An extended cost-effectiveness analysis of decentralised TB diagnostic testing with Molbio Truenat MTB/RIF versus hub-and-spoke GeneXpert MTB/RIF in Mozambique and Tanzania.

BACKGROUND

In low- and middle-income countries (LMICs), tuberculosis (TB) regained its status as the leading cause of death from a single infectious agent in 2023, surpassing COVID-19. Rapid and accurate diagnosis is critical, with decentralised diagnostic strategies offering a promising solution. Although decentralised diagnostic strategies have been proven cost-effective, further evidence is needed on affordability and equity in high-burden settings. This study, part of a multicentre real-world cluster-randomised controlled trial (cRCT), assessed implementation costs and out-of-pocket (OOP) expenditures across socioeconomic status (SES) groups from a societal perspective.

METHODS

The TB-CAPT Core trial compared decentralised point-of-care TB testing using the Molbio Truenat platform (intervention) with the hub-and-spoke Xpert MTB/RIF Ultra model (control) in Tanzania and Mozambique. Economic data were collected as part of the trial along with asset ownership information using an equity tool. Multiple correspondence analysis was used to construct an asset index for each country. Extended cost-effectiveness analysis estimated incremental participant costs for TB diagnosis and health outcomes (number of participants who initiated treatment within seven and sixty days) across SES groups. Distributional cost-effectiveness analysis assessed facility-based diagnostic cost per treatment initiated from a societal perspective across SES groups. Regression analyses explored the intervention's impact on direct, indirect and total costs.

RESULTS

Average OOP expenditures were lower in the intervention arm (US$8.82) than in the control group (US$13.61). Regression analysis confirmed a significant cost reduction. Least poor participants experienced greater cost-savings (-US$6.36 vs -US$2.93), while the poorest had a higher number of TB treatment initiations within 7 days of diagnosis (poorest vs least poor: 28 vs 8). The incremental cost-effectiveness ratio for the poorest group was US$778, whereas for the other two groups, the intervention showed higher treatment initiation (52 vs 36 for middle, 33 vs 25 for least poor) at lower costs than the standard of care.

CONCLUSION

The intervention reduced patient costs and improved outcomes across SES groups. Decentralised TB testing with the Molbio Truenat platform is both cost-saving and more effective and cost-effective compared with a hub-and-spoke model in Mozambique and Tanzania.