Progress towards the End TB Goals in South Africa: A comparative cost-effectiveness analysis of tuberculosis interventions

BackgroundSouth Africas National Tuberculosis (TB) Programme aims to achieve targets set by the World Health Organizations End TB Strategy, including an 80% reduction in TB incidence and a 90% reduction in TB mortality by 2030, compared to 2015 levels. We were tasked to evaluate 1) the impact on TB incidence and mortality of scaling up individual interventions focusing on TB preventive therapy, screening, testing and linkage to treatment as included in the 2023-2028 National Strategic Plan for TB (NSP); 2) the budget required for implementing the NSP; 3) the cost-effectiveness of scaling up individual interventions aligning with NSP targets, and 4) the health impact and cost-effectiveness of additional aggressive screening scenarios to reach the End TB targets. MethodsWe used the Thembisa TB model and public-sector costs to estimate the incremental cost-effectiveness per life year saved (LYS) for individual and combined interventions under various scenarios from 2023 to 2042. The NSP scenario included expansions of TB preventive therapy (TPT), symptom screening at primary health clinics and testing with Xpert, screening of household contacts of people with TB, and community-based screening (door-to-door and digital chest X-ray), as well as targeted universal testing for TB (TUTT) in people living with HIV (PLHIV), household contacts of people with TB and individuals with a history of TB, and reduction of initial loss-to-follow-up (ILTFU), all to levels deemed feasible by NSP stakeholders. We also assessed two hypothetical scenarios that largely maximise screening: Max scenario 1 (TPT for PLHIV, quarterly symptom screening for all adults) and Max scenario 2 (TPT for PLHIV, with eligible adults receiving yearly chest X-rays, TB testing, and TPT for household contacts). ResultsThe NSP achieved a 44% reduction in TB incidence and a 55% reduction in mortality by 2030 relative to 2015, while the aggressive scenarios (Max 1 and Max 2) achieved 57% and 56% reductions in incidence, respectively; and 75% and 71% reductions in mortality, respectively. Among individual NSP interventions, over the period 2023-2043, TPT for PLHIV and Xpert testing for symptomatic individuals seeking TB care were cost-saving. Symptom screening for household contacts ($12/LYS), ILTFU reduction ($13/LYS), TUTT for household contacts ($84/LYS), and TPT for household contacts ($106/LYS) were the most cost-effective interventions. Compared to baseline, the NSP scenario increased cost by 57%, saving 6.6 million life-years at $308/LYS over the period 2023-2042. The Max 1 and Max 2 scenarios increased the cost significantly, by 329% and 1526%, respectively. Max 1 saved 16.5 million life-years at $712/LYS, while Max 2 saved 14.6 million life-years at $3,774/LYS. ConclusionScaling up TB interventions to NSP coverage targets will substantially save lives and reduce TB morbidity. However, the End TB targets will not be met even under additional scenarios considered, including more aggressive prevention and screening. Accelerating progress would require both improving efficiencies of existing strategies and developing new tools including low-cost diagnostics and effective TB vaccines.