MEET: Catherine Bass, Ph.D., director of informatics at Onlife Health in Nashville, Tenn.
THE CHALLENGE: The Centers for Disease Control and Prevention (CDC) is constantly collecting and analyzing data that can be used by health professionals, government leaders and the general population to guide public health policy, programs and action.
But collecting this data through traditional telephone and in-person interviews is becoming more challenging and costly due to declining participation and changes in the way people communicate. At the same time, new technologies and data sources—such as wearable devices, apps and social media—offer an opportunity to improve the reliability and usefulness of health-related information.
The Centers for Disease Control and Prevention (CDC) encouraged innovators to demonstrate new ways to address the challenges and limitations of self-reported health information and unleash the potential of new data sources and methodologies for public health surveillance.
THE PRIZE: $45,000
THE SOLUTION: Catherine and team’s proposal, “Closing the Loop: Augmenting Mobile Data Sources for Public Health Surveillance,” aims to improve the quality and quantity of health behavior data collection by aggregating personal wearable device and mobile health app data, and augmenting it through user validation and survey responses.
The solution successfully demonstrated the collection of data related to nutrition, sleep, physical activity and sedentary behavior, and included a series of short, just-in-time survey questions delivered through push notification to supplement mobile data sources.
The system cross-validated the collected information from mobile and survey data sets, as well as with the Behavioral Risk Factor Surveillance System (BRFSS), which tracks health risks in the United States through a series of telephone surveys.
IN HER OWN WORDS …
How did you hear about this challenge?
I heard about the challenge through our product and marketing team. They will often seek out opportunities such as these as a way to keep up with industry trends and to look for unique ways to innovate.
A very basic way to look at this challenge is to see it as an attempt to find new technologies and approaches that can invigorate a 360-degree stream of health data — from the people to health experts and back to the people. Where does Onlife Health fit into this cycle and how does the solution you proposed for this challenge improve the collection and flow of information?
Onlife’s solution puts real-time data in the hands of the individuals who are making decisions about the health care of the nation. Rather than having to wait a year or more to collect data, analyze it and understand trends that drive decisions, this system allows for the collection of data while it’s happening so patterns can be identified more quickly.
Additionally, our solution is improving the quality of the data by providing more context around each of the measures. It provides a survey component that allows for collection of data that cannot be measured through devices. Coupling survey data with validated device data provides a comprehensive view of a person’s health behaviors.
A major shift in the collection of health data in recent years has been the proliferation of wearable and mobile devices. How does that help or complicate the process and how does your solution address these issues?
Wearable devices have revolutionized the world of data collection by making objectively collected data and self-reported data available in a relatively normalized format on a grand scale.
There are some barriers to overcome with device data, which we strove to address with our solution. When thinking about how an individual may use the device, device data can be incomplete, inconsistent and lack context. Our solution is designed to supplement mobile data sources through the use of validation grids. These grids prompted the user to confirm, edit or supply new information to the data that was tracked with their wearable device.
This additional data filled in the gaps in what we were able to see when looking at the data on its own.
Who provided the data through your solution? Did people sign up? How many?
For the challenge, people downloaded our app through the app store and connected their device to the app so we could use their device data with our validation grids. We partner with a data aggregator, which allows people to use whichever device they prefer.
We had 262 people download the app and 154 complete the whole process, including a survey and use of the validation grids.
What benefits did the open challenge concept provide for a smaller company likes yours?
The open challenge concept allowed a small team of people to innovate outside of our normal development cycle. Doing so allowed us to pilot technology and have results to share ahead of developing a long-term feature. We now know what worked well and what tweaks we’d like to make. It was a fun project!
What’s next in terms of your solution?
We’re looking forward to working with the CDC to understand how they may use our app to enhance their data collection methods for the BRFSS.
What ultimately do you hope to contribute to overall public health data and how will that benefit physicians and other health experts, CDC and the U.S. government, and the general public?
We are hopeful that this technology will provide more high-quality data to the professionals who can use it to positively influence the health of our nation, whether that be on a one-on-one basis through physician-patient interactions or larger initiatives supported by government efforts.
Our methodology is immediately applicable for any health behavior for which wearable devices, mobile apps and remote monitoring devices track data. In the same way, many chronic conditions benefit from improved lifestyle behaviors and can be tracked using a similar solution. Understanding more about the populations’ behaviors can inform campaigns designed to improve chronic conditions. This information also lends itself to monitoring progress in improving chronic conditions.
Thinking in terms of infectious diseases, pushing a simple survey question through push notification would be one useful application. Questions and monitoring could also be targeted to populations based on location. If a participant is near a doctor’s office or hospital, the app can push a quick question to verify their location, and then ask if their presence at the medical building is related to a particular disease or set of chronic conditions that the CDC wishes to track. Similarly, if a user is in a region suffering from an epidemic, the solution could be used to track symptoms.
Another thought is tracking and improving people’s medication adherence. This could be useful for certain populations, particularly those with chronic conditions such as diabetes or persons taking a course of antibiotics. Clinical testing may also benefit from such a solution with the ability to not only track the person’s data around the clock, but to push survey questions and push notification reminders.
We believe that the general framework of passive data collection combined with a limited number of active questions to close the loop, has nearly unlimited potential.
We are a society that is increasingly driven by data, which makes it all the more important to have high-quality data in the hands of those making decisions.
Let’s end with a free-for-all. What else would you like to add about your participation in this challenge?
I’d like to recognize that this was a team effort and I was lucky to work with smart, talented and dedicated individuals on this project:
- Erica Zuhr – data scientist
- Tommy Strickler – statistician
- Sean McCoy – mobile app developer
- Erik Hiser – database architect
- Biaunca Edwards – UI/UX graphic designer
Everyone contributed something critical to the effort and I’m so proud of our work. I think our solution can significantly improve the quantity and quality of data collected by the CDC.
Since its inception in 2010, GSA’s Challenge.gov program has provided resources and training to federal agencies using crowdsourcing competitions to solve critical problems. More than 835 competitions have been listed on Challenge.gov, where members of the public can find and participate in crowdsourcing events open to the public. Success Stories is an ongoing series that highlights past winners and their work.