Now published in IJTLD Open: results from the first cluster-randomized trial to evaluate nurse case management (NCM) for people initiating MDR-TB treatment.
Across 10 district hospitals in KwaZulu-Natal and Eastern Cape, South Africa, we enrolled 2,844 participants (2,134 analyzed) between 2014 and 2019 — among the largest MDR-TB trials published to date. Our primary hypothesis was that NCM would improve MDR-TB treatment success compared with standardized programmatic care. It did not (64.5% NCM vs 61.9% SOC; aOR 0.79, 95% CI 0.56–1.12; P=0.17).
In pre-planned secondary analysis, NCM was associated with a 44% reduction in the odds of treatment failure (aOR 0.56, 95% CI 0.31–0.99), and a 49% reduction in a sensitivity analysis adjusting for baseline cavitary disease. This is hypothesis-generating; we believe the mechanism was closer adverse-event monitoring and adherence support consistent with the Chronic Care Model.
What NCM could not do was overcome poverty, transport costs, distance to care, and late presentation to care. Each of these are structural drivers of loss to follow-up and death in this population. That is where the field needs to move next: decentralized, patient-proximal models with financial support integrated into care.