Statistical analysis and decision trees to identify risk factors in the Mexican population due to COVID-19 pandemic

Abstract

Introduction: the COVID-19 pandemic caused by the new SARS-CoV-2 virus was a big challenge to the world and was responsible for a vast number of deaths in a brief period; one of the countries with the greatest number of deaths was México. For this reason, studying this emergency is crucial.
Objective: study and compare the available statistics for Mexico about the COVID-19 pandemic and build a machine learning model that helps to identify the risk factors of the Mexican population.
Methods: this research is structured into three sections. Firstly, a worldwide and national statistical analysis, then a decision tree-based model, and lastly, research about the results of the vaccination campaign. Different databases were used to fulfill the objectives of each section. 
Results: with international information, the number of cases and deaths were studied for a group of countries; in addition, this study compared daily cases and deceases in México, Colombia, and Spain. The national data was used to obtain different statistics and a decision tree-based model. For the vaccination campaign, various statistics were gathered.
Conclusions: even though international statistics did not help determine if comorbidities had a significant effect on deceases, national statistics indicate that they were a risk factor for passing away due to COVID-19. Similarly, the decision tree model indicated that hospitalization was a common characteristic among deceased people. For the vaccination campaign, the lack of data was a problem in identifying the role this event had in the development of the pandemic; nevertheless, the international surveillance systems received an exceptional number of reports about adverse events; for this reason, each person should decide if they need a vaccine