Technical approach: We are developing a computational infrastructure and a related web-based dashboard to provide real-time public access to mobility data, commuting, and contact indicators for a representative sample of the US population. The web dashboard will include specialized ad-hoc metrics developed in collaboration with public health stakeholders that will be shared in a controlled access section of the platform.
Innovation: Our privacy-enhanced dataset includes information for more than 45M anonymized users in the United States and will span a time horizon from January 2019 up to the current date. The mobility metrics will be aggregated at different spatial resolutions, ranging from ZIP code and county level resolutions, to custom administrative boundaries of interest to public health stakeholders.
Epidemics and pandemics can induce major changes and disruption to collective human behavior. By using high-resolution, de-identified, privacy-preserving location data from millions of opted-in mobile devices it is possible to measure the extent and nature of mobility and behavior of entire populations.
By using curated data from panels of millions of opted-in users, we provide measurement five different mobility and proximity indicators that can be used to quantify daily changes in collective physical distancing: 1) commute volume, 2) mobility range (radius of gyration), 3) traveling flows between major metropolitan areas, 4) average distinct number of contacts per individual, and 5) average contact duration.
These indicators capture the strength of the behavioral responses and the changes in aggregated metrics of mobility, commuting, and contacts for geographic location (states and cities), and type of community (urban versus rural)[1].
Anonymous aggregated geolocation data provided by Cuebiq through their Social Impact program, darker green areas denote higher levels of activity. Data are aggregated in 15 minutes windows for a panel of representative users.
We are building an infrastructure to provide near-real-time access to the suite of mobility, commuting, and contact indicators to public health stakeholders and data scientists working on outbreak analytics. We will continuously improve the data analysis and computational infrastructure and develop a public dashboard to access data and visualizations.
Our ability to generate timely and realistic epidemic forecasts in the early stages of an outbreak relies heavily on the modeling infrastructure and data analytics systems developed during non-crisis periods. While numerous modeling efforts have been sustained for years by research teams across the U.S. and internationally, continuous access to real-time mobility and contact data remains largely undeveloped. With this activity, we aim at establishing and maintaining a real-time mobility and contact data platform for epidemic modelers and public health stakeholders. This platform will serve as a critical resource for the analysis and forecast of outbreak and health emergencies.
Large-scale epidemics models rely on time varying mobility data to explicitly model disease and outbreak dynamics within/between populations[3], and high resolution mobility data can be used to inform policy-oriented modeling studies[2]. Furthermore, real-time mobility and contact data to power statistical and machine-learning forecasting models.
1. Klein, B., LaRock, T., McCabe, S., Torres, L., Friedland, L., Kos, M., et al. (2024) Characterizing collective physical distancing in the U.S. during the first nine months of the COVID-19 pandemic. PLOS Digital Health 3(2): e0000430.
2. Nande, A., Sheen, J., Walters, E.L. et al. The effect of eviction moratoria on the transmission of SARS-CoV-2. Nat Commun 12, 2274 (2021).
3. Chinazzi, M., Davis, J.T., et al. (2024) ‘A multiscale modeling framework for scenario modeling: Characterizing the heterogeneity of the covid-19 epidemic in the US’, Epidemics, 47, p. 100757.
.png)
