Exercising caution in correlating COVID-19 incidence and mortality rates with BCG vaccination policies due to variable rates of SARS CoV-2 testing

Janine Hensel; Daniel J McGrail; Kathleen M McAndrews; Dara Dowlatshahi; Valerie S LeBleu; Raghu Kalluri

Next, the following parameters were assessed in the univariate analysis: (a) population density (population/km 2 ), (b) sex (ratio of female and male inhabitants), (c) heartassociated mortality, (d) percent smokers, (e) percent urban population, (f) percent migrants, (g) age, (h) rate of CoV-2 diagnostic testing, and treating a nation's BCG vaccination policy as a monotonically increasing continuous variable (never BCG = 0, prior BCG = 1, current BCG = 2). Univariate regression analysis on all nations showed that percent urban population, percent migrants, age, and rates of testing confound the analysis (Figure 2A) . The correlation between specific CoV-2 testing and BCG policy was robust (Pearson ρ = 0.82, p = 2.4x10 -19 , Figure 2B) . A similar strong, but negative relationship, was observed between percent mortality and testing rates (ρ = -0.42 p = 1.5x10 -4 , Figure 2C ). CoV-2 testing rates were significantly different between countries with distinct BCG vaccination policies (one-way ANOVA p = 6.5x10 -5 ). The non-overlapping values between different countries with different BCG policies preclude robust multivariate analysis. To account for this in the subsequent analysis, we only included "high CoV-2 testing" countries with 2,500 or more tests per million (N: Current universal BCG vaccination policy = 21, universal BCG vaccination policy in the past = 17, Never universal BCG vaccination policy = 6), resulting in equivalent distribution of CoV-2 testing rates across countries with different BCG policies (oneway ANOVA p = 0.17). When comparing countries with high CoV-2 testing, current universal BCG policy is no longer significantly associated with a reduced number of All rights reserved. No reuse allowed without permission. the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.