Malaria morbidity and mortality in Ebola-affected countries caused by decreased health-care capacity, and the potential effect of mitigation strategies: a modelling analysis

Abstract

Background: The ongoing Ebola epidemic in parts of west Africa largely overwhelmed health-care systems in 2014, making adequate care for malaria impossible and threatening the gains in malaria control achieved over the past decade. We quantified this additional indirect burden of Ebola virus disease.

Methods: We estimated the number of cases and deaths from malaria in Guinea, Liberia, and Sierra Leone from Demographic and Health Surveys data for malaria prevalence and coverage of malaria interventions before the Ebola outbreak. We then removed the effect of treatment and hospital care to estimate additional cases and deaths from malaria caused by reduced health-care capacity and potential disruption of delivery of insecticide-treated bednets. We modelled the potential effect of emergency mass drug administration in affected areas on malaria cases and health-care demand.

Findings: If malaria care ceased as a result of the Ebola epidemic, untreated cases of malaria would have increased by 45% (95% credible interval 43-49) in Guinea, 88% (83-93) in Sierra Leone, and 140% (135-147) in Liberia in 2014. This increase is equivalent to 3·5 million (95% credible interval 2·6 million to 4·9 million) additional untreated cases, with 10,900 (5700-21,400) additional malaria-attributable deaths. Mass drug administration and distribution of insecticide-treated bednets timed to coincide with the 2015 malaria transmission season could largely mitigate the effect of Ebola virus disease on malaria.

Interpretation: These findings suggest that untreated malaria cases as a result of reduced health-care capacity probably contributed substantially to the morbidity caused by the Ebola crisis. Mass drug administration can be an effective means to mitigate this burden and reduce the number of non-Ebola fever cases within health systems.

Funding: UK Medical Research Council, UK Department for International Development, Bill & Melinda Gates Foundation.

Figure 1. Distribution of Ebola virus disease, malaria prevalence, malaria treatment coverage, and estimated effect of Ebola virus disease on malaria mortality in Guinea, Liberia, and Sierra Leone

(A) Cases of Ebola virus disease up to Feb 1, 2015, with data from patient databases except for Liberia from Nov 17, 2014, for which data are from situation reports only. (B) Prevalence of malaria in children younger than 5 years according to slide microscopy. (C) Proportion of fevers treated with malaria in children younger than 5 years, from both population-based surveys, and our estimates of the additional malaria mortality caused by Ebola virus disease (D).

Figure 2. Effect of health-systems failure on incidence of untreated malaria

For Guinea (A), Liberia (B), Sierra Leone (C), and the combined total (D). Pink bars show the number of cases untreated and red bars show the number of cases treated when the system is functioning normally, blue bars show additional cases caused by increases in transmission from the additional untreated cases. The green lines show the present status of the Ebola epidemic (probable and confirmed cases from patient databases), the blue lines show Ebola cases from WHO situation reports.

Figure 3. Effect of health-systems failure on malaria deaths

For Guinea (A), Liberia (B), Sierra Leone (C), and the combined total (D). Red bars show additional deaths in individuals who would otherwise have been treated with an artemisinin-based combination therapy (ACT) and recovered, pink bars show additional deaths in individuals who would not have received ACT or failed to respond to ACT but would have otherwise recovered after hospital care, and blue bars show additional deaths caused by the additional malaria cases attributable to increased malaria transmission. Green lines show probable and confirmed Ebola cases from patient databases, blue lines show Ebola cases from WHO situation reports.

Figure 4. Estimated effect of mass drug administration on malaria deaths during 2015

For Guinea (A), Liberia (B), Sierra Leone (C), and the combined total (D). Red bars show predicted monthly malaria deaths in the absence of mass drug administration or any malaria treatment, blue bars show predicted monthly malaria deaths with mass drug administration with dihydroartemisinin–piperaquine for 3 months beginning in January 2015, the green bars show mass drug administration for 3 months beginning in June 2015, and the purple bars show mass drug administration for 6 months beginning in January 2015. All scenarios assume 70% coverage of the population and that the health system does not continue to function throughout the year.