PIP: A large BCG trial was undertaken by the Indian Council of Medical Research with the assistance of the World Health Organization and the US Public Health Service in Chingleput district in an effort to obtain reliable evidence of the efficacy of BCG in providing protection against tuberculosis and to investigate if there are differences between strains of BCG in their efficacy, if the vaccinating dose influenced the protective effect and whether nonspecific sensitivity interfered with the protective effect of BCG. A Leprosy Prevention Trial was superimposed on this study to determine if BCG provided protection against leprosy. The entire population of the Chingleput district -- about 360,000 persons -- was included in the trial. 2 BCG strains -- Danish and French -- were used to vaccinate randomly 2/3 of the population; the remaining 1/3 was administered a placebo injection. The vaccine was administered either in a normal dose of 0.1 mg or a lower dose of 0.01 mg, again by random allocation. The study design included an initial survey of prevalence of infection and disease and subsequent surveys at 2 1/2 year intervals to determine the incidence of tuberculosis in the vaccinated and control subjects. Approximately 270,000 subjects were vaccinated, of which about 1/3 were uninfected initially. These subjects formed the population for analysis of the protective effect of BCG. There were only 285 cases of tuberculosis within a 12-1/2 year period among 90,000 subjects, comprising about 1/5 of the expected incidence on the basis of a prevalence of 10.7 culture-positive cases/1000 subjects. There were at least as many cases of tuberculosis in each of the vaccinated groups -- 93 and 99 -- as in the placebo group -- 93, demonstrating that BCG vaccination either in a dose of 0.1 mg or 0.01 mg failed to protect against the development of bacteriologically positive pulmonary tuberculosis. Detailed analyses indicate that BCG most likely provided some protection. There was clear evidence of the presence of environmental mycobacteria, which caused sensitization and thus could detract from the possible beneficial effect from BCG by preempting BCG. Such an effect, even if it had occurred, could not account for the total absence of the protective effect of BCG. There was a 3% incidence of infection as assessed by tuberculin conversions among those subjects initially tuberculin-negative. The BCG protection provided at least 30% protection against leprosy in the vaccinated subjects.