Transmission dynamics and forecasts of the COVID-19 pandemic in Mexico, March 20-November 11, 2020.

Tariq, A.; Banda, J. M.; Skums, P.; Dahal, S.; Castillo-Garsow, C.; Espinoza, B.; Brizuela, N. G.; Saenz, R. A.; Kirpich, A.; Luo, R.; Srivastava, A.; Gutierrez, H.; Chan, N. G.; Bento, A. I.; Jimenez-Corona, M.-E.; Chowell, G.

The ongoing coronavirus pandemic reached Mexico in late February 2020. Since then Mexico has observed a sustained elevation in the number of COVID-19 deaths. Mexicos delayed response to the COVID-19 pandemic until late March 2020 hastened the spread of the virus in the following months. However, the government followed a phased reopening of the country in June 2020 despite sustained virus transmission. In order to analyze the dynamics of the COVID-19 pandemic in Mexico, we systematically generate and compare the 30-day ahead forecasts of national mortality trends using various growth models in near real-time and compare forecasting performance with those derived using the COVID-19 model developed by the Institute for Health Metrics and Evaluation. We also estimate and compare reproduction numbers for SARS-CoV-2 based on methods that rely on both the genomic data as well as case incidence data to gauge the transmission potential of the virus. Moreover, we perform a spatial analysis of the COVID-19 epidemic in Mexico by analyzing the shapes of COVID-19 growth rate curves at the state level, using techniques from functional data analysis. The early estimate of reproduction number indicates sustained disease transmission in the country with R~1.3. However, the estimate of R as of September 27, 2020 is ~0.91 indicating a slowing down of the epidemic. The spatial analysis divides the Mexican states into four groups or clusters based on the growth rate curves, each with its distinct epidemic trajectory. Moreover, the sequential forecasts from the GLM and Richards model also indicate a sustained downward trend in the number of deaths for Mexico and Mexico City compared to the sub-epidemic and IHME model that outperformed the others and predict a more stable trajectory of COVID-19 deaths for the last three forecast periods.