The global tuberculosis (TB) control plan has historically emphasized passive case finding (PCF) as the most practical approach for identifying tuberculosis suspects in high burden settings. PCF is defined as diagnosing TB among symptomatic patients who self-present to a health provider. It now appears, at least in some settings, that more intensified case finding approaches (such as active case finding) may be needed to control tuberculosis transmission. Active case finding (ACF) requires health providers to seek out TB suspects in the community rather than to just wait for symptomatic individuals to present to a diagnostic facility. While ACF may detect individuals earlier in their course of disease, and hence may reduce the risk of transmission, it is more labor intensive compared to PCF and is usually too expensive to be continuously sustained in resource-limited settings. Given that tuberculosis control programs are resource constrained and that the incremental yield of ACF is expected to wane over time as the pool of undiagnosed cases is depleted, a tool which can help policymakers to identify when to implement or suspend an ICF intervention would be valuable.
In this talk, I will discuss dynamic TB case finding policies that allow policymakers to use accumulating observations about the epidemic (e.g., reported TB cases) and resource availability (e.g., budget) to determine when to switch between PCF and ACF in order to optimize the overall population health. Using simulation models of TB/HIV co-epidemics, I demonstrate that these dynamic policies strictly dominate static policies that pre-specify a frequency of rounds of active case finding. This implies that following any feasible dynamic policy will result in statistically superior health outcomes in comparison with a static policy that satisfies the same budget constraint. Our numerical investigation further demonstrate that the performance of dynamic TB case finding policies is significantly increased in the presence of more accurate diagnostic tests (such as Xpert MTB/RIF) and in settings with high HIV prevalence.