COVID-19 have caused a global pandemic that has taken the lives of millions. This studyexamines factors that contribute or inhibit its contagious nature. It is considered for the analysisfeatures of three groups: air pollutants, meteorological measurements and human mobilityindices. The research analyses the relationship between those features and the mortality ofCOVID-19 measured by daily excess of deaths between March 19, 2020 to July 20, 2020 (120observations). The results reveal air pollutants such as PM2.5, PM10 and CO have statisticallysignificantly positive relationship with excess of deaths. Additionally, meteorological variableslike temperature, solar radiation and relative humidity are negatively correlated to Coronavirusmortality. In consequence, it should be noted that cold and dry weather favors COVID-19lethality, while solar radiation in high degree, diminish it. Moreover, it was found that humanmobility is positively linearly correlated with excess of deaths. It is worth to mention that allthe dependencies anlysed, depicted delayed effects along 25 days lag represented by highrelative risk of excess of deaths in upper or lower extremes of the factors. Finally, the machinelearning algorithm, gradient boosting machine, reported the best accuracy on predicting theexcess of mortality shifted backwards 25 days (exposure to death period) from theaforementioned factors. The model depicted as the most important variable to google mobilityindex, explaining 75% of the variance and its marginal effect on the mean response indicatesif quarantine was maintained more than twice less people would die from COVID-19.