Population-level data have suggested that bacille Calmette-Guerin (BCG) vaccination may lessen the severity of COVID-19; prior reports have demonstrated conflicting results. We leveraged publicly available databases and unsupervised machine learning, adjusting for established confounders designated a priori, to assign countries into similar clusters. The primary outcome was the association of deaths per million related to COVID-19 (CSM) 30 days after each included country reported 100 cases with several factors including vaccination. Validation was performed using linear regression and country-specific modeling. This protocol details the statistical analyses used to establish an association between BCG vaccination and CSM, which includes : Definition of the target function, data processing, exploratory factor analysis for variable selection, k-means clustering and step wise linear regression for validation. This protocol is differentiated from previous works on the same subject by its' comprehensive nature which considers the effect of several confounding variables while studying the association between BCG vaccination and CSM. There are still several potential measured and unmeasured confounding variables which could not be included in this study. It is also unclear if the protection from neonatal vaccination with BCG is transferable to those receiving vaccination as an adult and how long such protection lasts. The authors advise caution against routine BCG vaccination for the prevention of COVID-19 until prospective trials are completed.