Project profile: A novel platform to support contact tracing during the COVID-19 pandemic

Status: Complete

Early in 2020, facing uncontained spread of the SARS-CoV-2 infection in North Carolina and with testing and contact tracing resources severely constrained, researchers at Duke designed the Snowball study around the urgent need to diagnose cases, calculate the prevalence of the disease, and understand its spread. The study deployed a respondent-driven sampling design, in which individuals infected with COVID-19 serve as seed cases and recruit their own social contacts or peers for testing and enrollment in the study. To support the study, a team from Duke Crucible and Duke AI Health designed and implemented an integrated cloud platform to manage the recruitment, enrollment, and management of study candidates and their data.

Leveraging technology to make enrollment and data collection as frictionless as possible for study participants and researchers, the platform was integrated with the Duke Health electronic health record (EHR) to automatically identify all COVID-positive cases who met the study’s inclusion criteria. It was also designed with participant-facing websites and applications to guide research subjects through an electronic process of consent, enrollment, and peer recruitment. Lastly, the platform managed all of the study data, including the linkages between seed and peer cases to support the network analyses conducted by the study investigators.

With the platform in place, the study enrolled over 500 participants and collected data on over 2,500 individuals for network analysis, helping researchers better understand how the virus was spreading and how behaviors might be impacting its transmission. As a key deliverable of this CDC-funded project, the Duke AI Health team made the platform publicly available by publishing all of the code for a General Release version of the platform that can be deployed quickly by health systems and public health agencies as a tool for responding to infectious disease outbreaks in the future.

The development of the Snowball platform was supported by the Centers for Disease Control and Prevention under contract 75D30120C0955

Snowball platform website: https://sites.duke.edu/snowballstudyplatform/

An open-source General Release version of the Snowball platform is available for deployment by any development team seeking to support respondent-driven sampling:

Snowball General Release API Github Repo: https://github.com/duke-crucible/Snowball-API

Snowball General Release UI Github Repo: https://github.com/duke-crucible/Snowball-UI

ClinicalTrials.gov ID: NCT04437706

Principal Investigator: Dana Pasquale, Assistant Professor in Population Health Sciences, Duke School of Medicine, Department of Population Health Sciences

Publications:

Pasquale DK, Welsh W, Olson A, Yacoub M, Moody J, Barajas Gomez BA, Bentley-Edwards KL, McCall J, Solis-Guzman ML, Dunn JP, Woods CW, Petzold EA, Bowie AC, Singh K, Huang ES. Scalable Strategies to Increase Efficiency and Augment Public Health Activities During Epidemic Peaks. J Public Health Manag Pract. 2023 Nov-Dec 01;29(6):863-873. doi: 10.1097/PHH.0000000000001780. Epub 2023 Jun 28. PMID: 37379511; PMCID: PMC10549909.

Pasquale DK, Welsh W, Bentley-Edwards KL, Olson A, Wellons MC, Moody J. Homophily and social mixing in a small community: Implications for infectious disease transmission. PLoS One. 2024 May 28;19(5):e0303677. doi: 10.1371/journal.pone.0303677. PMID: 38805519; PMCID: PMC11132460.


AI Health Spring 2023 Virtual Seminar Series
Title: Design and Implementation of a Novel Platform for COVID-19 Contact Tracing and Testing
Presenter: Andrew Olson, MPP,with host Hyeon Ki Jeong, PhD
Date: April 12, 2023
Format: Virtual session
Hosted by: Duke AI Health

View the recording here: https://duke.hosted.panopto.com/Panopto/Pages/Viewer.aspx?id=c9adbedc-a06e-4e28-84c4-afe2015e63b6