To the Editor.
—Against a background of increasing incidence of tuberculosis (TB) in the United States, Dr Colditz and colleagues1 present a meta-analysis of BCG protection. They state that "[q]uestions about BCG vaccination that remain unresolved include its overall efficacy." Not so. The consensus is that the effect of BCG vaccine varies; this is why the World Health Organization2 recommends that countries undertake their own program evaluation studies. The scientific community is not searching for the overall protective effect. It believes that BCG vaccine efficacy varies, and it is looking for mechanisms and reasons for this variation.3In a meta-analysis last year,4 we explored one aspect of this variation: site-specific differences. We applied meta-analytic methods to substantially the same studies as Colditz et al, separating BCG vaccine efficacy against pulmonary TB from that against miliary and meningeal forms. Protection against meningeal and miliary TB appeared homogenous
Wheeler JG, Rodrigues LC, Diwan V. Efficacy of BCG Vaccine. JAMA. 1994;272(10):765-766. doi:10.1001/jama.1994.03520100029015