Elie Hassenfeld: [00:00:00] Hey everyone, this is Elie Hassenfeld, GiveWell's co-founder and CEO. Today I'm going to be joined by Alex Bowles, a Senior Program Officer on our malaria vector control team, to talk about the largest grant that GiveWell has ever made in our history. It's a $276 million grant to the Against Malaria Foundation to fund the purchase and partial delivery of insecticide-treated nets that prevent malaria. This will support their distribution in the Democratic Republic of Congo between 2027 and 2029, and we expect that this grant overall will avert 67,000 deaths.

We've talked a little bit already about our plans to scale at GiveWell and to be in a position to direct significantly more funding in the future. We directed more than $400 million last year. We've already surpassed that number in 2026 to date. [00:01:00] This grant is a really big part of that story. We now think that our grant making this year will approach, and may exceed $1 billion. 

I think a question that I get a lot from people is: as GiveWell grows, are we able to maintain the quality of our work? Are we maintaining that rigor? And one thing that I think is interesting is that as we've grown, I think we've not only maintained that rigor, but increased it. The quality of the decision-making and the depth of the research that we do is much larger today, you know, much better today, than it even was two years ago, never mind in GiveWell's past.

And so today, as we talk about this grant to AMF, we're going to talk about a question that I think many people have, which is: GiveWell's been supporting AMF for so many years, for more than 15 years, what are the questions that we still have? And as we go through questions that relate to how confident can we be in estimating the level of malaria in DRC, which affects our cost effectiveness estimates and then [00:02:00] the amount that we want to grant. How much do we know about the way in which nets are delivered in the real world and the issues that people face in obtaining and utilizing their nets? How long do nets last, and what does that mean for the most cost effective to distribute them? You know, all of these are questions that we weren't in a position to answer as effectively two or three years ago even, as we do today, and our larger, more experienced team has enabled us to answer those questions with more confidence, with greater depth, and enabled us to deliver a very large amount of money to this program, which will help tens of thousands of people in DRC.

So today I'm chatting with Alex Bowles about this grant. Alex, can you introduce yourself?

Alex Bowles: Yeah, thanks Elie. I'm Alex Bowles. I'm a Senior Program Officer at GiveWell, and I lead our malaria vector control grantmaking.

Elie Hassenfeld: So tell us a little bit more about Against Malaria Foundation and this large grant that you [00:03:00] led.

Alex Bowles: Absolutely. So the Against Malaria Foundation is a UK-based charity that has been working to distribute nets, principally in Africa, for about 15 years. They buy nets, and they work with partners in a range of countries across the continent to distribute these nets to prevent malaria.

And we've been working with them for a long time. We funded them in a lot of countries. This particular grant is our biggest yet. We're really excited about it, and we think it's going to save a lot of lives.

Elie Hassenfeld: And so, I imagine a lot of people who are listening to this know about the Against Malaria Foundation—we often call them AMF—because GiveWell has supported them for a very long time, more than 15 years. I think I first talked to AMF's CEO in 2008. So this is a group that we have this longstanding history with. What does AMF do to enable malaria net distributions to take place? Like, what is their particular value add?

Alex Bowles: Sure. They do a [00:04:00] lot of things, but the core thing that they do is they buy nets. So they work with the companies that make those nets. They negotiate prices for them. They work out how many nets are needed. They buy the nets, and then they make sure that they get to the countries where they're going to be used. They work with the national malaria programs and their implementing partners to make sure that the campaigns to distribute them are well designed. Then they work with partners to make sure that we do surveys afterwards to see how much the nets are being used. Did the nets get out there? Are they providing the protection that we hoped that they would provide?

So they really do that whole sequence of things from buying, which is their really core function, but they also work closely with partners, and particularly in DRC, where they've been working for a really long time, they are really core partners working with everyone involved in distributing these nets to the people who need them.

Elie Hassenfeld: Yeah. That's great. So, you know, I want to dig in a bit on this grant, and I imagine that a question a lot of people might have, because [00:05:00] I get it periodically from donors, is something along the lines of: "GiveWell, you've supported AMF, for coming on 15 years at this point. The malaria vector control team right now is six people. You know, like, what are you doing when you're investigating this grant? What questions could you still have about AMF?"

And so I would love to just hear from you, who led this work, what are the big questions that were on the team's mind as we assessed whether and what kind of grant to make in this place, and how do we answer them? So let's just go through them, I'd love to hear what was there.

Alex Bowles: Absolutely. And it is the case that with a long-standing partner there are some things that are easier, but we always have questions. And I think the things that we were thinking about in this case were things like, how confident are we that nets in the real world really work the way that we model them to work? Can the Against Malaria Foundation scale up to this size? This is bigger than anything they've done before, so how can we be confident they'll scale up? What do we know about malaria burden [00:06:00] in this particular context, in this country, in the DRC, and in the different parts of it?

And then there's a sort of broader question, which is, what happens if we, together with AMF, become the biggest funder of malaria vector control in DRC? Does that change how we should think about the grant? And then there's a couple of other things, one of which is, how long do these nets last? If we distribute them, how long are they going to protect people for? And that's a question that the answer to which changes over time. So that's something that we wanted to understand better for this grant.

Elie Hassenfeld: All right. That's great. So we're gonna go through these. Before we do, we've said this a few times—malaria vector control. So can you just define what that means so people know what you're communicating when you use those words?

Alex Bowles: So malaria vector control is anything that prevents malaria by preventing people from being bitten by mosquitoes. So in this case we're talking about nets—people hang nets over their beds so it protects them at night so they're not bitten by mosquitoes that transmit malaria. But there are other things beyond nets, like spraying walls with insecticides or treating breeding grounds, that kind of thing. [00:07:00] All of those things we would consider vector control.

Elie Hassenfeld: And then there's other types of malaria control activities. There can be providing treatment if people contract malaria. Malaria vaccines, which are trying to reduce the burden of malaria through mechanisms that don't interact with the mosquito that transmits malaria.

Okay, so we're gonna go through these five questions that were on your mind. The first question you raised cuts right to the core of it, which is, we expect that this grant is going to have a huge effect. You know, we estimate that it will avert about 67,000 deaths. And your big picture question is like, well, should we really believe that number? Do we buy it? So yeah, just tell me how you thought about that question, both why that's a concern…I don't know, I think maybe some people would say like, “What do you mean? Like, don't you know that nets are having an effect? Why are you even worried about that?" And then how we went about answering it.

Alex Bowles: So there are quite a few different parts to this as you hinted at. One of them is, if you think about the way that GiveWell often models the impact of our interventions, we're [00:08:00] often looking at the evidence from a randomized control trial that was done to test how effective, in this case, are nets? And then we extrapolate from that, we make a lot of adjustments for different contexts and so on, to reach a conclusion about this particular grant in this particular context.

But there are lots of parameters of that model that we want to test that really apply in the real world. And so, for example, we want to be confident that those nets are being used by people. So if a net is delivered to a household but isn't used, it's not going to protect anyone against malaria, and therefore we should be less confident about the impact. And so we wanted to make sure that we had evidence that people use nets if they're delivered to them. We want to have evidence that the nets are actually reliably being delivered to households, that they're not going missing en route, they're not being delivered to the wrong households or missing certain households.

Elie Hassenfeld: That makes sense, but I guess I wonder, don't we know that from years of looking at AMF's track record and their monitoring? So, AMF is a [00:09:00] group that has asked and is trying to answer these same questions. What's missing there in your confidence that people are using the nets and that they're reaching the people who need them?

Alex Bowles: Sure. So one element of this is looking for ways to understand the impact in the world. So the things that you're describing are sort of monitoring, so looking at those different steps, are the nets getting out into the world? Are they being used? Those are the things I talked about before. So one way we can build confidence and triangulate on that is to look for, well, are they preventing malaria? Is there less malaria where they're used? 

And that was something that we were able to do for this grant, we were able to look back at a previous grant that we had made to support malaria nets in DRC, and see whether they had actually prevented malaria. And this was new for us, and we were able to do this because there was a big survey, a big national survey in DRC, a few years after those nets that we funded were distributed. And so we were able to retroactively do a study, essentially, a [00:10:00] quasi-experimental study, to see whether the nets actually had that health impact. And fortunately, it turns out that they did, and it was really close actually to what we had modeled when we'd made that initial grant, the impact that they would have. So that gave us a lot more confidence that all of these pieces in the chain were adding up to impact. And that was something we'd not been able to do in the past because we didn't have that kind of survey evidence.

Elie Hassenfeld: Right. And I think that's like something that's just important to know about GiveWell's work, and one of the things that make international giving challenging, that there's no source of truth that very straightforwardly says, "This program had the effect that you expected."

And so we do a lot of work upfront to try and predict what will happen or the impact the grant will have. Then, we do everything we can to gather data that helps us determine how the grant went, but there still are going to be missing pieces. And so this was a unique case where, because of the way in which DRC does net distributions, and then the presence of this survey, and the data we were able [00:11:00] to analyze, we could really see that the nets had the effect we expected in the real world. And in some ways, that's this very powerful, confidence-increasing piece of data to the work we're doing because we know that there are, let's say unknown unknowns, about the way in which the grant could be failing to have the impact we expect. And so it was really powerful to have that in this case.

A second question that you raised was, this grant was much larger than grants that we have made in the past, and much larger than distributions that AMF had supported. So was AMF going to be able to successfully operate at this much larger scale? So tell me more about that question, what made you wonder about it, and how you went about answering it.

Alex Bowles: There are a few parts to this. One is, as you say, this ended up being a much bigger grant than we had made to AMF in the past, nearly three times as big as our previous largest grant, which we made to them a couple of years ago, and that one was spread across several countries. So really this is a very big bet on their ability to deliver in [00:12:00] DRC in particular. And so it was really important to us that we understood how they were going to scale and what that looked like.

And so there are a few things that we did as part of this investigation. One thing is we just talked to them about their plans for staffing. Staffing is important because AMF is a small organization. They have fewer than twenty people, and so it really, really is important to understand the different jobs that the people there do, and how they expect to work with partners to scale. AMF, as I described earlier in terms of their structure and what they do, they buy the nets, but they work really closely with other partners. And so we talked to them in a lot of detail about what the different partners would be doing in this case, where they were scaling up a lot. We talked to them about what was gonna be different this time.

For example, in the past, other funders have really been principally responsible for funding the actual delivery of the nets. So AMF would buy the nets and other funders like The Global Fund would pay for the delivery of those nets. In this case, [00:13:00] because this is a much bigger commitment, there isn't enough other funding to deliver all these nets, and so we are funding AMF to do that. So we talked to them about how they would find partners who could deliver those nets, how they would fund them, how they would identify the most effective partners to give us that confidence that they had the systems to enable that, to make that work effectively.

We also talked to other partners involved in the space, to The Global Fund, to the National Malaria Program, to the implementing partners, to get a range of different perspectives on AMF's role in DRC and understand the kind of advice that they were able to give, the role they could play beyond simply delivering the nets. And we pretty universally found that they are, partly because they've been such a long-standing partner in DRC net campaigns, that they were a really important part of the ecosystem and that everyone really valued their contribution beyond just providing many millions of nets.

Elie Hassenfeld: For a group like AMF, as it has to [00:14:00] scale up to larger size, what are the places where this scale starts to strain AMF's ability? Like, were the areas where you were most worried about their ability to be able to handle this larger scale, or where they most needed to adjust to be able to address a larger scale grant? And just as part of this, if you can say how much larger scale was this for them than what they had supported previously?

Alex Bowles: So one part of this is the net ordering. So this is just a much larger volume of a particular commodity than they have had to order before. And so, we wanted to be confident that they had the procurement processes and capacity to engage with manufacturers on this much larger scale. I don't have the exact numbers at my fingertips, but I think in this, the three-year period that this grant covers, we're talking about maybe three times as [00:15:00] many nets that they're procuring as last time. That's on the sort of procurement end.

Elie Hassenfeld: And then let me just ask you about what are the kinds of things that you imagine could go wrong or they would have to be able to handle? Like when you think about like concretely, what could be challenging about ordering three times as many nets, what are the kinds of things that come to mind?

Alex Bowles: So some things that come to mind, for example, I expect that you are more likely to have to order from multiple different suppliers in that context, which means that you are going to be managing supply of these nets coming in from different locations, shipping from different locations. You're going to have to manage the quality control sampling. And so some of that is just more processes, and more work, and more chances for things to go wrong. In some ways, it's not even more complicated than that.

Elie Hassenfeld: Got it. And so I guess I sort of entered this conversation with the overarching frame of, “Well, GiveWell supported AMF forever. What are the questions we'd have today?” And like one of the questions is, well, we haven't supported AMF at this scale in the past, and AMF has not [00:16:00] operated this scale in the past. And, even just one little piece of what they do, this purchasing nets and ensuring that effective nets get to the place in the country where they are needed, is more challenging at this scale than at their previous scale. And so that was a meaningful part of the investigation that we went through before committing to this grant.

So we've gone through the first two of the questions you raised. The third question you raised was this question about what is the malaria burden in DRC? So just explain first what that is and why it matters.

Alex Bowles: The malaria burden, when we talk about that, what we really mean is, who is getting ill, and who is dying of malaria, and how many people is that affecting? And that seems like a simple question, but it's actually really hard to answer, particularly in a place like DRC. And that's because DRC doesn't have good health records. It is really hard to get direct evidence of who is dying of what. So that [00:17:00] means we rely on more indirect methods. So there are some surveys, where they go and ask parents about the life histories and death histories of their children. And children are the ones who are most likely to die of malaria.

We're quite reliant on modeled estimates. So these are estimates that use all sorts of different inputs about the burden of disease to produce these estimates that account for the fact that there will be underreporting, because some people will have died and it'll never have been reported and things like that, to produce overall estimates for how many people are getting sick, how many people are dying, both at the country level and to some degree, at the subnational level.

We don't fully trust those modeled estimates because there's a lot of things that go into those, and it's just a really hard thing to do. So we try to triangulate there's more than one model. We try and sort of weigh them against each other. And in particular, for subnational, so for how many people are affected by malaria in a given province, we try to, to the greatest extent [00:18:00] possible, rely on these survey data. So things that are more rooted in data rather than models. But it's a really difficult problem, particularly when you get below the national level.

Elie Hassenfeld: And I guess like just to help describe how this sort of matters for GiveWell overall. You know, at GiveWell our job is to direct money where we can save or improve lives the most. And that might mean giving $276 million for nets in DRC as we did here, but it also might mean giving some larger amount to cover even more people with nets, and it might also mean giving a lesser amount so those funds can be used elsewhere.

And so a big question that we're asking in our grant investigations is, how much impact per dollar will this grant have? And so those burden estimates, both overall and then in the individual provinces, make a big difference to the question of how much of funding should we be directing to this program versus all the other programs that we could potentially support.

I think this even predates your time at GiveWell, but [00:19:00] doesn't predate my time because that would be hard to do. But you know, a few years ago we did this project that we called red teaming, where we tried to look for problems in our analysis, and I think…I believe we did this back in 2023. And at that point, one of the primary findings, which if anyone is interested you can see on our mistakes page, is that we were relying too heavily on a single source for estimates of malaria burden. And for the reasons that you just described, when we looked more closely at it as we grew as an organization and had more capacity and time to dig into things, we recognized that this was a big mistake because there were so many different sources of information that could lead us to a better estimate.

And so this is just one way in which the GiveWell of 2026 is doing a much better job than, say, the GiveWell of 2022. And I think it's really amazing to see that improvement where, as we've grown, we have this increased rigor in our work, where I think we've been able to scale to do more, but also do it at a higher level of quality than we were able to do in the [00:20:00] past.

I wanna keep moving through your questions. So your fourth question was this really big, kind of meta question, which is, GiveWell is now the biggest funder of vector control for DRC. That's different than where GiveWell started, which was making grants of tens of thousands of dollars to organizations all the way back in 2007. So, talk more about this question, how it comes up for you, and what's on your mind when you think through it.

Alex Bowles: So the way I think about it is, and you were getting at this in your framing of the question, Elie, is for a lot of GiveWell's history, we've been accustomed to think of ourselves as operating as a marginal actor. So we're adding on a little bit to, and particularly in net campaigns, we're adding on a bit to what is already there. So, if with nets, and this is the way we've historically often worked with AMF, if there is a campaign happening anyway, maybe we can make it bigger, so we can provide more nets, and that campaign is going to be a bit bigger and cover more people.

But in this [00:21:00] circumstance, we have to think about things a little bit differently, because if we are providing—and directly with AMF, we are providing all of the nets for this campaign—we're not quite operating at the margins in the same way. We still have to think about it that way in some ways, right? There is a campaign that would happen if we didn't support it, and there's a campaign that happens if we do, and it is bigger. So that is still useful, but all of the other partners behave a little bit differently, I think, when you are the biggest funder. And so it becomes really important to have those relationships and for AMF, in this case, to work well with all the other parties involved, because they become more and more essential to the overall success of the campaign.

And in DRC, I should emphasize for our listeners, this is a really complicated logistical thing to do. DRC is a really big country. It has very poor road infrastructure. And so we're moving these nets around to really difficult places to reach. And so they do these rolling campaigns. So, rather than one big national campaign, there are campaigns [00:22:00] happening kind of all the time in one province, then another province, then another province, and this is just a huge logistical undertaking.

So where we are the single biggest funder, I do think it becomes much more important to understand the overall structure of how this works, how all the parties work together. Can they be effective together? Because that makes or breaks our support in a way that I think just doesn't operate the same when you are a relatively marginal extra factor.

Elie Hassenfeld: In some ways it's clear that GiveWell's relationship vis-à-vis this distribution is very different than what GiveWell's was in the past. It's different to come in and add the last 1% of funding needed for something to take place, versus some very large proportion of the funding where you're…you know, our donors' funding is the single largest contributor to this whole project.

And then I'm curious about how big a difference this makes. And I guess my hypothesis is that it does mean that we have an [00:23:00] opportunity to contribute to how the whole distribution takes place. It's like, we can suggest and support things that we think will make the whole program work better, while obviously listening to people when they disagree. Like, we're one voice, not the only voice.

But then I would guess that in many ways, it doesn't really change what GiveWell is at the end of the day. I mean, when I just think about the three open questions that you listed, in some ways they just go back to the basics of GiveWell, which are, do we really know that it works? Can we understand this organization and its value added in, in whether it's operating effectively? And how cost effective will this program be? It's like we're asking all the same questions. The questions themselves are operating at a much, you know, much larger scale, but the sort of core of does it work, what do you get for your dollar, you know, et cetera, are just the questions that we've been asking forever.

And so, I'm kind of most curious about the extent to which this feels like a big change versus it's…I don't know, an additional opportunity to support effective [00:24:00] international programs where we now have a seat at the table we didn't have before, and it's worth recognizing the impact we can potentially have and the responsibility that comes with that.

Alex Bowles: I think that's right. I think the core of what we are thinking about is still, is this among the most cost effective things we can do with the resources we have? And the basics, as you say, of, is there a need? Is this going to effectively address the need? Those are still there, and being confident in those was absolutely still at the core of this investigation.

I think what has changed in my view—and I obviously have less view of the long run than you do at GiveWell—is we have to be more aware of how what we might do affects others. And so there's upside to that, as you said, it gives us, in some ways, influence to suggest improvements that we think could help make things more cost-effective, reach more people. And I think that we also have to be [00:25:00] aware that if we are coming in with a lot of funding, that might change other funders' behavior, and that's something that we've had to think about a lot as well, and might change their decision-making. And so that is something that is just a bit different when you're operating at the scale we are now.

Elie Hassenfeld: Yeah, can you say more about that? Like, this is another area where, for a long time, even at smaller scale, GiveWell was still providing a pretty significant amount of support to country malaria programs, to malaria net distributions. And so we were, and have been, concerned about this question that we often call “funging,” the extent to which money we give could cause another funder to take their money and do something else. And so in that way, the money we give might not have the effect that we expect. And so we've been thinking about that for a long time. Obviously, at larger scale, those effects are even larger, perhaps even more obvious, and it will affect others' behavior even more. Say more about how you think about this at larger scale, and what's most on your mind beyond the like, or maybe it is just [00:26:00] like, bigger scale, bigger problem.

Alex Bowles: I think in part it's bigger scale, bigger problem, but I think there are a few other things. So, one is that it becomes more important for us to think about and to know about and learn about the other things that the money could do.

And what I mean by that is if we are, in the way that you described, influencing another funder to withdraw funding from something, from a net campaign or something like that, it matters a lot what else they're going to do with the money. Because if they are going to withdraw it from this and spend it on something else that we think is similarly cost-effective, then we don't really mind very much, because we're just causing a transfer from one cost-effective thing to another cost-effective thing. If, however, we think they're gonna move it to something that we think is much less cost-effective, that's obviously more of a problem for us. And that means we should understand, for example, in this context in malaria, we want to understand more, well, what are the other things [00:27:00] that we do in malaria? How cost-effective do we think those are? So that we can reach more of a judgment about the effect on our overall cost-effectiveness of that effect. Because you can't ultimately prevent, entirely, this “funging,” as we call it. It is a fact of the world, people act in response to the things that we do, and we wouldn't expect otherwise.

Elie Hassenfeld: And so do you have thoughts about this problem of funging with respect to this particular grant? How did you think about that question here at this large scale?

Alex Bowles: So the other major funder of malaria programs and net distributions, in particular in the DRC, is The Global Fund. And so, it has been important for us to get a sense of what The Global Fund, which works closely with the national government to set its priorities, what they were planning to do, because they have three-year cycles, they're at the beginning of their three-year cycle, so they are also making decisions about what their funding could support. They're still in the final process of doing that, so I can't really say [00:28:00] anything too specific about where exactly they've ended up. But it was front of mind for us to think about what they might want to do, and how The Global Fund and the government might adjust their plans on the basis of these funds.

Elie Hassenfeld: And so, how did you think about…I mean, it just seems like a really hard question to answer, especially when you're going to come through and commit a very large amount of money up front. So yeah, how did you go about trying to answer this question?

Alex Bowles: One of the reasons that we wanted to explore making a grant of this size was that we do believe there is a significant advantage to providing clarity about the available funds at the beginning of a cycle like this, because it makes planning easier, essentially.

As I mentioned earlier, the logistics of net campaigns like this are really tough, and so if you can avoid the program having to replan as new funds come in, that's a good thing. We think that that is more likely to make campaigns effective. So we came [00:29:00] in with that mindset. We knew, however, that also coming in earlier carried this risk that it made it probably more likely that other funders might adjust.

So we just tried to trade off these risks, I mean, one thing that is important is that there is a lot of need for funding. There's a big funding gap for various malaria programs in DRC. And so we don't expect that those funding gaps are going to be filled. There is still need, we don't expect that money is going to be moved away from malaria by The Global Fund, so that does reduce the potential consequences from our perspective.

Elie Hassenfeld: Yeah. Let's talk about the last one of the things that we listed out at the beginning, which is this question of how long nets last, and whether the default distribute nets every three year approach is really the one that seems optimal.

Alex Bowles: So this is a problem that we have been thinking about for a long time, and AMF has been thinking about for a long time, [00:30:00] because we know—thanks in part to the monitoring that AMF does after the distributions it supports and thanks in part to surveys funded by other groups like the US government—that nets don't last three years in many contexts. It varies by country, but usually they don't. They often last maybe two years, and that varies between nets, so I'm talking about averages. But that means often there is a gap of maybe a year in a three-year cycle, where children are not protected by the nets that were distributed two years ago. And you can see this in the burden. So you see more kids getting sick and dying in the third year after a distribution, it's very clear in the data.

So we've seen that the lifetime of nets is shorter in DRC than in other countries. There's a lot of discussion about why this might be, and I think there are a lot of factors that go into net durability, including climate. So, in more humid contexts, nets don't last as long. But it's also about how they're used, the bedding they're with, the housing they're with, all these kind of things.

[00:31:00] So DRC had already taken steps to address this, they had changed their policy to aim for distributions every thirty months, so every two and a half years in the last cycle. But we did think that, given that we think that nets last maybe a year and a half or two years in DRC, that there was room for improvement there. There's still this gap, right? And we saw this in the lookback that we talked about earlier. Because we had the potential to bring these extra resources for these campaigns, thanks to our donors, we wanted to look at, do we think it would be cost-effective to reduce that even further? Do we think it would be cost-effective to support the country to try distributing every two years?

And our best guess is it is cost-effective to do this. The burden is sufficient, and it really will help to protect children during that kind of gap. So what we have, in collaboration with AMF, decided to do is make funding available so that the government can, working with AMF, working with The Global Fund, test this approach of distributing nets every two [00:32:00] years in a small number of provinces.

To return to the theme of logistics, this is a big logistical undertaking. You're adding more campaigns in a place where it's hard to do campaigns. But we think the health impact is worth it if it can be done. And so we wanted to start, and we understand the government is on board with starting with these five provinces, to see if we can successfully have these nets distributed more frequently. We're going to make sure that we are supporting evidence gathering so we can see whether it worked. We want to provide support for that and learn from that. And hopefully that'll have a big impact, and that's something that we and others will be able to support in the future.

Elie Hassenfeld: Yeah, and it's really interesting because it's just a way in which, there's this default mass campaign every three year approach that, I guess I remember being the default for a very long time. And I don't know, using the lens of cost effectiveness that we have, the question is just, well, what is the optimal timeframe for net distribution to save and improve lives most with the money that we have? And, that might be three years, [00:33:00] that could be more, it could be less. And so, based on what we've been able to find, we think it's fewer years, it's doing it more often. You know, hopefully we'll learn more about how this goes in the future. Do you know when we're gonna have some initial results from that to see how it's gone?

Alex Bowles: I think we should be able to get some initial results in about two years, because some of the first campaigns hopefully will happen two years after those previous campaigns that'll happen in the next couple of years. It depends a bit. So the country is in the process of planning its campaigns for '27, '28, '29 now, and so it's hard to be definitive exactly on the timelines there.

Elie Hassenfeld: Yeah. Well, it's very cool. I mean, I think it's just interesting to think about for a lot of people on the outside, it probably seems like GiveWell's been recommending AMF forever, what is there even left to do? And there are both a lot of questions that I think need to be answered, especially when we're talking about funds of this scale, and ways in which we're able to take the things we learned from the grants that we've made in the [00:34:00] past and hopefully support even more effective malaria programming in the future.

I wanted to shift gears a bit away from what were our questions and instead, just looking at the short period, at least relative to GiveWell history, the roughly two-year period where you've been doing these malaria-related investigations at GiveWell. I'd just like to hear a little bit about what has changed in that small period of time. I think we've grown our team a lot during that period, and we grow the team so we can make better decisions about where to direct funding. And so just would love to hear some of what, what you've seen, ways that we're able to do more or do it better, relative to maybe the first investigation you did, which I think was almost exactly two years ago at this point.

Alex Bowles: That's right. And that was for a grant to AMF as well. 

So I think the first I would say is one that you anticipated, which is more capacity matters a lot. And an example of this is the lookback we were talking about earlier. Two years ago, we would not have been able to dedicate the staff time that it [00:35:00] took to look in detail at that previous campaign. And when I say look in detail, there are a few things. So one was the sort of quasi experimental study of looking really at the data about when that campaigns happened, and how much disease they prevented. And that took quite a bit of staff time. And it also…that extra staff time enabled us to commission these short-term qualitative field surveys, that in a very short amount of time, a couple of months, were able to help us answer questions about what it is really like to deliver nets in DRC, the concrete, specific challenges that they face doing that.

And one of the reasons that matters is because in our cost effectiveness analysis, a parameter we have, which is essentially, a discount because it is really hard to deliver nets in DRC. And this extra capacity and this ability to put surveys into the field was able to really help us substantiate what had been, I think in the past, something of an informed intuition, with more [00:36:00] concrete examples, which helped us feel more confident about that.

Elie Hassenfeld: So tell me, about what that was. So, to just walk through the process, like at some point in this grant investigation, you, someone on the team decided it would be great if we could get this qualitative information. And then we went out, like what information did we get back on what time frame, and how did it affect our eventual decision?

Alex Bowles: Sure. So we decided that it would be a good idea to get independent information from DRC in places where AMF supported net campaigns had happened. And so we identified a firm based in DRC that could conduct these kind of field surveys, so send people out into rural areas where campaigns had happened to ask households about their experience receiving nets, whether they did receive nets. We were able to cross-check that with records from the campaigns, and so that's a sort of check on the quality of those records. But also to ask [00:37:00] qualitative questions about why nets might have missed a particular household. And to talk to community leaders about challenges that they'd had with campaigns.

And so we got back these rich, dense, qualitative reports about just how the campaign seemed to have gone and what the obstacles were. And that has a kind of twin function. One is that we're able to share these findings with AMF so that they can learn from them. But it also helped us understand, and in some ways feel confident about what had happened, because I've described problems so far, but a big thing that came out of it is, most people got nets. The nets came to their household. They, you know, generally this was working the way we'd hoped. But it also included some of the challenges. And so it's this kind of extra confirmation of the general picture that we had to complement this more econometric approach that we were taking in terms of understanding the health impact of the campaigns. And that just takes more people and more time.

Elie Hassenfeld: Yeah, and I think those sort of qualitative, or just real world [00:38:00] observational information to pair with the intense quantitative approach is really important to make good decisions.
You know, in some ways, the lookback and the qualitative surveys, we're still digging into that study and getting it replicated and reviewed, but the estimate of mortality reduced in the lookback—pretty close to what our model was. If I recall correctly, the overall take of wastage, meaning nets that don't get where they're supposed to, from the qualitative survey was pretty close to the number that we had in our model. And we've obviously had conversations at GiveWell about times that our models were off, so that is not always the case. But maybe that is surprising, or it's even not surprising. We put a lot of work into figuring out what is happening here, and so to some extent it's heartening that we were pretty close in our forecasts to what actually occurred in this situation.

Alex Bowles: I agree. And we were talking earlier about what we can do in 2026 that we maybe couldn't do a couple of years ago. But I would also say one of the outcomes from this is, oh, we were doing some pretty good work in 2024 too. [00:39:00] Yeah. And it put us in a really good spot to be able to expand in the way that we have.

Elie Hassenfeld: So, I want to kind of move off this and ask you about what's next for malaria vector control. But before we do that, anything else you wanna add about AMF or this grant in particular?

Alex Bowles: I mean, I would just say we are really excited about this grant. It's just a chance to have a huge impact in a country that can really benefit from these nets. People in DRC will, we hope, really benefit, and we're really glad to be able to support AMF and all of the partners in the country, who will deliver these nets, the government that will manage these campaigns. So, yeah, just really excited about that.

Elie Hassenfeld: Yeah, for sure. Okay, so I mean, to the extent you can just share briefly, what is next for malaria vector control? And I think in particular, one narrower question here is, to what extent do you think malaria nets will be the primary program that malaria vector control supports versus others? And are there other vector control [00:40:00] interventions that you think are particularly promising and might be on GiveWell's radar, GiveWell supported in the future?

Alex Bowles: One of the nice things about having more capacity now is we can keep more things on our radar, which is great. I will say I expect nets to be at the core of what we do in vector control in the short term. They're a really effective tool. The newest nets we have are particularly effective, and I expect us to support them and for them to be a core tool for some years to come. That said, there are other new tools that we are looking at really closely.

So one is a class of tool called spatial emanators. So the way they work is they're these little squares of plastic basically, or rectangles of plastic, impregnated with an insecticide, that emanates into the room where they're used. And we think that that affects mosquitoes both by repelling them, but also by interfering with them in ways that stops them biting [00:41:00] people. There is some evidence that they're pretty effective in preventing malaria, so people hang them in their houses, and it seems to prevent bites and prevent malaria. We still have some questions about that evidence base, and so we are looking closely at supporting more randomized trials to improve the evidence base, particularly in the contexts we care about, high-burden areas in Western Central Africa in particular, because the existing evidence is more from East Africa.

But I definitely think there's a world in which we want to really support the scaling up of that tool in the future. It has some advantages over nets. You don't need to actively choose to sleep under it. You just hang it on your wall, and it's working twenty-four hours a day.

Elie Hassenfeld: Insecticide emanating inside of people's homes sounds very frightening. How is that safe? And what've they done to check that?

Alex Bowles: Yeah, that's a really fair question. So they have done studies to measure the accumulation of that insecticide in indoor spaces, and the accumulation of it on surfaces, so it does accumulate a little bit on [00:42:00] surfaces. I was actually at a facility in Liverpool recently where they have these metal boxes that are specifically designed to do this kind of study to measure it landing on surfaces and things like that. And so there is some evidence being built there about the safety side of that, but I think there are still some questions. I will say this class of insecticide that's being used is a pyrethroid. They've been used for quite a long time. It's a similar class to what's being used on nets, and so we feel reasonably comfortable with that. But I think more work could be done.

Elie Hassenfeld: Are there other new programs potentially on our funding agenda in the future?

Alex Bowles: So something that's a little further out, both in terms of, it'll take longer for it to be ready, and also it's a wider departure from what we've supported in the past, is gene drive mosquitoes. So this is a technology for genetically modifying mosquitoes so that either they cannot transmit malaria, so they're modified so they're no longer a hospitable [00:43:00] host for the malaria parasite, or they're modified so that they can't reproduce properly and so that you crash the population of that particular mosquito species.

These technologies are a little bit away from prime time, so this is not something we've been involved in, but they are potentially very promising, because they are the sort of most fundamental way to interfere with that part of the life cycle of malaria, is preventing mosquitoes from carrying it. And there's some modeling that suggests they could be really, really effective in the fight against malaria. So that's definitely something we're keeping an eye on, trying to keep abreast of the evidence as it emerges, the trials as they emerge, and want to be ready to assess properly if it matures as a technology.

Elie Hassenfeld: What are the big open questions right now about gene drives?

Alex Bowles: One is technically, so it's genetic modification including a gene drive, which is the idea of propagating a particular trait to all of the offspring, and that's how you spread it through a population. So can scientists create a reliable drive that can reliably [00:44:00] propagate this trait through a population and it stays stable?

Another is what is the distribution of different mosquito species in the different countries that are most affected by malaria? Because if you have a gene drive mosquito, that's just one species, right? But there are multiple species that are responsible for spreading malaria, and so you need to understand what the distribution of species is. And we're quite interested, because this matters for other vector control as well, in supporting better understanding of the distribution of mosquito species, particularly in Africa. So, that's an open question and an area of interest.

And then a third one is acceptability. You know, this is a new technology that not everyone is likely to be on board with straight away. And so I think there's a lot of work to do to help assess people's willingness to have these technologies deployed in their communities, to help people learn about the technology. So I think that's a big factor as well.

Elie Hassenfeld: Great. Well, thanks so much, Alex. This was really interesting to go through this grant. Thanks for all the work [00:45:00] that you did to figure everything out and make this grant possible that's going to do a lot of good.

Alex Bowles: Thanks so much, Elie.
–

Elie Hassenfeld: Hey everyone, this is Elie again. I hope that conversation was interesting.

You know, one of the things that's really on our minds, and on my mind at GiveWell right now, is, how can we scale most effectively? And that means both doing a better job at answering the kinds of questions we've had in the past, where we have the capacity to go deeper by deploying these rapid surveys into the field that give us qualitative information back. But also questions like, how does the role that GiveWell plays in the world, how does that change as we reach this larger scale where we have an opportunity to interact with the decision makers about how a distribution happens, and how often it happens in a country like DRC to improve its overall impact and hopefully help even more people.

And so we're very [00:46:00] excited about where we are today and our ability to do more work in the future, both with the increased funding we've received, and with the larger team that is enabling us to cover more ground and go deeper than we have in the past. Thank you so much as always for your support and for your interest in the work that we do.