Abstract
BACKGROUND: Many individuals previously infected with Mycobacterium tuberculosis (Mtb) may clear their infections or experience substantially reduced progression risks over time. Such dynamics suggest that recent transmission is more important in driving TB incidence in high-burden settings than previously estimated; thus, the impact of interventions to reduce transmission (eg, community-based active case finding) may also be greater than previously thought. METHODS: We constructed 2 models of Mtb transmission that differed only in the inclusion of a clearance mechanism. We calibrated these models independently to the same set of epidemiological data representative of a high-TB-burden setting (India) and used the calibrated models to project the impact of illustrative biennial active case-finding campaigns (75% coverage; 65% sensitivity). RESULTS: The estimated annual risk of Mtb infection and prevalence of recent infection were substantially higher in the model with Mtb clearance, despite being fit to the same data. The clearance model projected a greater impact of case finding on the incidence of TB disease: 45% [95% uncertainty interval 28%-57%] reduction compared to no intervention after 10 years versus 11% [6%-18%] in the model without clearance. CONCLUSIONS: Models that allow for Mtb clearance are evidence-supported and project greater impact from active case finding than models that do not include these dynamics.