Developing a holistic systems-based response to the complex challenge of malaria in Burundi
“This is not a difficult but a complex problem. We need a holistic systems-based response moving from blanket interventions to risk-stratified planning that addresses multiple reinforcing factors simultaneously across ecological and social dimensions,” said David Niyukuri of the Department of Mathematics at the University of Burundi.
“Malaria in Burundi is driven by a range of factors rather than a single cause. Transmission is heterogeneous (varies sharply by place and season), a large asymptomatic reservoir sustains onward spread, and health-system bottlenecks (diagnostics, supply chains and data loops) weaken the ability to target interventions precisely,” he continued. “Climate variability, population mobility and economic precarity further amplify risk. Even when bed-nets, treatment and spraying campaigns scale up, rebounds or persistent hot spots are common, unless the system is addressed holistically. Thus, the path to malaria control in Burundi is not one big lever, but a coherent package with targeted layered interventions, better detection of hidden infections, climate-informed timing, and fast feedback loops that let teams shift resources to where they’ll do the best before the peak arrives.”
And, creating the framework for such a package is what Iso Lomso fellow Niyukuri hopes to achieve in his STIAS project.
Burundi is a landlocked country located in East Africa, bordered by Rwanda to the north, Tanzania to the east and south, and the Democratic Republic of the Congo to the west, covering an area of approximately 27,830 square kilometres. It is densely populated with around 13.6 million people, predominantly rural country with a young population, 45% of the population is under 15 years old, indicating a high birth rate. The country is situated in the highlands of East-Central Africa, and its climate is tropical and conducive to agriculture but also favours malaria transmission, making vector-borne diseases a major public-health issue.
Despite significant government initiatives, such as expanding healthcare facilities from 500 in 2000 to 1500 in 2020 and increasing testing (20-fold between 2000 and 2020), there has been a substantial increase in malaria infections since 2000 – in 2019 there were 843 000 cases per million inhabitants more than twice that in the early 2000s. Over 9 million cases were recorded in 2019. That highlights gaps, inconsistent interventions, diagnostic limitations, and the potential effect of the high proportion of asymptomatic people.
Pregnant women and children under five carry the highest burden of malaria infection – around 50% of cases are in children under five with those in the 12-to-59-month category particularly vulnerable. Around 50% of chronic malnutrition is also found in children under five.
Niyukuri explained that part of the reason for the increased reported cases in this period is more testing and a threefold increase in treatment centres from 500 to 1500. “Diagnosis is by microscopy or by rapid diagnostic tests (RDTs) which were introduced from 2006 and are now widely used,” he said.
But diagnosis is not all straightforward with a many false negatives in RDTs. HRP2/3 gene deletions are one of the causes of false negatives and require ongoing surveillance. There is currently a pilot study underway to understand the extent of false negatives. “The World Health Organization recommends microscopy but a study in Ghana has shown that this also misses cases.” (See: https://malariajournal.biomedcentral.com/articles/10.1186/s12936-023-04506-5.)
Treatment was originally chloroquine, but artesunate/amodiaquine was introduced in 2005 and artemether/lumefantrine in 2019 due to chloroquine resistance. The country also uses Sulfadoxine-pyrimethamine (SP) as recommended by the World Health Organization for pregnant women as part of the Intermittent Preventive Treatment in pregnancy (IPTp) strategy. Recent results (https://www.medrxiv.org/content/10.1101/2025.06.22.25330092v1.full.pdf). indicate high levels of resistance in Plasmodium falciparum parasites to chloroquine and sulfadoxine-pyrimethamine (more than 80% of the samples), identifying a novel A675T mutation in the Kelch13 gene associated with artemisinin resistance. This widespread resistance, including specific gene mutations for sulfadoxine-pyrimethamine resistance, suggests compromised effectiveness for malaria prevention strategies like IPTp-SP and highlights the need for molecular surveillance
“Artemisine-based combination therapies remain the mainstay treatment but resistance markers and delayed parasite clearance elsewhere in Africa signal an ongoing need for vigilance,” said Niyukuri.
Treatment also needs ongoing monitoring, that is why Therapeutic Efficacy Studies (TES) need to be conducted on a regular basis. But also going beyond to understand socio-economic factors and programmatic challenges influencing malaria prevention and treatment adherence.
Vector control relies on long-lasting insecticidal nets which were widely introduced from the early 2000s and selective indoor spraying, but insecticide resistance is widespread and heterogeneous. The insecticide resistance problem is also enhanced by the fact that there is only one insectary in the country to monitor emergence of resistance. .
“There have also been some issues with spraying,” said Niyukuri. “Outdoor spraying can also harm other insects including bees so it should only be deployed for serious outbreaks not for first-line control.” Though the cost of spraying might be high, during malaria outbreaks, indoors help to cut down transmissions. Thus, regular research on vector behaviour indoor and outdoor is critical for tailored vector control measures.
“Overall, access to healthcare, diagnosis and vector control have all improved but malaria is still a challenge,” he said.
Climate and geography
Niyukuri explained that geography and climate play an important role in malaria transmission. Burundi is a landlocked, mountainous country. “In the lowlands there is perennial transmission and seasonal surges due to stable breeding conditions and in the highlands temperature and rainfall variations can trigger explosive outbreaks when conditions align,” he explained. “Different mosquito types are found in the different areas – and they don’t all behave the same. Variations in temperature and rainfall have big implications for mosquito vectors.”
Climate change is increasing the risks – “a small temperature increase can make previously unsuitable highland areas conducive to transmission,” he said. “There’s a need for early warning systems and preparation planning for districts using weather forecasts to predict malaria transmission surges.”
“Some of this work previously relied on USAID funding which is now gone – it’s very concerning,” he added. “The entomological surveillance and TES which were ongoing stopped.”
Hidden human reservoirs
Another major problem is the existence of a large reservoir of infected but asymptomatic humans.
“Studies in communities in healthy people have shown high levels of asymptomatic carriers – over 50% − who might not use bed-nets or other prevention and continue to contribute to the cycle.”
“We are only currently treating pregnant mothers and children. The human reservoir is infected but healthy people. They are missing in the fight and need to be reached. This large, asymptomatic reservoir sustains onward spread in communities. There are currently no specific interventions in the National Malaria Control Programme to reach them,” explained Niyukuri.
Health-system bottlenecks include access-to-care barriers, supply chain issues including drug stock outs and delays, and lack of data to inform interventions. “Ideally we need real-time data to advise understanding,” he said.
A USAID funded study published in 2017 (https://journals.plos.org/plosmedicine/article?id=10.1371/journal.pmed.1002234) showed that modern housing in sub-Saharan Africa (SSA) was associated with a 9% to 14% reduction in the odds of malaria infection. It highlights the impact of social and economic vulnerability by looking at the factors within households that influence getting or not getting malaria. “Poverty, housing conditions and livelihood patterns create complex exposure risks that traditional interventions may not adequately address in vulnerable populations,” said Niyukuri.
“The critical gaps we need to close include mapping and surveillance, operational research, diagnostics via enhanced surveillance, education plus screening and preventive therapy pilots in schools, and supply chains with reliability metrics tied to actions and responses,” he said.
His work at STIAS is about leveraging data – including surveillance and meteorological data − to build a framework to advise policy makers and the National Malaria Control Programme.
“We need good, reliable scientific evidence on which to base interventions.” he said. “There has not been enough recent research in Burundi.”
And what would he do immediately if he was given $10 million?
“The driving question is how can we control transmission,” he said. “I would look at transmission dynamics across the districts and the behaviour of the mosquitoes to deploy tailored vector-control interventions. We need to put together the pieces of the puzzle at the district level to be able to implement tailored interventions. We also need to do something about the human reservoirs. It’s still not clear to what extent humans are driving the cycle. These are people in schools, communities, workplaces – so we must find effective interventions for all of these. There is good research showing that countries that target schools can reduce infection rates.” Local capacity building in research is really needed to be able to generate evidence and skilled human resources for the government and partners in the public- health domain.
Michelle Galloway: Part-time media officer at STIAS
Photograph: Igshaan Abrahams