BCG not tied to risk of childhood asthma in latest study
Bacillus Calmette-Guérin (BCG) vaccination is not associated with any changes in the risk of developing childhood asthma, suggesting that any associations previously reported may be attributed to confounding factors, according to the QBCIH* study.
“Unlike most other vaccines, which stimulate Th2 lymphocytes and antibody production, the BCG vaccine induces Th1 lymphocytes and cell-mediated immunity. This drives the immune response towards a Th1 response rather than the Th2 response that is prototypical of atopic disorders, suggesting that BCG vaccination could contribute to asthma prevention,” explained the researchers.
The retrospective, population-based birth cohort study involved a 2-stage sampling ─ first using administrative data of 76,623 individuals enrolled in the QBCIH, and the second stage from telephone interviews with 1,643 participants. [Am J Epidemiol 2017;186:344-355]
In the first-stage sampling, asthma risk was 10 percent lower in the vaccinated vs the unvaccinated individuals after adjusting for variables in administrative data (odds ratio [OR], 0.90, 95 percent confidence interval [CI], 0.85–0.95).
However, the protective association between BCG vaccination and asthma development disappeared after additional adjustment for potential confounders in stage-2 sampling in the final model, ie, data not available from administrative databases (OR, 0.95, 95 percent CI, 0.87–1.04).
When the analysis was stratified by age of vaccination, data from stage-1 sampling suggested that a lower asthma risk was associated with vaccination during the first year of life (OR, 0.90, 95 percent CI, 0.85–0.95) but not later (OR, 0.91, 95 percent CI, 0.79–1.06), when compared with unvaccinated individuals.
Again, early-life BCG vaccination was not associated with a reduced risk of asthma after additional adjustment in the final model.
“The initially observed reduced risk of asthma among BCG-vaccinated individuals was no longer present when we adjusted for several important potential confounders unavailable from administrative databases, highlighting the benefit of a 2-stage sampling,” said the researchers.
Previous studies have shown conflicting results of either a protective association or no association between BCG vaccination and asthma development during childhood, but a sensitivity analysis suggested that the observed association was due to publication bias. [Pediatrics 2007;120:e1269-e1277; J Allergy Clin Immunol 2014;133:688-695; Int J Epidemiol 2010;39:469-486; J Allergy Clin Immunol 2011;127:246-253]
The current study employs a 2-stage sampling design, which according to the researchers, potentially “allows correction for the ‘selection bias’ introduced by oversampling individuals belonging to strata with smaller counts in the original sample table.” [Epidemiology 1998;9:309-315; Am J Epidemiol 2005;162:1225-1234]
Nonetheless, they also acknowledged several study limitations, including lack of data on prescription drug claims and potential misclassification of asthma due to lack of differentiation between allergic and nonallergic asthma.