Association of BCG vaccination policy with prevalence and mortality of COVID-19

There is some evidence that tuberculosis vaccine bacillus Calmette-Guerin (BCG) has non-specific beneficial effects against non-related infections. Here, we examined the possible association between BCG vaccination with prevalence and mortality by COVID-19 by using publicly available data of COVID-19 in 199 countries/regions and the BCG World Atlas. By using linear regression modeling, we found that the number of total cases and deaths per one million population were significantly associated with the country9s policy concerning BCG vaccine administration. The amount of variance in cases and deaths explained by BCG vaccination policy ranged between 12.5% and 38%. Importantly, this effect remained significant after controlling for the country9s life expectancy and the average temperature in February and March 2020, which themselves are significantly correlated with the cases and deaths indices, respectively. By contrast, the ratio between deaths and cases was weakly affected. This latter outcome suggested that BCG vaccination may have hindered the overall spread of the virus or progression of the disease rather than reducing mortality rates (i.e., deaths/cases ratio). Finally, by roughly dividing countries into three categories showing high, middle, or low growth rate of the cases, we found a highly significant difference between the slope categories among the BCG groups, suggesting that the time since the onset of the spread of the virus was not a major confounding factor. While this study potentially suffers from a number of unknown confounding factors, these associations support the idea that BCG vaccination may provide protection against SARS-CoV-2, which, together with its proven safety, encourages consideration of further detailed epidemiological studies, large-scale clinical trials on the efficacy of this vaccine on COVID-19, and/or re-introduction of BCG vaccination practice in the countries which are currently devoid of the practice.