Dear Editor,
We read the article on BCG vaccination policy and COVID-19 spread by
Ozdemir et al. with interest and found it very
thoughtful.1 Our few concerns are as follows.
Firstly, the authors have classified the countries in two groups
according to the presence or absence of recent routine childhood BCG
vaccination policy. The countries having universal BCG coverage above
90% previously and stopped routine vaccination practice in the last one
or two decades only (Czechia, Finland, France, Ireland, Norway,
Slovakia, United Kingdom) were classified into the countries with “no
BCG-vaccination policy”. On the contrary, some of the countries with
routine BCG vaccination policy but poor coverage (Nigeria, Somalia,
South Sudan) were included in the category of countries following
“routine BCG-vaccination policy”. Do the authors mean to confer that
the protection offered by BCG (by its non-specific effects on immunity)
is sustained for one to two decades only post vaccination and the actual
immunization coverage of a country does not play a role? In that
scenario, BCG vaccination shall be protective for children only (the
least-affected age-group with COVID-19), whereas the adults and
geriatric population (most-affected) will not be benefitted by it
against COVID-19. Such classification (according to recent BCG
vaccination status) that completely ignores the previous vaccination
status, actual BCG-vaccination coverage, and generalizes the results
across all age-groups is scientifically wrong and leads to false
conclusions. Therefore, an age-specific comparison of COVID-19 and BCG
immunization coverage might be more meaningful.2
Secondly, the authors did not adjust for any confounding factors like
the number of tests done at that time, population demographics,
co-morbidities, health infrastructure, reporting bias, etc. (though they
mentioned it in limitations) that can significantly alter the
results.3 A recent study observed positive correlation
(opposite to what we think) between the COVID-19 related parameters
(cases, death-rate, and case-fatality rate) and BCG vaccination coverage
of various countries across the span of four decades. However, with
adjustment of confounders, there was actually no correlation between
COVID-19 and BCG coverage.4
Thirdly, at the time of analysis (April 16, 2020) the pandemic was
limited to the northern hemisphere which now has rapidly evolved. For
now, four (India, Brazil, Russia, Peru) out of the five most affected
countries have routine BCG vaccination policy with more than 90%
coverage.5,6 Therefore, the analysis favouring BCG
vaccination in the initial stages of pandemic was too early to predict
and was affected by the limited spread of COVID-19. To test this
hypothesis, we analyzed the correlation between BCG coverage (2010-2018)
and COVID-19 related parameters (Cases per million, and deaths per
million) of various countries at two different time-points (March 01,
2020, and June 29, 2020) using various datasets.5,6 We
observed a weak but significant positive correlation (spearmen rho=
+0.2-0.4, p< 0.05) between the BCG vaccination coverage and
COVID-19 cases and deaths(as of March 01, 2020). However, this
correlation was not seen on June 29, 2020 (Table 1).
Based upon the above facts and observations, we conclude that as of now
there is no correlation between recent BCG vaccination coverage of a
country and COVID-19. As the protective efficacy may not last beyond
childhood, we should not equate the childhood vaccination policy of a
country with the recent vaccination being done in ongoing trials. Until
we have the results of ongoing randomized clinical trials, routine use
of BCG vaccine in COVID-19 management should be discouraged and
restricted to research purpose only.