The contribution of TB rapid diagnostic testing in reducing TB-related mortality in Sub-Saharan Africa- in both Person-Living with HIV and HIV-Negative populations: A 9-year quantitative retrospective analysis.

Abstract

Background: A potential contributor to achieving WHO's "End-TB" goal of 90% reduction in TB related mortality by 2030, is scale-up of TB Rapid Diagnostic Testing (RDT). Our study evaluated the contribution of RDTs' in reducing TB-related mortality in both PLHIV and the HIV-negative population, from 2015 to 2023 in Sub-Saharan Africa (SSA).

Methods: We carried out an 9-year quantitative retrospective analysis of country-level data (annual WHO TB reports) for all countries in SSA reporting to the WHO. We estimated the following parameters: incidence, notification, percentage of undiagnosed TB patients, percentage diagnosis with RDTs, and TB-related mortality. We stratified the reports according to TB incidence (creating incidence strata) and limited further analysis to reports where the percentage of undiagnosed individuals was 30% or less. We then used scatter plots to examine the existence of a relationship between the use of RDTs and TB-related mortality, and quantified the observed relationships via linear regression models.

Results: Over the nine years, SSA made great strides toward the 2025 milestones of End-TB disease burden-related targets; TB disease incidence decreased by 14%; TB-related mortality decreased by 27.2%; and TB/HIV-related mortality decreased by 64.1%. Similarly, RDT became the priority TB disease diagnostic modality (66.0% in 2023). We found a consistent inverse relationship between RDT scale-up and TB-related mortality in the HIV-negative population, which was significantly stronger in the higher TB incidence settings (R2 = 0.692, P = 0.003). Following adjustments (R² = 0.883, P = < 0.001), independent predictors of TB related mortality in this population were TB RDT use, TB incidence, TB notification, percentage undiagnosed TB and percentage with drug resistant TB. In contrast, the relationship was weaker and inconsistent in the PLHIV population and was significant only where the TB incidence among PLHIV was very high (R2 = 0.541, P = 0.0239). Following adjustments (R² = 0.944, P < 0.001), just TB incidence and TB treatment coverage in PLHIV were independent predictors of TB mortality in this population.

Conclusions: This study provides support about the anticipated contributions of RDTs in decreasing TB-related mortality in SSA, highlighting the importance of maximum scaleup (addressing underdiagnosis of TB) and limiting the biased prioritization of PLHIV for these RDTs.