University of Pretoria Study Highlights the Limitations of Global Malaria Models for African Contexts
Malaria simulations that don’t consider each country’s unique population growth dynamics will not be enough to help African nations eliminate the disease, UP mathematicians Dr Stéphane Tchoumi and Professor Jacek Banasiak warn.
Alongside North-West University’s Professor Rachid Ouifki, Tchoumi and Banasiak recently published a novel predictive model that plugs in “demographic parameters”, like the pattern of births and deaths in a country.
Their model is the first to compare how different natality, mortality and other population dynamic measures might affect the pace of malaria elimination when using transmission-blocking drugs (TBDs).
“In our model, we show that depending on the demography, the impact of the disease is not the same,” says Dr Tchoumi, adding that health interventions should be adapted accordingly.
Prof Banasiak says their work is a warning to health officials who make long-term predictions about malaria control: make sure that the assumptions about demographics, such as exponential growth, reflect the actual dynamics of the population.
“Don’t simply use off-the-shelf solutions,” he says.
Real-world predictions
Unfortunately, while local population data are relatively easy to obtain from sources like the United Nations’ population data portal, information about in-country malaria infections and TBD use is difficult to access, says Prof Banasiak.
This is, for instance, the case in Dr Tchoumi’s home country of Cameroon, where the malaria burden is one of the highest in the world.
“It is difficult even to collect patient data,” explains Dr Tchoumi. He says malaria has done terrible harm to many of his fellow Cameroonians, and believes that mathematical models are critical weapons that decision-makers need to combat the scourge effectively.
Banasiak agrees. Quipping that although some think they are “transporting mosquitoes in abstract spaces,” he says that once actual data is plugged in, the theoretical simulation turns into a real-world, helpful prediction for a particular country.
He cautions, however, that mathematical modelling is a tool and, as with any tool, it requires a deep understanding of its limitations and ranges of application.
African models
In his capacity as the DSI/NRF SARChI Chair in Mathematical Models and Methods in Biosciences and Bioengineering, Prof Banasiak has been working closely with the UP Institute for Sustainable Malaria Control.
Prof Banasiak and Dr Tchoumi say that while the collaboration has made major strides in accessing large-scale data about local malaria infections and treatment, red tape in the public health sector, as well as silo-working in academia, are serious frustrations.
Dr Tchoumi says he wants to appeal to the general public, malaria researchers and health officials to put more trust in mathematical models that are suited to their particular environment.
He says given that the African population is expected to double in the next two decades, it is high time to develop models incorporating demography. Such models factor in nation-specific growth models at the same time as the effect of supplying the drugs that prevent the spread of malaria.
This means health officials will have a much better idea of how their interventions will fare in the long term. “We can also see how long it might take to eliminate the disease completely,” says Dr Tchoumi.
Dr Tchoumi and Prof Banasiak will now test and retest the model using actual data. At the same time, they are working on cross-border migration models to assess the impact of possible country-to-country malaria transmission.
“A lot of good models are coming from Africa, despite malaria research being underfunded on the continent,” says Prof Banasiak.
He says Africans are nevertheless doing the work needed, and UP plays a leading role in connecting malaria research on the continent to disciplines like the arts, engineering, and, of course, mathematics.