Patient-Generated Data: How Your Health Information Will Be Managed in Fast-Growing Market of Wearable Medical Devices
By Lori Cameron
Share this on:
From fitness to remote patient monitoring to home healthcare, wearable medical devices are about to substantially improve the way you are treated by your doctor.
The market is expected to more than double to over $14 billion by 2022, however, leaving researchers scrambling to figure out how to coordinate and manage the massive amounts of data these devices will generate.
Toward that goal, the 11th EAI International Conference on Pervasive Computing Technologies for Healthcare held 23-26 May 2017 in Barcelona gathered technology experts, practitioners, industry stakeholders, and international authorities to discuss the development of pervasive medical-based technologies, standards and procedures.
Nuria Oliver, lead researcher on Data Science at Vodafone, Chief Data Scientist at Data-Pop Alliance, and General Co-Chair of PervasiveHealth 2017 discusses the goals of the PervasiveHealth 2017 workshop.
In their article “Harnessing the Power of Patient-Generated Data,” researchers from industry and academia discuss several key takeaways from the conference and outline a future research agenda that will harness the power of medical and health data.
During the breakout sessions, participants at the so-called PervasiveHealth 2017 workshop designed three scenarios that demonstrate several contexts in which PGD-sharing can be beneficial.
Scenario 1: Chronic Patient Care of an Older Adult with Low Motivation and Low Technology Literacy
Sarah is a low-income, 68-year-old woman living alone who suffers from depression and was diagnosed with Type 2 diabetes 15 years ago. She has limited insurance coverage and worries about her medical bill. Sarah is not familiar with high-tech devices and does not own a smartphone, though she does have Internet access. Her lifestyle is sedentary and she has low motivation to manage her health.
Sarah’s goal is to avoid diabetes symptoms such as extreme fatigue, blurry vision, foot problems, and kidney failure. Starting off slow, she can learn to track her blood-sugar levels, medication taking, diet, mood, stress, and physical activity to see more clearly how these factors might contribute to her symptoms.
Scenario 2: Casual Data Tracking for Later Use in a Clinical Setting
Oliver is in his mid-thirties and runs his own hair salon. He has a passion for healthy diets and fashion and loves tech. He has been exploring a broad range of activity tracking including heart rate, sleep, food, and mood as well as more unconventional things like gut bacteria and social activity.
Business has slowed and plunged Oliver into depression. The doctor suggests Oliver collect self-tracked data from devices and other sources like Facebook, and upload them to an informatician—someone who is skilled in health data analyses. Later, the doctor shows Oliver different examples of data correlations—high blood pressure during periods of low social interaction, high levels of blood glucose associated with bad mood, and increased use of mobile apps correlated with low concentration. He is diagnosed with mild ADHD and treated.
Scenario 3: Parents with a Newborn Baby Capturing and Sharing Health and Developmental Data
Jessica and John want to figure out how to feed their baby and train him for sleep. They are also anxious to know whether the baby is growing according to standard milestones or having any problems. So, in addition to logging daily feedings, diaper changes, sleep, and so on, as the pediatrician suggested, they install a baby monitor on the crib, allowing them to see their baby 24/7.
One of Jessica and John’s main challenges is to share the data and ask questions without overwhelming the pediatrician. To reduce the information overload while increasing the utility of the video data, they want data-capturing tools that help them create a succinct summary from the video. Once the clinician sees the benefit of the video data, she can suggest other markers Jessica and John should look for.
Each scenario underscores, not only the various ways PGD devices can be used by people with different health goals, but also the wide variety of data that can be gathered by them. They also demonstrate the various players involved—including doctors, patients, caregivers, and data analysts.
PGD research conference yields four takeaways
Supporting Clinicians’ Goals through PGD
Physicians might be reluctant to introduce their patients to PGD tech because of misperceptions about workload. To successfully leverage PGD in patient–clinician collaboration, more work is needed to better understand clinicians’ goals and desires by discussing their needs and goals when developing design requirements and solutions.
Facilitating Actionable Insight Gaining from Multiple Data Sources
Collecting personal data from multiple sources has become easier and more prevalent. However, these data can be scattered across many devices, apps, and platforms, making it difficult to get a good overview of one’s health and well-being. Despite the emergence of systems taking integrative approaches to data collection, such as AWARE and OmniTrack, and visualization, such as Visualized Self and Exist, further research is needed to find easier ways for lay individuals and clinicians to gain insights from multiple sources of data.
Cultivating Sustainable Data Collection
The quality of data plays an important role in data-driven communications. To ensure data quality, we need to better motivate patients to collect their data diligently. For example, creating a beautiful and unique artifact to express oneself using personal data can draw people to self-tracking. For example, Dear Data demonstrates that personal data can be visualized in beautiful, creative, and compelling ways.
Ensuring the Clinical Relevance of Collected Data
To make PGD useful for clinicians, we need to make it easy and not time-consuming to take action. There should be a mechanism in place to involve clinicians throughout patients’ data-capture process. In this way, clinicians can guide patients to see how certain data is relevant to their disease.
“In the future, tracking and sharing PGD might change the practice of clinical consultations as we know it: patients and clinicians have a data-driven medical consultation, improving patient engagement and speeding up the diagnosis. We hope to continue discussing ways to leverage PGD for better care with more clinician engagement, and encourage other researchers to contribute to future conferences including PervasiveHealth,” the authors say.
The researchers who wrote the study are Eun Kyoung Choe, an assistant professor in the College of Information Studies at the University of Maryland, College Park; Bongshin Lee, a senior researcher at Microsoft Research; Tariq Osman Andersen, an assistant professor at the Department of Computer Science at the University of Copenhagen; Lauren Wilcox, an assistant professor in the School of Interactive Computing at Georgia Institute of Technology; and Geraldine Fitzpatrick is Professor of Technology Design and Assessment at TU Wien (Vienna University of Technology).
Research related to digital health in the Computer Society Digital Library:
Lori Cameron is a Senior Writer for the IEEE Computer Society and currently writes regular features for Computer magazine, Computing Edge, and the Computing Now and Magazine Roundup websites. Contact her at firstname.lastname@example.org. Follow her on LinkedIn.