Background

Bacillus Calmette-Guérin (BCG) vaccination is a widely-used public health intervention for tuberculosis (TB) control. In Taiwan, like other intermediate TB burden settings, steadily declining TB incidence raises important questions on whether universal BCG vaccination should be discontinued. Recent surveys on adverse events following immunisation, such as BCG-induced osteomyelitis/osteitis, also suggest a need to re-evaluate the vaccination programme.
Methods

We developed an age-structured transmission dynamic model, calibrated to population demography and age-specific TB notification rates in Taiwan. We adopted ‘weak-protection’ and ‘strong-protection’ scenarios, representing a range of characteristics including the duration of BCG protection and vaccine efficacies against TB infection and progression. We estimated averted disability-adjusted life years (DALYs) and incremental costs over 10 years after discontinuing universal BCG vaccination in 2018, 2035, and 2050. We also examined the potential impact of ‘surveillance-guided’ discontinuation, triggered once notification rates fall to a given threshold.
Results

In the weak-protection scenario, discontinuing BCG would result in 2.8 (95% uncertainty range: 2.3, 3.1) additional notified TB cases and −4.1 (−7.7, 0.8) net averted DALYs over 2018–2027. In the strong-protection scenario, 82.9 (72.6, 91.6) additional cases and −402.7 (−506.6, −301.2) averted DALYs would be reported, suggesting a robustly negative health impact. However, in this vaccine scenario, there could be an overall health benefit if BCG is discontinued once TB notification falls below 5 per 100,000 population. The most influential vaccine characteristic for the net health impact is the vaccine efficacy against progression to pulmonary TB. In financial terms, the eliminated cost of the vaccination programme substantially outweighed the incremental cost for TB treatment regardless of BCG protection.
Conclusions

BCG discontinuation may be warranted in intermediate burden settings, depending on the quality of vaccine protection, and the potential for refocusing on other TB control activities for earlier detection and treatment.