Equity and Efficiency of Using Wearables Data for COVID-19 Monitoring (2020-2021)
The coronavirus pandemic has resulted in over 3.5 million infections and over 138,000 deaths in the U.S. alone as of mid-July 2020. To quickly identify and isolate new infections and clusters, public health officials are seeking new tools to target diagnostic testing for individuals who exhibit symptoms.
Previous research has demonstrated that wearable technologies can detect physiologic and behavioral changes when a user becomes infected with the influenza, including a heightened resting heart rate, lower heart rate variability, decreased blood oxygen saturation, disturbed sleep, decreased physical activity and changes in wear habits. Together, these “digital biomarkers” form a signature of infection.
In April 2020, Duke launched CovIdentify to test the viability of using wearables to quickly identify individuals who may have contracted the coronavirus. The CovIdentify platform integrates information from widely used wearables with simple daily electronic self-reports on symptoms and social distancing, for up to 12 months. CovIdentify’s overarching objective is to implement existing digital biomarkers and establish new digital biomarkers to develop, validate and translate CovIdentify as a continuous screening tool. Since April, the study has collected data from over 4,500 individuals using a “bring-your-own-device” (BYOD) model.
This project team will improve and expand the CovIdentify study by designing a new database system suitable for large-scale data analysis and recruiting members from underserved populations to participate in the study.
Database restructuring: Team members will redesign the database that stores wearable and survey data from participants into a structure that efficiently stores and retrieves data, with the goal of digital biomarker development. The team will:
- Write and test functions in Python for data retrieval and restructuring into tabular formats for four data types (sleep, heart rate, activity and survey) from three wearable device types (Fitbit, Garmin and Apple Watches)
- Research and discuss the pros and cons of multiple different database models (relational, hierarchical, entity-relationship, etc.) and experiment with a set of them to select the method that best suits the needs of the study
- Write the equivalent functions in SQL code to automatically map the existing database into a new database with a new model and structure
- Adjust Microsoft Azure design parameters so that data storage from the old database into the new one is space- and time-efficient as more data comes in
- Perform economic, computational time and storage space cost analysis, comparing the old database structure with the newly designed options
- Construct a guide for best practices in storing and organizing large-scale health data
Diversify the study population through community outreach: The study’s current BYOD model has resulted in a skewed participant demographics. Team members will work to diversify the study population through improved recruitment and outreach efforts. The team will:
- Generate and disseminate blog posts with visualizations from the enrollment survey and daily response surveys to engage study participants on how their data is utilized and how it can be helpful to combating COVID-19
- Reach out to citizen scientists with opportunities to engage with this research
- Establish a community advisory board consisting of study participants and community leaders from the Latinx, Black and low-income communities in Durham
- Devise strategies to improve the study advertising, participant recruitment and wearable device distribution to underserved populations at risk for COVID-19 who live or work in high-density facilities by engaging with the Duke Clinical and Translational Science Institute, Community Engaged Research Initiative and the Recruitment Innovation Core
- Roll out studies targeting groups in high-density housing settings
- Set performance criteria to evaluate study adherence and digital biomarker accuracy
Database system for storing, securing and retrieving study data easily; development of strong relationships with Durham community partners to establish trust and a culture of participatory research; distribution of more than 1,000 wearable devices to underserved populations at high risk of exposure to coronavirus; identification and validation of digital biomarkers; blog posts and publications.
Fall 2020 – Spring 2021
- Fall 2020: Write and test functions in Python for data retrieval and restructuring into tabular formats; research and discuss the pros and cons of different database models; generate blog posts with visualizations from enrollment and daily surveys; write the equivalent functions in SQL script to automatically map the currently existing database into the new structure; adjust design choices so that data storage from the old database into the new one is space and time-efficient; conduct focus group study; perform economic, computational time and storage space cost analysis; determine underserved populations at high risk of exposure to COVID-19; draft publication for best practices in storing and organizing large-scale health data; form community advisory board
- Spring 2021: Develop strategy with the community advisory board to advertise the study and distribute wearable devices to underserved populations at risk for COVID-19; generate figures for database-related publication; revise publication and submit for review; continue recruitment efforts with community advisory board and develop a strategy for improving study adherence
- Peter Cho, Pratt School of Engineering-Biomedical Engineering
- Jessilyn Dunn, Pratt School of Engineering-Biomedical Engineering
- Ryan Shaw, School of Nursing
/graduate Team Members
Karnika Singh, Biomedical Engineering-PHD
Will Wang, Biomedical Engineering-PHD
/undergraduate Team Members
Yen Dinh, Biology (BS)
Libba Lawrence, Electrical & Computer Egr(BSE), Computer Science (BSE2)
Amanda Stern, Computer Science (BS)
Bilge Tatar, Computer Science (BS), Neuroscience (BS2)
Jaclyn Xiao, Biomedical Engineering (BSE), Computer Science (BS2)
Jaehan Yi, Economics (BS)