Net gains: modelling malaria

There are hundreds of millions of cases of mosquito-borne malaria every year and up to half a million children are dying of the disease. Until we have a vaccination, the mainstay of malaria prevention which accounts for enormous international investment are bed nets laced with long acting insecticides like pyrethroids. These prevent people being bitten in the first place and they also help to supress the mosquito population which protects people who are not under a bed net. But mosquitoes are becoming increasingly resistant to the pesticides although not everywhere and not at the same rate. So policy makers need to know how to adjust their strategy in the most cost-effective way. But the data on bed net performance is patchy and noisy. Imperial College's Tom Churcher tells Chris Smith about his new model that combines these weaker, smaller studies into a much more powerful predictive mechanism.

Tom - So there's many ways in which an insecticide might not impact a mosquito in a way it did previously. The way we measured this is using something called experimental hut trials where you have a fake house and you have somebody sleeping under a net and you see how many mosquitos come in the window, how many successfully feed, how many die, how many leave the house without biting. So this is called an experimental hut trial. There's been quite a lot of these done in different areas. So, what we wanted to do was, we wanted to collate all this information together and we wanted to see how mosquito behaviour as measured in these hut trials, these fake houses, how that has changed with the level of insecticide resistance. Problem is, it's actually very difficult to assess the public health impact of resistance on malaria directly. Malaria, data is very noisy and is influenced by lots and lots of things. So it's not obvious when you start having failure of the nets. It could increase the number of cases but it might not be so obvious.

Chris - What has emerged then? What are the findings from the study?

Tom - What appears to be the case is a little bit of resistance and bed nets are still working well, protecting the person sleeping under them. When you have a little bit of resistance, you might not be killing the mosquito but the person underneath the net is still protected. However, there's a slight issue there that because you're not killing the mosquito, that mosquito survives to fight another day or bites somebody else, might bite you when you get out of your bed net. And so, a little bit of resistance. The nets are still working when you're under them but they're not doing as well as they were before by killing the mosquito. Only when you get relatively high levels of resistance, do actually mosquitos seem to actually come away and bite through the net, and come away fed when somebody is sleeping under the net.

Chris - Just define for me, what is a little bit of resistance and what is a lot?

Tom - So, a little bit of resistance and still the majority of your mosquitos that are exposed insects will die i.e. 80, 90 per cent of the mosquitos will die. A lot of resistance, very few of your mosquitos will die. Only 10, 20, perhaps 0 per cent of your mosquitos will die.

Chris - And so the take home message is that we can tolerate a little bit of resistance because there still overall net benefit - excuse the pun - even in the face of resistance. But once you start to get very high levels of resistance in the population, we've got to rethink the strategy.

Tom - Yes. Though even a little bit of resistance, you will perhaps lose some of the protection provided by the net by killing mosquitos. So if you're under a net, yes, a little bit of resistance is fine but if for some reason you're not under a net, then a little bit of resistance still might be a problem for you because there'd be more mosquitos flying around.

Chris - Can your study, also take into account the equivalent to herd immunity because if we look at the human population, we vaccinate 90 per cent of people against MMR for example, the 10 per cent who can't be vaccinated for whatever reason are protected by the overwhelming level of resistance in the herd. If you've got lots of people under bed nets and those bed nets are killing off mosquitos, does there therefore become a fringe benefit for people who don't have a bed net but because mosquito population is being decimated by insecticidal bed nets, they're protected.

Tom - It's completely the same principle. If you have a bed net in the community, you're going to be increasing the death rate of the mosquitos. There'll be fewer mosquitos around. There's going to be fewer mosquitos and this is the critical thing that survive long enough to transmit malaria because it takes 10 days for the parasite to develop in the mosquito. So, if you've got lots of old mosquitos in your population, you're in trouble, lots of young mosquitos, you're okay. And so, the idea is that you get a large number of bed nets in the community and mosquitos, they're dying in larger numbers and you get this kind of community's herd benefit for everybody, not just those people sleeping under nets.

Chris - So, if the WHO were to walk into your office tomorrow and say, "Tom, what do we need to do?" What's the summary take home message from this?

Tom - Like every scientist, you ask for more evidence, but there needs to be more evidence for the actual public health impact because so much money is spent on them that it's really important that we're A.) Buying the right net and B.) Utilising them in the most appropriate way.