The theme of this year’s World Malaria Day (marked on 25 April) is ‘Driven to end malaria’. One way that scientists and health workers are doing this is by giving anti-malarial drugs (medicines that kill the malaria parasites) to healthy young children during the malaria ‘high season’ - before they actually get sick from malaria. Every year, 50 million children in Africa are treated with this seasonal malaria chemoprevention.
This is a really practical intervention: it uses relatively cheap medicines, and it targets the most vulnerable group – young children - at times of the year when their risk of getting malaria is highest. Crucially, health workers give these medicines directly to children in their communities - parents don’t to travel to health centres to get these life-saving medications, an important consideration in many rural parts of world, where health centres are a long way away and difficult and expensive to get to.
Making things better
At the Infectious Diseases Data Observatory at Oxford University, we’re working to make this malaria preventative treatment even better. We do this effectively and cheaply by bringing together data from multiple studies, then analysing it to produce robust scientific evidence that health policy makers can actually use.
As part of my DPhil, I am evaluating what combinations of antimalarial drugs, including combinations with malaria vaccines, work best for this kind of preventative malaria medication. Normally, you might figure this with a clinical trial where we test out all of these combinations to find out which ones work best. But clinical trials are expensive, even more so in a resource-limited settings in Africa, and the few clinical studies that do exist aren’t enough to provide a comprehensive answer.
But when we pool data from many studies, we can see patterns emerging that aren’t visible when looking at the studies individually. I use a statistical technique called network meta-analysis, which enables me to compare many treatments to each other - even if they have not been directly tested against each other. Network meta-analysis cleverly uses shared comparisons between treatments to indirectly estimate how they would perform against one another. This means that we can build a more complete picture of how different treatment combinations work (even if they haven’t been tested directly in a clinical trial) and rank them based on their effectiveness.
Optimising malaria prevention
We can use statistics to work out the ideal combination of medications that can save even more people from malaria, but in the real world, many other factors influence how effective this preventative malaria medication actually is.
For example, does this prevention still work when the malaria parasites becoming resistant to antimalarials? Does it work equally well in areas with low, moderate or high malaria transmission? Does the intervention protect young children of all ages, or do some groups benefit less? And if so, why? If a child is malnourished, does the preventative medication still work? Or could combining the medication programme with nutritional support make it even more effective? These are some of other questions that I’m attempting to answer as part of my DPhil. Once again, bringing together individual participant data from many studies allow me to answer these questions; something a single study would be unlikely to do.
By generating this kind of evidence from combining data, we can help health policymakers and governments carrying out seasonal malaria chemoprevention make decisions that are actually guided by the best available data. This means being able to choose the most effective drug combinations, optimising the intervention to real-word factors at specific locations, and making the most of limited resources to save lives.
Dhruv Darji is a DPhil student at Oxford's Nuffield Department of Medicine and Infectious Diseases Data Observatory (IDDO).
This article originally appeared on the University of Oxford website.