Evaluation of BCG vaccination among Puerto Rican children.

In 1949-51 a clinical trial of BCG vaccine was conducted among 191827 children in Puerto Rico 1-18 years of age. The trial was designed to simulate a mass vaccination campaign. Follow-up was continued until 1969. Among the 1976 tuberculosis cases estimated to have occurred without vaccination 29% were nonreactors to 10 TU of tuberculin and thus were eligible for vaccination. The average annual tuberculosis rate was 27.6/1 among controls and 19.7/1 among vaccinees a reduction of 28.7%. However the overall reduction in tuberculosis that would have resulted from a complete vaccination program was less than 9%. Vaccination did not reduce the severity of cases that occurred among nonreactors. The reduction in tuberculosis attributable to vaccination showed only minor variations by age race and urban versus rural residence; however BCG effectiveness was significantly greater among males than females. Screening with 100 TU of tuberculin would not have resulted in a significantly different estimate of the effectiveness of BCG. It is noted that high tuberculosis case rates and low risks of acquiring new infections can coexist in the same population. In such circumstances BCG vaccination can do little good and can actually do harm by interfering with the identification of infected persons. The Puerto Rican experience demonstrates the need to estimate as carefully as possible the actual risk of infection before a vaccination campaign is commenced.