Costs and cost-effectiveness of an infection prevention bundle to reduce neonatal sepsis and mortality in Zambia: The Sepsis Prevention in Neonates in Zambia (SPINZ) trial

Abstract The Sepsis Prevention in Neonates in Zambia (SPINZ) trial was a prospective observational cohort study conducted in the neonatal intensive care unit of the University Teaching Hospital in Lusaka, Zambia. Introduction of an infection prevention and control (IPC) bundle reduced hospital-associated mortality, total mortality, suspected sepsis, and confirmed bloodstream infections. This companion analysis was undertaken to analyze intervention costs and cost-effectiveness in this low-resource setting. We conducted a retrospective cost analysis, using SPINZ study-related records, and expressed costs in real 2016 US dollars. We also estimated intervention cost-effectiveness using both outcomes from SPINZ (avoided deaths, confirmed infections, and suspected episodes of infection) and estimated disability-adjusted life years (DALYs) averted by the intervention. To provide data for policymakers, a future cost projection was undertaken to estimate costs of the program implemented nationally over a 10-year period in real 2025 US dollars. A total of 2,035 neonates were enrolled from September 2015 to March 2017. Total costs during implementation (introduction of the IPC bundle) (April-May 2016) and the subsequent intervention period were $17,641 and $5,265, respectively, of which most expenses were incurred during the preparation period due to travel and training. During the intervention period, the program’s running cost was approximately $478 per month. The estimated cost per death, confirmed infection, and suspected episode averted was $208, $204, and $32, respectively; the estimated cost per DALY averted was $9.5. The future model was estimated to cost an average of $107,561 annually to implement nationally. The analysis indicated that the IPC bundle to prevent sepsis-related neonatal mortality was highly cost-effective. Cost reductions from task-shifting, reduced preparation (start-up) costs, and longer intervention periods would further decrease cost per death averted. IPC bundle implementation can thus be recommended for resource-constrained settings where sepsis and other nosocomial infections are associated with high neonatal mortality.