 Thank you, and thank you, Nick, for the introduction and also for your tribute to Malcolm, who was a key collaborator for us and actually a strong source of inspiration for many of us. So I'm going to talk not about the whole programme, which Nick has already covered. I'm going to deal specifically with some questions on the epidemiology of malaria, and it's important ways to give credit where it is due. So the work I'm presenting is not really my own work. It's work that's been led by the three names you can see there, Abdi Salam Noor, Bob Snow and Alice Kamal. Alice Kamal being a PhD student, well, recently passed her PhD and is now working as a postdoc with us in the programme. So there are going to be some key concepts for the talk. So the first is this acronym PFPR 2 to 10. So that refers to posmodium felsiparum malaria prevalence in community surveys of children between the ages of two to ten. And the importance of that measure is that that is considered to be a good way of capturing how much transmission of malaria there is in a particular community, based on the idea that people are losing parasites all the time. So if you've got a particular percentage of parasites in that population, that implies that there's a sustained exposure to infected mosquito bites at a level sufficient to maintain that percentage. So that's just infection that we're talking about, and that's not necessarily symptomatic. Then on the right, there are three different disease states that I'm going to mention. So one is cerebral malaria, which at all intents and purposes means a child in a coma in Africa. And the coma is defined by the Blantyre coma score, the score that Nick mentioned, Malcolm Molyneux having been co-inventor of. There's severe anemia, which is caused by the parasites chewing up all the red cells and is defined as a hemoglobin less than five. And there's respiratory distress, which is caused by a high acidosis in leading to rapid breathing. So I'm going to keep going back to those four concepts throughout the talk, but I'm going to start by just mentioning why are we studying all of this. I'll come back to it at the end. But the point is that there are some control measures for malaria. Bed nets and residual spraying to kill mosquitoes are two key ways of protecting children from infected bites. Anti-malarial drugs can be given as treatment, but they can also be given to prevent malaria in the first place. And you're going to hear about vaccines from Adrian later on in the day. So the question I'm going to deal with is how do we monitor the impact of those? How do we look at Africa as a whole to see whether the control measures are effective in reducing malaria transmission? And then in the second part of the talk, I'm going to deal with the question as to how much good does that do? And there are reasons both for and against which I'll return to. So this is the 115 year dataset. You can see across these 115 years, there have been many surveys. Each of them is indicated with a green dot. So as you might expect, there are many more surveys done in the modern era than there were done previously. And so that means that one has to do some geospatial modelling to bridge the gaps between all of these surveys. It's complex. I mean, let me capture the idea for now as just saying it's pretty much smoothing between two points. If you've got two points either side of a district, which both show a very high malaria prevalence, then it's a safe bet that in the middle of the district, there's high malaria prevalence as well. So as you might imagine, that is a more speculative proposition between 1900 and 1944 when you've got relatively few surveys compared with 2010 to 2015. Nevertheless, if you do all of that, you do get a pretty consistent impression of the extent of malaria transmission in Africa. So on the left, there's the map as we would reconstruct it for 1900. And on the right, the map for 2020. So you can see that the central Western Eastern Africa has remained malaria throughout those 120 now years with the five year projection. If you look at North Africa, you can see that a lot of North Africa has succeeded in eliminating malaria. Some of those countries are certified by the WHO as having eliminated some are not certified, but nevertheless, to all practical intents, they have eliminated malaria. And then you can see in Southern Africa, the margin has retreated. So there's almost no malaria left in South Africa. And Namibia, Zimbabwe also looking pretty good, although Mozambique still with quite a lot of malaria. So that's sort of qualitative. Is there malaria there or not? We're also interested though by how much malaria there is. So this GIF is showing you the animation of how things have changed over time. And you can see that there are some areas. So if you cast your eye around the Horn of Africa, Ethiopia, Somalia, and the northern part of Kenya, you can see that malaria transmission towards the end of the 1900s and particularly into the year 2000 and beyond, there is now relatively little malaria in those areas. But if you look at Central Africa and the west of Africa, you can see there's a very dense area of malaria transmission where there's some up and down, there's some reduction in recent times, but there remains a lot of malaria in those areas and really insufficient progress. So one key question is what is causing the trends. So this graph at the top is showing you PFPR, again the prevalence of malaria in the community, on the y-axis. On the x-axis, you can see the year. So starting in 1900, finishing off in 2015. So the red bars are telling us how many surveys went into measuring malaria. So it gives you some idea of how much uncertainty there is. But the average PFPR for all of Africa is shown by the thick green line with the confidence intervals in lycra around it. So you can see that there have been some declines. There's a decline that occurs in the sort of run-up to 1950 and then there's another decline that occurs after the year 2000, which is much more dramatic. So the 1945 to 49 decline is probably largely driven by DDT and a WHO effort to control malaria. That effort to control malaria was abandoned and you can see that malaria then increased after that. The reduction in malaria after 2000 is probably related to at least three factors. So one is that the bed net, or ITN as I'm putting it here in the point, but it refers to bed nets, have really increased in use across Africa. So in the pre-2000, bed net ownership in Kenya was generally down at sort of 5-10%, whereas today bed net ownership in malaria-randomic areas is around 70%. So it's a huge difference and we believe that has had a substantial impact. Anti-malarials have become more effective. Anti-malarials have become available since 2000 and we think that that has also had an effect. I suspect that climate has had an effect also, but it's not a straightforward effect. So if you look at the four graphs that I've got on the bottom now, so again PFPR is on the y-axis for all of them, there are four different potential explanatory variables. So the minimum mean temperature across Africa, the mean rain as measured in the Sahel band, and then socioeconomic factors, so the percentage of the population that is in urban settlements versus rural, and the mean GDP across Africa. So you can see that there are trends in all of those things. I mean the world as we all know has got warmer between 1900 and 2010. It has got, well actually it's complicated to work out whether it's got drier or not, just looking at the mean Sahel rain because there are opposing effects of climate change. There have probably been trends in two different directions. It's become more urban and there's been an increase in GDP. So while all of those things are trends that broadly speaking might go along with the reduction in malaria, you can see that if you look at the year-to-year variations, none of those things tell a consistent story. You know just tell a consistent story, what you would want to see would be 1900 in the top left of the graph, 2010 in the bottom right of the graph, and a straight line joining them in a sort of linear progression between those two points. And you can see that that isn't the case. So it remains complex to predict what's going to happen with malaria, and I don't think we are able to model the future based on that, but we can at least monitor the past and the present. So in case people are wondering about how the modeling is supported by observations on the ground, these are longitudinal data from the Kenyan coast. So this is an area where we've got a lot of data. We've done lots of malaria parasite surveys during the period between 1974 and 2014. And you can see that with the one Sentinel location in Africa, you can see that there definitely are trends. It's not just sort of random noise that we were looking at previously with the African map, and that you can relate some of the events that I was describing to the trends that we were seeing. So I'm going to leave talking about parasite prevalence now in the community, and I'm going to talk about the severe malaria phenotypes that I described earlier. And so I want to tell you about two observations that have led people to be concerned about the outcomes of malaria control in Africa. So this sort of graphic is showing you what happens to the epidemiology of severe malaria as you go from a PFPR, which is very high on the left of the graph, to one which is very low on the right of the graphic. So there's the picture for severe malaria anemia towards the higher end and cerebral malaria towards the right. And that's because it's been a long established observation that as malaria transmission goes up, you see more severe anemia and less cerebral malaria. And that is a concern because cerebral malaria has a much higher mortality rate than severe anemia. If children are admitted with severe anemia to hospital with blood available, the mortality is almost 0%. Whereas with cerebral malaria, the mortalities that we see are depending on the hospital range between 10 to 30%. Along the bottom, I'm showing you histograms which are the distributions of the ages of children admitted to hospital. So in the far left, you can see a histogram with a very big peak down at the 0 to 1 age group. And as you move right, you can see that that distribution flattened. And that, again, is a long established observation that with more malaria, the presentations are in more malaria transmission, I should say. The presentations tend to be in younger children. And as malaria goes down, you see older children presenting with malaria. And that makes sense because with malaria exposure, children become immune to malaria. And so as they become older, they get less malaria. That effect is less marked at high transmission intensity. So both of those effects made people worry that if you improved malaria control in Africa, you might actually have a harmful effect. And so this is an old graph. This is a graph from 1997. The initials on the graph are referring to different places in Africa. There is, again, the parasite prevalence in the community along the exact axis of the graph. I mean, here is actually 0 to 9 years old. This was an earlier era when that was the standard way of doing things. But it would be very, very similar to the 2 to 10-year-old estimate. And on the y-axis, you are seeing the disease rate among children. So what you would expect to see on this graph would be a straight line from the bottom left corner, where you would see sort of zero parasites in the community and zero parasites in hospital, to a straight line to the top right, where you would see the maximum number of children being admitted with malaria and the maximum transmission in the community. But you're not seeing that. So I mean, B is in the bottom left-hand corner. Katie and then H do tell that story that as parasites, transmission in the community goes up. So the disease rate goes up as well. But after that point, you don't see that. In fact, if anything, the disease rate starts to go down again. And so that led people to conclude that in a community with more transmission, the mortality that you might have seen with malaria is offset by the acquisition of immunity in childhood. And worse than that, that actually reducing malaria might even increase the mortality due to malaria. And this caused a very significant controversy in the malaria field with quite, you know, I mean, the sort scientists took very polarized positions in the same way as you're sort of seeing around COVID at the moment. And this was a decision that was of extreme public health urgency. So here is an updating in 2021 of that of that relationship. So you can see that, you know, the ability to draw graphs in the community is clearly sort of improved in terms of the mathematics of it and the presentation of it. But it's basically the same thing. What you're seeing on the bottom is the parasite prevalence rate in the community. And what you're seeing on the left are the admissions per thousand children per year. And you can see that in this graph, you are seeing what I was saying we would hope to see, i.e. that as parasite prevalence goes up, the admissions per child per year go up as well. So it raises the question, were we just wrong in 1997? Did we, you know, choose the locations badly? Or was it the fact that it was a smaller data set, perhaps less precisely measured? I think there's another factor, though, as well, which has changed. And that is that back in 1997, the access to treatment was very poor. Children would frequently be left in the community for many days, if not a week or more, with a fever without access to anti-malarial. Whereas in 2021, the average child in Africa lives five kilometers or less distance from a dispensary. And the likelihood of that dispensary being open and able to give them an anti-malarial is much higher. And their access to hospital care is better. And so my own view is that that is what's changed the relationship. That back in 1997, the main survival advantage a child infected with malaria had was their immunity. Whereas in 2021, it's the health system that is there for them. And it is less likely that immunity is going to be the determining factor. And hence, we can now be confident that reducing malaria is unequivocally a good thing and reduces the risk of severe disease. Now, my slides seem to have frozen. Ah, got it. So I'm now breaking that relationship down by severe malaria, respiratory distress, and cerebral malaria. And you can see that what I said is true for severe anemia and for respiratory distress. For cerebral malaria, it's hard to know what is true, because actually it's just a very rare outcome. It was an extremely noticeable outcome for us in the early days because it's such a dramatic thing. And there was probably quite a lot of misdiagnosis of severe anemia as well in the old days. But when you measure it carefully, although cerebral malaria is very dramatic and those children need treatment, it's actually not the thing that drives the epidemiology of malaria. And so the potential increase in cerebral malaria's transmission goes down. Turns out not to be such a problem for us. Now, what about the age distribution thing? Well, here that is again with data compiled in 2021. And you're seeing those graphs again as transmission, I'm sorry, perhaps rather confusingly, I put it the other way around. So on the left, this time we've got PFPR, the parasite prevalence being less than 5%. And then moving over to on the right, being higher than 30%. So you can see that it is true that as you get more malaria transmission, it's increasingly younger children that you're seeing and the older children are relatively protected. But what you can see, I think, from these graphs is that there's so much more malaria at over 30%, compared with below 10% say, that actually the relative protection of older children doesn't even come close to undoing the impact of the overall increase in how much malaria there is. So I'm going to come back now to some longitudinal data. That was data across East Africa that I was showing you previously. I'm going to come back to longitudinal data in a single site, and that being Kalefi County Hospital. So the graph on the left is showing you the median age of children admitted to hospital dying. And they're separated in the red line with those who die with a positive malaria slide and those in the blue line with those who have a negative line. And you can see that again on a local level, we are seeing that as malaria transmission goes down in Kalefi, the average age of children with malaria, that's the graph on the right, and the average age of children dying with malaria, that's the graph on the left, does increase. So the older children are relatively less well protected. However, in Kalefi, we've also got the advantage of being able to match the children who are admitted to a population. And that means that we can get quite an accurate estimate of the number of children compared with the population denominator who are admitted to hospital. So the graph to focus on here, I think is the top left one mortality and the blue line for mortality. And you can see that although, as I showed you in the previous graph, the average age of children dying has gone up. There are many fewer children dying of malaria in Kalefi as time goes by. And that is true in the other graphs, you can see that that is true of other forms of severe malaria. And so I think we can be again reassured that the reduction in malaria has been a good thing. It has had positive impacts on public health. So that brings me onto my conclusion slide that I think that we can see that malaria transmission in Africa as measured through parasite surveys shows some reductions over 115 years. And we can see that in certainly in East Africa, and we expected to generalize to Africa as a whole, we can see that more malaria transmission leads to more severe malaria. And so even though immunity is acquired with more malaria, nevertheless, that isn't enough to undo the harmful effects of malaria. And so I'd leave you with the hopeful conclusion that malaria control is possible, and it reduces deaths, and is no doubt a good thing.