Abstract
The rapid early spread of COVID-19 in the U.S. was experienced very differently by different socioeconomic groups and business industries. In this study, we study aggregate mobility patterns of New York City and Chicago to identify the relationship between the amount of interpersonal contact between people in urban neighborhoods and the disparity in the growth of positive cases among these groups. We introduce an aggregate Contact Exposure Index (CEI) to measure exposure due to this interpersonal contact and combine it with social distancing metrics to show its effect on positive case growth. With the help of structural equations modeling, we find that the effect of exposure on case growth was consistently positive and that it remained consistently higher in lower-income neighborhoods, suggesting a causal path of income on case growth via contact exposure. Using the CEI, schools and restaurants are identified as high-exposure industries, and the estimation suggests that implementing specific mobility restrictions on these point-of-interest categories are most effective. This analysis can be useful in providing insights for government officials targeting specific population groups and businesses to reduce infection spread as reopening efforts continue to expand across the nation.
Introduction
Early in the COVID-19 pandemic, it became evident that the virus did not impact all urban populations equally. Lower-income neighborhoods consistently reported higher infection rates and mortality. While mobility was widely identified as a primary driver of spread, traditional metrics—such as total trip counts or distance traveled—often failed to capture the actual risk of interpersonal contact. For instance, a long solo commute poses far less risk than a short stay in a crowded indoor environment. This study addresses the need for more granular, "quality-based" mobility metrics to explain the disparities in COVID-19 spread at the neighborhood level (Zip Code Tabulation Areas, or ZCTAs).
The Contact Exposure Index (CEI)
To better quantify infection risk, the researchers developed the Contact Exposure Index (CEI). This metric estimates the "worst-case" total contact duration at a Point of Interest (POI) by integrating visitor volume, venue square footage, and "dwell time" (the duration of stay). By assuming a uniform distribution of visitors, the CEI calculates the sum of overlapping minutes between all pairs of visitors. This approach highlights that risk is a function of both density and time, identifying locations like schools and full-service restaurants as high-exposure environments due to the long durations people spend within them.
Data and Methodology
The study utilized SafeGraph mobility data, which tracks anonymized GPS foot traffic to over 4 million POIs, including industry codes and floor area. This was combined with ZCTA-level epidemiological data from the NYC Department of Health and the Illinois Department of Public Health. Socioeconomic context was provided by the 2017 American Community Survey (ACS). To establish causality, the authors employed Structural Equation Modeling (SEM), treating "exposure" as a latent variable influenced by CEI and stay-at-home metrics to test the pathway: Income → Latent Exposure → Case Growth.
Results: Temporal Trends and the Impact of Lockdown
The implementation of stay-at-home orders in March 2020 led to a significant shift in urban behavior. As illustrated in Figure 4, while the proportion of devices staying home increased and the time spent at home rose, the CEI saw a dramatic plummet. The SEM confirmed that the effect of exposure on case growth was consistently positive and significant in both cities. Notably, NYC was found to be more "sensitive" to mobility changes than Chicago; a smaller relative increase in contact exposure in NYC resulted in a larger jump in positive cases.
Results: Socioeconomic Disparities and Income
A critical finding of this research is that income serves as a proxy for unavoidable risk rather than just behavioral choices. Residents of lower-income neighborhoods were significantly more likely to come into close contact with others throughout the study period. As shown in Figure 5, there is a clear spatial correlation between lower household income, higher CEI, and higher total case counts. In NYC, the disparity in exposure between high- and low-income groups actually increased after stay-at-home orders were issued. This suggests that while wealthier residents could transition to remote work, lower-income residents—often essential workers—remained exposed to maintain city functions.
Industry-Specific Exposure
The CEI allowed for a granular analysis of different business sectors. Schools and fitness centers experienced the most significant declines in exposure following lockdowns. Conversely, supermarkets saw the smallest declines, and industries like fast-food joints and hospitals saw exposure levels rebound quickly. The high dwell times associated with schools (4–5 hours) and restaurants explain why restrictions on these specific industries were particularly effective in reducing the overall "exposure budget" of a city.
Conclusion and Policy Implications
This study demonstrates that the disparity in COVID-19 spread is deeply rooted in the nature of mobility and contact exposure associated with socioeconomic status. By using the CEI, policymakers can move beyond blunt city-wide lockdowns and instead target high-exposure industries and support vulnerable neighborhoods. While the model has limitations—such as assuming uniform visitor distribution and relying on smartphone-using populations—it provides a robust framework for understanding how urban mobility patterns drive epidemiological outcomes.
Summary updated