Malaria vaccine rollout averted about one in eight child deaths in three countries

The four-year evaluation of what happens when a malaria vaccine is put into a routine national immunisation programme — rather than into a clinical trial — has reported its primary result. Across 158 districts and subcounties in Ghana, Kenya and Malawi, introducing the RTS,S/AS01E vaccine in 2019 was associated with a mortality rate ratio of 0·87 among children old enough to have received three doses, a 13% reduction that the authors describe as averting roughly one in eight deaths [s1].

The number matters because of how modest the delivery was. This was not a study of what the vaccine can do under ideal conditions. It is a measure of what it did with real-world coverage, an incomplete four-dose schedule, and health systems that were absorbing a new product for the first time.

What was measured, and how

Malaria vaccines have now been added to immunisation schedules in 25 sub-Saharan African countries [s1]. The three-country pilot that preceded that expansion was designed as a cluster-randomised implementation programme: administrative units — districts in Ghana, subcounties in Kenya, and groups of immunisation clinics in Malawi — each covering an annual birth cohort of roughly 4,000 children were randomly assigned 1:1 either to introduce RTS,S in 2019 or to introduce it later [s1]. That produced 158 clusters, 66 in Ghana and 46 each in Kenya and Malawi, split evenly into 79 implementation areas and 79 comparison areas [s1].

RTS,S was given in four doses: at 6, 7, 9 and 24 months of age in Ghana and Kenya, and at 5, 6, 7 and 22 months in Malawi [s1]. Mortality was tracked not through hospital records but through a network of 26,000 local reporters who notified deaths in their communities, after which study staff visited the family to confirm the details and complete a verbal autopsy [s1]. Severe malaria and other conditions were monitored separately at 18 sentinel hospitals over the 46-month evaluation period [s1].

The primary outcome was all-cause mortality excluding injury, in children eligible to have received three doses. Crucially, the analysis did not simply compare deaths in vaccinated versus unvaccinated areas. It compared the ratio of deaths in vaccine-eligible age groups to deaths in non-eligible age groups, between implementation and comparison areas [s1] — a design intended to absorb the background differences in child mortality that would otherwise confound the comparison.

The findings

By the end of the evaluation, 1,289,504 children had received a first dose of RTS,S, 1,158,850 a second, 1,068,039 a third, and 436,527 a fourth [s1]. Coverage assessed in a 2022 household survey was 82·8% (95% CI 80·7–84·9) for the first dose, 71·1% (68·8–73·5) for the third, and 39·9% (36·9–42·9) for the fourth [s1].

Excluding injury deaths, there were 5,576 deaths in implementation areas versus 6,152 in comparison areas among children eligible for the third dose, against 7,534 versus 7,044 deaths among children not eligible [s1]. The resulting mortality rate ratio was 0·87 (95% CI 0·77–0·97; p=0·016) [s1].

The confidence interval is worth reading carefully. Its upper bound sits at 0·97, close to the null. The result is statistically significant but not enormously precise: the data are consistent with a mortality reduction as small as 3% or as large as 23%. What the evaluation establishes with reasonable confidence is direction and rough magnitude, not an exact figure.

Why the fourth-dose number is the interesting one

Only about two in five eligible children received the fourth dose [s1]. The schedule as designed runs to four doses, with the last given at 24 months in Ghana and Kenya and 22 months in Malawi [s1], and the gap between 71·1% third-dose coverage and 39·9% fourth-dose coverage is the largest single delivery shortfall in the programme [s1]. The mortality benefit reported here was therefore achieved in a setting the authors characterise as having moderate three-dose coverage and low fourth-dose uptake [s1].

That cuts two ways. It suggests the programme is delivering measurable benefit despite an incomplete schedule, which is reassuring for the 25 countries now rolling the vaccine out. It also implies that the ceiling has not been reached, and that the gap between 71% and 40% is where additional lives are likely to be found. An accompanying commentary in the same journal frames the result as evidence that RTS,S implementation reduces mortality in African children [s2].

What this does and does not settle

This is an observational evaluation of a randomised introduction, not a randomised controlled trial of the vaccine itself. Individual children were not randomised; areas were. Deaths were ascertained through community reporting and verbal autopsy rather than certified cause of death, a method that is standard in this setting but less precise than hospital-based ascertainment.

The evaluation also cannot separate the vaccine's contribution from everything else that changed in these areas over four years — though the eligible-versus-non-eligible age-group comparison is designed specifically to limit that problem, since both age groups in a given area experience the same bed-net campaigns, the same rainfall, and the same clinic staffing.

The evaluation is registered as NCT03806465 and is now complete [s1]. The authors' stated conclusion is that these results argue for accelerating malaria vaccine deployment in the parts of Africa where malaria remains a leading cause of child death [s1]. The open operational question for the countries already deploying is narrower and more concrete: whether the fourth dose, currently reaching under half the children who get the third, can be brought closer to the rest of the schedule.

Sources

• Impact of introducing RTS,S/AS01E malaria vaccine on mortality in young children in Ghana, Kenya, and Malawi: an observational evaluation of a cluster-randomised implementation programme, The Lancet, 7 May 2026
• RTS,S/AS01 implementation reduces mortality in African children, The Lancet, 7 May 2026