Source-linked AI summary
Mapping county-level mobility pattern changes in the United States in response to COVID-19
Song Gao, Jinmeng Rao, Yuhao Kang, Yunlei Liang, Jake Kruse
TL;DR
The paper addresses the need to quantify how U.S. communities changed mobility in response to COVID-19 social-distancing and stay-at-home measures. It develops an interactive county-level dashboard from aggregated smartphone mobility data and finds substantial but geographically variable adherence, with mobility changing around gatherings and reopening. The authors conclude that mobility tracking is valuable for infectious-disease research and public-health policy, while noting that the data cannot identify physical distancing itself.
Problem
The study seeks timely quantitative evidence on how people across U.S. counties and states responded to COVID-19 social-distancing guidance and stay-at-home orders.
Method
The authors combine GIS with daily anonymized, aggregated smartphone data on median travel distance and stay-at-home dwell time in an interactive county-level web portal.
Results
High-level adherence occurred at the beginning of stay-at-home orders, but geographic variation existed; mobility also changed around gathering events and state reopening.
Takeaways & Limitations
Tracking human mobility patterns can support infectious-disease research and public-health policy making.
Takeaways & Limitations
Aggregated smartphone mobility data cannot identify physical distancing because of GPS error and uncertainty, and their use raises privacy and ethical concerns.
Abstract
from arXiv · showhide
To contain the Coronavirus disease (COVID-19) pandemic, one of the non-pharmacological epidemic control measures in response to the COVID-19 outbreak is reducing the transmission rate of SARS-COV-2 in the population through (physical) social distancing. An interactive web-based mapping platform that provides timely quantitative information on how people in different counties and states reacted to the social distancing guidelines was developed with the support of the National Science Foundation (NSF). It integrates geographic information systems (GIS) and daily updated human mobility statistical patterns derived from large-scale anonymized and aggregated smartphone location big data at the county-level in the United States, and aims to increase risk awareness of the public, support governmental decision-making, and help enhance community responses to the COVID-19 outbreak.
1 Introduction
The paper presents an interactive county-level mapping platform that combines GIS with daily smartphone-derived mobility statistics to monitor responses to COVID-19 social-distancing guidance. It is intended to support public risk awareness, governmental decisions, and community responses.
- The platform maps how people in different U.S. counties and states reacted to social-distancing guidelines and stay-at-home mandates.
- It integrates GIS with daily updated median travel distance and stay-at-home dwell-time patterns derived from anonymized, aggregated smartphone location data.
- Its stated aims are increasing risk awareness, supporting governmental decision-making, and enhancing community responses to the COVID-19 outbreak.
- The online platform can monitor spatiotemporal mobility changes and the effects of social-distancing policies on movement behaviors.
- The platform can inform analyses of factors associated with compliance, including beliefs, social disparities, political contexts, and geographic contexts.
2 Methods
The study quantifies county-level mobility responses to stay-at-home orders using travel-distance and home-dwell-time measures from two smartphone mobility datasets. It preprocesses, aggregates, and visualizes these indicators in an interactive dashboard for monitoring and analysis.
- 2.1 Data Sources: The study measures responses to stay-at-home orders through changes in median travel distance and stay-at-home dwell time.
- 2.1 Data Sources: Individual mobility is measured as each mobile device’s daily maximum travel distance from its initial location, compared with a county baseline.
- 2.1 Data Sources: The baseline is the median weekday maximum-distance mobility from February 17 to March 7, 2020, within each specified county.
- 2.1 Data Sources: Percent change in mobility is computed from daily median mobility relative to baseline, while SafeGraph data provide county-level median stay-at-home dwell time.
- 2.2 System Design: ArcGIS Operational Dashboards integrate maps, time-series plots, gauges, queries, and interactive interface components for mobility monitoring and analysis.
- 2.2 System Design: The dashboard preprocesses and quality-checks mobility data, aggregates them to counties, spatially joins them to county maps, and calculates mobility indicators.
3 Mobility Changes and Insights
County mobility changed substantially across successive pandemic periods, with active reductions after stay-at-home guidance, geographic heterogeneity, event-related increases, and renewed travel during reopening. Stay-at-home dwell time was generally higher in the most infected states than in the least infected states.
- Before stay-at-home orders: Most U.S. counties increased median travel distance on March 6, 2020, except some outbreak areas such as King County, Washington.
- During stay-at-home mandates: About 73%: New York’s median maximum travel distance decreased relative to baseline, reaching less than 0.1 km on March 15.
- During stay-at-home mandates: Geographic adherence varied on March 15, with reduced mobility across many regions but increased mobility in Florida counties and several Mountain West states lacking statewide lockdown orders.
- During stay-at-home mandates: Wisconsin’s April 7 election coincided with increased travel in several counties, including Waukesha County from 1.1 to 1.9 km and Lafayette County from 2.7 to 4.5 km.The paper also reports longer-distance travel in Rock County, from 2 km to 2.66 km.
- Reopening period: Since early May, median travel distance rose as states lifted stay-at-home and non-essential business restrictions after a period near minimum values.
4 Conclusion and Discussion
The study developed a smartphone-derived mobility tracking portal and analyzed mobility changes associated with statewide stay-at-home orders. It found early adherence with geographic variation, while noting that mobility reductions cannot by themselves identify physical distancing.
- The study used aggregated smartphone location data and user-friendly design principles to develop a human mobility tracking web portal for the COVID-19 epidemic.
- High-level adherence to stay-at-home orders occurred initially, although mobility responses varied geographically across the nation.
- The portal showed the value of tracking human mobility patterns for infectious-disease research and public-health policymaking.
- Reduced mobility does not necessarily ensure physical social distancing, which the aggregated mobility data cannot identify because of GPS error and uncertainty.