The invisible burden of malaria-attributable stillbirths

This correspondence was originally published by The Lancet.

By Freya J I Fowkes, Eliza Davidson, Kerryn A Moore, Rose McGready, Julie A Simpson


The highly anticipated publication of the updated malaria maps in The Lancet shows that despite the substantial progress made since 2000, mortality gains have stalled and there were 618 700 (95% CI 368 600–952 200) deaths due to malaria globally in 2017, most of which occurred in sub-Saharan Africa (545 200 [88%]).

However, mortality calculations ignore malaria-attributable stillbirths because data on fetal death are not captured in standard estimates of infant and under-5 mortality. Using global estimates of malaria, stillbirth, and estimated relative increase in risk of stillbirth associated with malaria exposure, we estimate that across all WHO regions in 2015, between 126 109 (95% CI 51 632–218 474) and 207 971 (125 723–303 396) stillbirths were attributable to malaria, depending on the treatment coverage of malaria in pregnancy (appendix).

Ignoring stillbirths can significantly underestimate the true burden of malaria-attributable mortality by up to a third. The large number of stillbirths due to malaria, even when treated, are indicative of the failure of health strategies available for women only when they are pregnant. These include bednet distribution and intermittent preventive treatment of malaria (which is further compounded by increasing resistance to sulfadoxine-pyrimethamine) delivered at antenatal care in Africa.

Reducing malaria in the general population will reduce malaria deaths, including stillbirths. However, policy makers and funding agents cannot overlook the need to identify preventive interventions that are cost-effective and efficacious and diagnostic tools targeted for malaria in pregnancy to achieve WHO and UNICEF’s Every Newborn Action Plan to end preventable stillbirths by 2035.

We declare no competing interests.


Author’s reply included in the original publication.

Post a Comment