Current issues of ACP Journal Club are published in Annals of Internal Medicine


Review: BCG vaccine reduces the risk for tuberculosis and the risk for death from tuberculosis

ACP J Club. 1994 July-Aug;121:22. doi:10.7326/ACPJC-1994-121-1-022

Source Citation

Colditz GA, Brewer TF, Berkey CS, et al. Efficacy of BCG vaccine in the prevention of tuberculosis. Meta-analysis of the published literature. JAMA. 1994 Mar 2;271:698-702.



To determine the efficacy of the BCG (bacille Calmette-Guérin) vaccine in preventing tuberculosis (TB) using meta-analysis.

Data sources

Relevant citations were identified by searching MEDLINE using the index terms BCG vaccine, tuberculosis, and human; by scanning references of retrieved articles; by reviewing previously compiled lists of BCG studies; and by contacting experts. The World Health Organization and the Centers for Disease Control and Prevention were contacted to ensure that all published and nonpublished articles (in English or other languages) were considered for review.

Study selection

Studies were selected if they were randomized or quasi-randomized controlled trials measuring the efficacy of BCG vaccination in preventing TB or death, with one group receiving the BCG vaccine and the other group receiving no vaccine. The control and the intervention groups had to have equivalent surveillance procedures and similar lengths of follow-up to be considered for review. Case-control studies were also included in the analysis if the criteria for selecting cases and controls and the method for determining their BCG vaccination status were defined.

Data extraction

Data were extracted on study design, age range of study participants, location of study, strain and dose of BCG used, route of administration, outcomes measured, and efficacy of vaccination. The data were independently extracted by 2 readers. Efficacy data were combined separately for the trials and the case-control studies using random-effects models.

Main results

14 trials and 12 case-control studies met the selection criteria. Combining data from 13 trials, fewer patients vaccinated with BCG developed TB than those not receiving the vaccine (Table). Combining data from the case-control studies, the odds ratio (OR) for BCG vaccination against TB was 0.50 (CI 0.39 to 0.64, 50% protective effect). Combining data from 7 trials reporting on deaths from TB, fewer patients vaccinated with BCG died from TB than those not receiving the vaccine (Table). 5 case-control studies reporting on tuberculous meningitis showed a 64% protective effect of BCG vaccination, and 3 case-control studies reporting BCG efficacy in preventing disseminated TB showed a 78% protective effect. Geographic latitude of the study site and the study validity score explained 66% of the between-study variance in the trials.


BCG vaccination reduces the risk for active tuberculosis, on average by 50%, and the risk for tuberculous death, meningitis, and disseminated tuberculosis.

Sources of funding: Centers for Disease Control and Prevention and Association of Schools of Public Health.

For article reprint: Dr. G.A. Colditz, Channing Laboratory, Harvard Medical School, 181 Longwood Ave., Boston, MA 02115-5899, USA. FAX 617-525-2000.

Table. BCG vaccine vs no vaccine*

Outcomes RRR (95% CI) NNT (CI)
Tuberculosis 51% (30 to 66) 142 (99 to 251)
Death from tuberculosis 71% (47 to 84) 87 (54 to 230)

*Abbreviations defined in Glossary; RRR, NNT, and CI calculated from data in article. Follow-up not reported.


Although BCG vaccine has been used for more than 70 years, its efficacy in preventing active TB remains controversial. The controversy stems from the use of different strains, the wide range of definitions and perceptions of reported efficacy, the apparent geographic variability in efficacy, and the abundance of inadequate studies (1).

The decision whether to use BCG vaccination depends on answers to several questions: the incidence of TB in a population; the ability to treat the disease; the efficacy, practicality, and cost of the vaccine; and the effect of BCG vaccination on the ease of interpreting tuberculin skin tests. These questions have become more difficult in the face of resurgent TB and multidrug resistance. The article by Colditz and colleagues uses a careful meta-analysis of the published literature to answer one of the questions, that is, the efficacy of BCG in preventing tuberculous disease. It shows clearly that BCG is modestly effective at reducing the risk for active TB and death.

The documentation of the methods, the methods themselves, and the discussion of the limitations of the study are excellent. It is perhaps telling that of 1264 citations examined and 70 studies reviewed in detail, only 26 studies were sound enough to be included. Geographic latitude and the study validity score explained most of the between-study variance. Different BCG strains did not appear to explain the variation. This study may not really change the limited indications for BCG use in the United States (2), but it does provide data that support policies recommending its use.

Andrew T. Pavia, MD
The University of UtahSalt Lake City, Utah, USA


1. Clemens JD, Chuong JH, Feinstein AR. The BCG controversy: a methodological and statistical reappraisal. JAMA. 1983;249:2362-9.

2. Centers for Disease Control. Use of BCG vaccines in the control of tuberculosis: a joint statement of the ACIP and the Advisory Committee for the Elimination of Tuberculosis. MMWR Morb Mortal Wkly Rep. 1988;37:663-4, 669-75.