy of devices
– Self-quantification experiment lack rigor and
controls
– Very few formal research studies exist
o Smartphones close to observations
o Wearable devices more variable
• Standards
• More rigorous research
15
16. Quantified Self – Challenges (2)
• Devices are not part of EHR
• Need for visualization tools
• Lack of device integration into Health IT
ecosystem
– No good mechanism for health information
exchange
– No bidirectional patient-provider exchange
16
17. Quantified Self – Challenges (3)
• Digital divide—disadvantaged populations
• Engaging the elderly
• Long-term utilization
– Gamification one potential strategy to
maximize engagement long-term
17
18. Activated Patients
Summary – lecture b
• Combination of wearable devices and wearable
computing
• Many devices and apps available
• Data used to optimize care and health
management
• Challenges still to be overcome
– Validation of technologies
– Bi-directional exchange of data with EHRs
– Addressing the barriers to adoption
18
19. Activated Patients
References – Lecture b
References
Accelerometer. Wikipedia. Wikipedia Foundation. Retrieved January 2016 from
https://en.wikipedia.org/wiki/Accelerometer
Activity Tracker. Wikipedia. Wikipedia Foundation. Retrieved January 2016 from
https://en.wikipedia.org/wiki/Activity_tracker
Altimeter. Wikipedia. Wikipedia Foundation. Retrieved January 2016 from
https://en.wikipedia.org/wiki/Altimeter
Case, M. A., Burwick, H. A., Volpp, K. G., & Patel, M. S. (2015). Accuracy of Smartphone Applications
and Wearable Devices for Tracking Physical Activity Data. JAMA: Journal of The American
Medical Association, 313(6), 625-626.
Castelao, L. (2012, March 3). The Quantified Self. The Economist. Retrieved January 31, 2016, from
http://www.economist.com
Cardiio. (n.d.). Retrieved January 31, 2016, from http://www.cardiio.com
EmWave2. (n.d.). Retrieved April 08, 2016, from http://store.heartmath.com/emwave2
Granado-Font, E., Flores-Mateo, G., Sorlí-Aguilar, M., Montaña-Carreras, X., Ferre-Grau, C., Barrera-
Uriarte, M., & Satué-Gracia, E. (2015). Effectiveness of a Smartphone application and wearable
device for weight loss in overweight or obese primary care patients: protocol for a randomised
controlled trial. BMC Public Health, 15(1), 1-6.
19
20. Activated Patients
References 2 – Lecture b
References
Guidi, G., Pollonini, L., Dacso, C. C., & Iadanza, E. (2015). A multi-layer monitoring system for clinical
management of Congestive Heart Failure. BMC medical informatics and decision making.
HapiFork. (n.d.). Retrieved January 31, 2016, from https://www.hapi.com/product/hapifork
Ithlete heart rate variability training tool. (n.d.). Retrieved April 08, 2016, from
http://www.myithlete.com/
Killingsworth, Matt. Track your happiness. Retrieved January 2016, from https://itunes.apple.com
Li, I. (n.d.). Moodjam. Retrieved April 08, 2016, from http://moodjam.com
Lumo Lift - Posture Coach & Activity Tracker. (n.d.). Retrieved April 08, 2016, from
http://www.lumobodytech.com/lumo-lift/
Marceglia S., Fontelo P., Rossi E., Ackerman M.J. A standards-based architecture proposal for
integrating patient mHealth apps to electronic health record systems. Applied Clinical Informatics.
Retrieved April 11, 2016, from http://www.ncbi.nlm.nih.gov/pubmed/26448794
Moodscope - Lift your mood with a little help from your friends. (n.d.). Retrieved April 11, 2016, from
https://www.moodscope.com/
Optimism Mental Health Apps for Self-Tracking. (n.d.). Retrieved April 11, 2016, from
http://www.findingoptimism.com/
20
21. Activated Patients
References 3 – Lecture b
References
PWC Health Research Institute. Top health industry issues of 2016. Retrieved January 2016 from
http://www.pwc.com
“Quantified Self.” Wikipedia. Wikipedia Foundation. Retrieved January 2016 from
https://en.wikipedia.org/wiki/Quantified_Self
Sequence Your Microbiome - Gut Flora, Microbiota. (n.d.). Retrieved April 11, 2016, from
http://ubiome.com/
Sleep Cycle alarm clock on the App Store. (n.d.). Retrieved April 11, 2016, from
https://itunes.apple.com/us/app/sleep-cycle-alarm-clock/id320606217?mt=8
Track anything, collect data and gain insight over time. (n.d.). Retrieved April 11, 2016, from
https://www.mercuryapp.com/
uBiome. Retrieved January 31, 2016, from http://ubiome.com/
Withings. Retrieved January 31, 2016, from http://www.withings.com
21
22. Patient-Centered Care
Activated Patients
Lecture b Quantified Self & mHealth
This material was developed by The
University of Alabama at Birmingham,
funded by the Department of Health and
Human Services, Office of the National
Coordinator for Health Information
Technology under Award Number
90WT0007.
22
Hinweis der Redaktion
Welcome to Person-Centered Care, Activated Patients. This is lecture b. In this lecture, The Quantified Self and mHealth, we will learn more about Activated Patients through the usage of mHealth and wearable technologies.
The objectives for this unit, Activated Patients, are to:
Bring a global perspective to Do-It-Yourself (DIY) medicine and describe the factors influencing its expansion
Discuss the impact of DIY medicine on both clinical practice and clinical research
Discuss the potential promise and peril of the changes that DIY will bring to healthcare
Discuss the role of the Quantified Self and mobile health applications in patient-centered care
Quantified Self is a movement to incorporate technology into data acquisition on aspects of a person’s life in terms of inputs (e.g., food consumed, quality of surrounding air), states (e.g., mood, arousal, blood oxygen levels), and performance (mental and physical).
Self monitoring often combines wearable sensors and wearable computing for data collection.
It may be conjoined with gamification approaches that allow everyday activities to be turned into games rewarding points to encourage people to compete with friends.
It is, in many ways, a natural extension of the trend of patients taking more responsibility for their own heath and reflects patients now collecting data that may have been previously collected by health professionals.
The combination of wearable sensors and wearable computing has been known by many names over time. Lifelogging, self-tracking, auto-analytics, body hacking, self-quantifying, self-surveillance, personal informatics, are all forms of the quantified self using wearables.
Patients often collect data using self-quantification devices in the service of DIY Medicine. Patients may use these data to monitor a chronic illness and to ascertain the impact of their daily activities on such an illness as they seek subtle clues on how to improve their functional status and quality of life.
Such activated patients often seek to share and discuss these data with their clinicians. Yet in an industry where standardized practices are still being formed, it is sometimes difficult for clinicians to engage with patients on the optimal way to interpret and act on their data.
Data that are collected by wearable sensors (or data that are logged manually) can be analyzed by researchers using traditional techniques such as linear regression to establish correlations among the variables. In essence, these innovative technologies are applying quantitative methods used in science and business to personal health data, often with the goal of detecting opportunities for health improvement. Data visualization techniques can also suggest hypotheses that can be tested via more rigorous methods.
It is estimated that nearly 60% of US adults track at least one health metric.
Mobile health application adoption doubled from 2013 to 2015 in the HRI Consumer Survey by PWC. In regard to sharing data, 83% of respondents were willing to share data to aid in diagnosing and treating themselves while 73% were willing to do the same to aid others.
Shifting attitudes to data tracking and sharing, readily available wearables connected to networks (smartphone or otherwise) empower this growing movement
With the current technologies, there are multiple aspects of human activities that can be tracked and monitored. We will now explore some facets of the quantified self movement and currently available tools and biosensors across the following domains:
Mood
Activity
Sleep
Biological
Other
There are applications that allow patients to track their mood for a variety of reasons. There is a spectrum of users from those seeking to understand or analyze factors that influence their mood for self-discovery to those diagnosed with mental health disorders like depression.
Several apps allow users to “journal” about their mood either by answering questions or logging data.
Many allow users to share mood data over social media and even have some interactions with other users, for example MoodPanda allows for sharing of “virtual hugs”.
Some apps seek to pool user data to glean insights into factors associated with happiness with the goal of improving this outcome in the general population
Many available, some examples include
Track Your Happiness, MoodPanda, Moodscope, Moodjam, Optimism, etc.
Activity trackers are wearable devices that monitor and record a person’s fitness activity and biologic response. The use sensors to calculate mileage, caloric expenditure, physical activity, heart rate, temperature, etc. Sensor examples include: pedometers for counting steps, accelerometers to detect acceleration, altimeters which measure the altitude of an object above a fixed level, and GPS for geolocation. Many commercial devices for sale include companion apps to track, visualize, analyze and share data. Gamification may also be used to promote competition with the goal of enhancing use.
Available examples include: FitBit, Amiigo, Jawbone, Strava, RunKeeper, Pebble, Apple Watch.
The overall goal of sleep monitoring sensors is to monitor and analyze sleep patterns, providing insight into ways to optimize sleep and waking. Multiple products use different approaches from smartphone sensors to separate alarm clocks to monitor sleep cycles have been utilized.
Several products have been discontinued like Zeo and Wakemate, while sleep specific apps continue to exist (e.g., SleepBot, Sleep Cycle), and sleep monitoring has been included in many activity trackers, for example , FitBit.
Biological tracking can involve any physiological output as long as there are sensors on devices that can capture the relevant data. Data can be captured with components available on smartphones; as part of independent sensors or as unique combinations of both. Some common biological tracking apps collect data on heart rate, blood pressure, and weight.
Cardiio uses a smart phone’s camera to detect heart rate. With each heartbeat more blood is pumped into our faces which changes how much light is absorbed and reflected back to the camera. These subtle changes are detected by the app and converted to an accurate heart rate.
Emwave2 is a separate device with sensors for respiratory and heart rate that allow their correlation with stress and encourage interventions to decrease it.
iThlete uses a chest strap or finger sensor that transmits data over bluetooth to the device that runs the app to visualize heart rate and help with training.
Withings combines a watch which tracks activity and sleep and a scale that wirelessly transmit data to a smartphone. The Withings app visualizes the data and allows sharing of data with friends.
MyFitnessPal is an app where one can easily enter foods consumed. It keeps track of prior entries and builds lists over time to facilitate subsequent data entry. Using data from MyFitnessPal or any such associated caloric intake and weight tracking apps allows users to quantify, monitor and share their caloric intake and promote weight loss. There are also multiple podcasts and web communities such as The Quantified Body dedicated to self-quantification of different diets and DIY approaches to weight loss and improving health.
Many examples exist of remote monitoring of chronic diseases such as congestive heart failure. Data are typically gathered by sensors at the patients residence and communicated wirelessly to a central location where it is visualized by providers who are sometimes alerted to issues via built in clinical decision support.
Such systems may spread further as the shift to accountable care organizations expands the emphasis for effective home monitoring solutions for early detection or exacerbations of chronic illness.
The opportunities for quantifying different facets of our lives will only grow as new sensors to collect new types of data become available. Some companies are providing quantification without sensor devices. For instance, there has been growing interest in research in the microbiome (pronounced micro-by-ome). The microbiome refers to the genome of the microbe population of parts of our anatomy (nose, gut, genitals, etc.). These microbes may perform essential functions (for example, in the gut they digest food and synthesize vitamins. uBiome allows sample swabs to be mailed in, additional survey data collected, the company sequences the users’ microbiome and users can compare their data to others.
LumoLift has a sensor that attaches to clothing near the collarbone for postural and activity monitoring. It vibrates if the person wearing it is slouching to remind them to sit differently.
HapiFork includes sensors in a fork to provide feedback on eating speed and Belty is a belt buckle sensor that adjusts at mealtimes and signals for activity after long sedentary periods. What is interesting about these devices is they both track data and provide feedback and prompts based on the data.
Self quantification opens the door to home follow-up at a hitherto unimaginable scale. Patients may capture and share more information than never before on how their condition is doing at home. Home monitoring technologies enable patient-centered care with previously unavailable monitoring, early detection and intervention possibilities that enhance patient morbidity (less severe exacerbations of chronic illness), cost and potentially mortality.
A looming challenge is how to integrate them into clinical care in a non-disruptive manner? How do we integrate these data in a way that does not cause alert fatigue which may lead to clinicians ignoring important alerts. How do we incentivize the uptake of these technologies and how do we determine and establish fair compensation for clinician time? Unanticipated impact on patients must also be considered – will those who complete newly monitored home goals (compliance with exercise or diet) be treated the same by their insurance providers or will these lead to financial penalties?
If patients are going to be using these devices and potentially sharing the data with their clinicians, one needs assurance that the data are accurate. Unfortunately, self-quantification experiments lack the rigorous controls and methodology of traditional research. Such controls might help mitigate placebo effects, which could be a significant factor. Scant data exist comparing these devices and their accuracy. Case and colleagues published a study in JAMA in 2015 that showed the variability in step count trials for 500 and 1,500 step trials across various commercially available wearable devices and smartphone apps. They found consistency between the 500 and 1500 step trials, and generally found the smartphone apps to be very close to the observed step count. However, the wearable devices showed the most variability.
Standards for scientific validation/comparison of these technologies are not fully formed and are needed to identify best practices. Increasingly randomized controlled trials to validate them are underway
One of the major problems is that most of these technologies exist independent of electronic health records and cannot easily be imported into them even if the clinician and patient were interested in having them be part of the record. In addition even if the raw data could be imported, there would also need to be good visualization tools within the EHR to make the data interpretable. Although the devices and apps often have such tools, they may not transfer to the EHR.
Their integration into the larger health-IT ecosystem and bidirectional data exchange between patient and providers remains limited and is an opportunity to maximize the impact of these tools and involve individuals more closely in their care.
Another challenge relates to the digital divide. Significant sections of the population either do not have access, or do not have the health and computer literacy to use these devices. How will we bring the benefits of these new technologies to economically disadvantaged populations who may not have access?
While the digital divide usually refers to disadvantaged populations, the elderly are another population group that can benefit greatly, yet may not use the devices and apps for self-monitoring. How do we bring the benefits of these technologies to elderly patients who often lag in regards to their adoption when compared to younger Americans?
Once in the hands of users, how will we ensure long term user engagement in the capture and sharing of data. Gamification has been proposed as one strategy, we need data to further understand how to maximize longitudinal utilization.
In summary, the Quantified Self refers to self-tracking of activities through a combination of wearable devices and wearable computing. There are a growing number of applications to measure a wide variety of body
Data captured in this fashion can be used by both individuals, clinicians and healthcare systems to optimize care and health management
Many challenges remain including, establishing robust methodologies to validate these technologies, the bi-directional exchange of data with EHRs and establishing how to bring the benefits of these technologies to all individuals regardless of socioeconomic or other barriers to adoption.
