Protective efficacy of neonatal Bacillus Calmette-Guérin vaccination against tuberculosis

This matched case-control study was performed in two hospitals in Bangkok to evaluate the protective efficacy of neonatal Bacillus Calmette-Guérin (BCG) vaccination in Thai children and to find factors that might explain the reported variation in estimates of the protective efficacy of BCG. Cases were defined as children 3 months to 14 years of age who had tuberculosis and controls consisted of one to four children who were born in the same year and had the same district of residence as the case. A matched analysis with a variable number of controls per case was performed on 75 cases and 207 controls. Conditional logistic regression was performed to adjust for the potential confounding effects of household tuberculosis exposure and socioeconomic status. Forty-eight percent of cases had nonrespiratory tuberculosis. Laboratory-confirmed evidence for tuberculosis was found in 48% of cases. The adjusted protective efficacy of neonatal BCG vaccination was 83% (95% confidence limits, 35%, 96%). It was 96% (95% confidence limits, 66%, 100%) when only 36 matched sets of laboratory-confirmed cases were analyzed. Subgroup analyses and literature reviews indicated that the accuracy of tuberculosis diagnosis, types of tuberculosis, duration after vaccination and household tuberculosis exposure contribute to variation in the reported protective efficacy of neonatal BCG vaccination.