Mobile phones and their use to study dynamics of the COVID-19 pandemic

Alex Berke

Alex Berke, Kent Larson, Chapter 3 - Mobile phones and their use to study dynamics of the COVID-19 pandemic, Editor(s): Rajkumar Rajendram, Victor R. Preedy, Vinood B. Patel, Features, Transmission, Detection, and Case Studies in COVID-19, Academic Press, 2024, Pages 25-37, ISBN 9780323956468,


The COVID-19 pandemic dramatically changed human behaviors around the world, particularly mobility, as policies restricting mobility and economic activity played critical roles in controlling the rate of infection. Unprecedented access to data collected from mobile phones enabled numerous researchers across academic disciplines to study the dynamics of the pandemic. This chapter describes how mobility metrics were derived from mobile phone data as well as applied, where use cases range from studying economic, social, and political impacts of the pandemic, to studying the link between mobility and infection rates, and creating more accurate epidemic models to simulate alternative scenarios. Limitations and privacy risks are also addressed. Many of the methods and themes covered can be applied to future crises.

Below we show the first 4 figures of the chapter, where further description and illustrations are in the chapter.

Figure 1. Mobility metrics from large providers for a single area (New York County, USA). 

Top panel shows mobility metrics from Apple: Relative change in routing requests for driving, transit, and walking. Middle panel shows mobility metrics from Google: Relative change in visits to POIs and time spent at home (residential). Bottom panel shows mobility metrics from Facebook: Relative change in movement and the estimated number of people staying put. 

Figure 2. Multiple mobility metrics derived from a single source.

The plot shows multiple metrics computed from Andorra Telecom data, from (Doorley et al., 2022). To show the metrics together, they are plotted as a change from a baseline value, computed as the average value over the first week of data in March 2020.

Figure 3. Toy diagram demonstrating origin-destination (O-D) matrices. 

Geographic space is divided into areas (marked A, B, C, D, E, F in diagram) and the daily number of mobile phone users making trips between areas is estimated and summarized in a series of matrices, each matrix representing a period of time. For example, daily O-D matrices are used to analyze the daily number of trips between regions.

Figure 4. Mobility compared across countries.

Apple Maps mobility Index (driving) compared across countries: United States, Italy, United Kingdom, Spain, France, Germany. The index is the percent change in driving route requests relative to January 13, 2020 (set as the baseline).

Related Content