Spatial scales of COVID-19 transmission in Mexico

Abstract

During outbreaks of emerging infectious diseases, internationally connected cities often experience large and early outbreaks, while rural regions follow after some delay. This hierarchical structure of disease spread is influenced primarily by the multiscale structure of human mobility. However, during the COVID-19 epidemic, public health responses typically did not take into consideration the explicit spatial structure of human mobility when designing nonpharmaceutical interventions (NPIs). NPIs were applied primarily at national or regional scales. Here, we use weekly anonymized and aggregated human mobility data and spatially highly resolved data on COVID-19 cases at the municipality level in Mexico to investigate how behavioral changes in response to the pandemic have altered the spatial scales of transmission and interventions during its first wave (March-June 2020). We find that the epidemic dynamics in Mexico were initially driven by exports of COVID-19 cases from Mexico State and Mexico City, where early outbreaks occurred. The mobility network shifted after the implementation of interventions in late March 2020, and the mobility network communities became more disjointed while epidemics in these communities became increasingly synchronized. Our results provide dynamic insights into how to use network science and epidemiological modeling to inform the spatial scale at which interventions are most impactful in mitigating the spread of COVID-19 and infectious diseases in general.

Keywords: SARS-CoV-2; epidemiology; mobility data; network science; spatial analysis.

Fig. 1.

Epidemiological situation of COVID-19 in Mexico. A) Map of cumulative cases per 100,000 people, as of 2020 September 1. B) Timeline of new cases per 100,000 population at the state level (7-day rolling average), highlighting the 15 states with the most severe cumulative outbreaks. C) Number of municipalities that reported confirmed cases of COVID-19 through time. D) Age and sex distributions of confirmed COVID-19 cases across Mexico, highlighting “early” and “late” periods during which the relative risk of infections were calculated. E) Age and sex relative risk ratios of infection, comparing the early vs. late periods from (D).

Fig. 2.

Human mobility and transmission of COVID-19 in Mexico. A) Prepandemic average of the inter-municipality mobility network, colored by network community (detected using the Infomap algorithm). Mobility flow data are based on the aggregated Google Mobility Research dataset (see Materials and methods). B) Deviation of weekly human mobility (number of flows within (grey line) and between states (black line)) from baseline (baseline mobility is calculated as the mean weekly mobility between 2020 January 12 and February 29). C) Evolution of the coefficients of mobility flow from Mexico City in (lagged) correlations with state-level case rates across the country, highlighting the key role that mobility from Mexico City played in the early stage of the epidemic. D) Average fraction of total outgoing mobility from each state that is to Mexico City (black) and the median entropy of states’ distributions of outgoing mobility. Error bands correspond to 95% confidence intervals.

Fig. 3.

Network structure determines the synchrony of epidemics. A) Grouping of municipalities based on the state administrative boundaries. Gray shaded municipalities are removed from downstream analyses as they could not be assigned a movement community (see Materials and methods). B) Example grouping of municipalities based on human movement data and a community detection algorithm (52) (see Materials and methods). Colors indicate movement communities. Grey municipalities have limited recorded movements and could not be assigned to a community and were consequently excluded from analysis. C) Synchrony of weekly growth rates of epidemics across municipalities as measured by the pairwise standard error between growth rates. The lower the error, the more synchronized epidemics are. Blue line shows grouping by network communities, and orange shows groupings by state administrative boundaries. The green dashed line shows the nationwide trend in reported cases during this period. For comparison, please also see differences in within-state and withing-community standard deviations of growth rates in Fig. S1.