Lives saved from malaria prevention in Africa-evidence to sustain cost-effective gains

Lives saved have become a standard metric to express health benefits across interventions and diseases. Recent estimates of malaria-attributable under-five deaths prevented using the Lives Saved tool (LiST), extrapolating effectiveness estimates from community-randomized trials of scale-up of insecticide-treated nets (ITNs) in the 1990s, confirm the substantial impact and good cost-effectiveness that ITNs have achieved in high-endemic sub-Saharan Africa. An even higher cost-effectiveness would likely have been found if the modelling had included the additional indirect mortality impact of ITNs on preventing deaths from other common child illnesses, to which malaria contributes as a risk factor. As conventional ITNs are being replaced by long-lasting insecticidal nets and scale-up is expanded to target universal coverage for full, all-age populations at risk, enhanced transmission reduction may-above certain thresholds-enhance the mortality impact beyond that observed in the trials of the 1990s. On the other hand, lives saved by ITNs might fall if improved malaria case management with artemisinin-based combination therapy averts the deaths that ITNs would otherwise prevent. Validation and updating of LiST's simple assumption of a universal, fixed coverage-to-mortality-reduction ratio will require enhanced national programme and impact monitoring and evaluation. Key indicators for time trend analysis include malaria-related mortality from population-based surveys and vital registration, vector control and treatment coverage from surveys, and parasitologically-confirmed malaria cases and deaths recorded in health facilities. Indispensable is triangulation with dynamic transmission models, fitted to long-term trend data on vector, parasite and human populations over successive phases of malaria control and elimination. Sound, locally optimized budget allocation including on monitoring and evaluation priorities will benefit much if policy makers and programme planners use planning tools such as LiST - even when predictions are less certain than often understood. The ultimate success of LiST for supporting malaria prevention may be to prove its linear predictions less and less relevant.