This transcript was automatically generated using software, with additional editing afterward, and may contain minor inaccuracies.
--
Elie Hassenfeld: [00:00:00] Hey, this is Elie. It's day two of our trip. We're still in Blantyre, and we're sitting outside our hotel in the morning before we start off on our day. And I'm here with Julie Faller, who's a Senior Program Officer at GiveWell. And today we're really gonna focus on one of the big questions of our trip, which is better understanding how the health system works and the effects of aid cuts on how the health system operates. In the morning we're gonna visit senior health officials and then after that we'll be heading to a couple health facilities that are also in this region of Malawi.

So Julie, today, like when we talk to the district health official, what do you want to ask them about?

Julie Faller: So there's two things on my mind. One is that this district, as we understand, had been part of a USAID-funded experiment that was sort of an experiment in trying to give more local decision-making control over use of health funds. And so I'm really curious to hear what they learned through that experiment and then also how that [00:01:00] made the effects of aid cuts different, if at all, because they had this more local control over that set of funds. Yeah, so that's one thing.

And then the second is that I'm really interested in kind of digging in and understanding more about the supply chain of health commodities. Often in our work we hear about stockouts, like the problem where people go to health facilities and then you know what the treatment is supposed to be, but it's not available at the health facility.

And one thing that came through in our discussion yesterday with households in the village that we visited where, to a T, every person we spoke with said that they preferred to go to private facilities, if they had the money available for the fees, because of stockouts, basically because of the inconvenience of going someplace and not having the treatment. We also heard a little bit about stockouts from the clinic staff at the clinic we were at yesterday, and so I'm interested with both the district official and in the hospitals that we visit today to understand more about their perceptions of what's going on with the supply chain, what has been tried to try to fix these problems of stockouts, why do they persist, and just getting more of a window of the health system by digging in there.

Elie Hassenfeld: Got [00:02:00] it. Can you say a little bit more about this USAID experiment to give more local control? This is sort of a big topic, like an area of criticism for USAID, that they worked through all the US contractors and US NGOs. And so presumably this is an attempt to try something different. Do you know more about the background of this and what they were trying to see and how they might find it out? And then what are the kinds of things that you could hear that would affect how you think about what we should be doing?

Julie Faller: I just briefly skimmed some project briefs on this yesterday. And so I don't know a ton about the background, but it seems to me like it was more responsive to the idea of local versus national planning rather than like local versus international NGO or something.

And so I think the idea is, basically, district officials are the people who have the best information about what is going on in their area and also the sort of language of the project was around efficiency and accountability. And so I think also the idea is that voters can [00:03:00] know who their district commissioners are and hold them more accountable for public services. Whereas if you have a parliamentary system, it's kind of difficult to figure out who should you punish if you're not happy with how your hospital is running, for example.

Elie Hassenfeld: And then do you think that could affect how—you know, you manage a lot of the different grant making teams like water and nutrition, et cetera—do you think learning about this could affect our work? And if so, how?

Julie Faller: Yeah, I think definitely. So we're, one question that's on our mind across the research team is just making sure that we're getting contextual on-the-ground information that affects program functioning and thinking about how to use that to help work with our implementing partners, our grantees, to help design better programs.

And so I think if there are examples from speaking with the district commissioners of specific solutions that they came up with, I think that's just more information that we should be leaning into this further and we might try to expand on in the future.

Elie Hassenfeld: Got it. Yeah, that makes sense. And then let's talk about stockouts just for a minute. So, Teryn and I were talking a little bit about that yesterday. It came up in the conversation at the health facility where they were talking about persistent [00:04:00] stockouts that were independent of aid cuts. And so, I'm kind of curious just like, how do you see that question? And if there are things you're going to be particularly focused on today and trying to get, like triangulate that issue more?

Julie Faller: Yeah, what I would really like to be able to do is have a sense of the magnitude of importance of different factors that drive stockouts, because I think probably there's at least some corruption or diversion, probably there's at least some problems with record keeping and inaccurate forecast for need, probably there's at least some problems with operational issues of, like, are the trucks able to get where you expect them to go.

And I think that trying to understand, particularly with the district commissioner, perception of the magnitude and the importance of these different issues feels like it would be really helpful in trying to think about what's tractable, what's solvable, and what's not.

Elie Hassenfeld: Right. And how about the health facilities? Do you think there are things we can learn there, or it's just maybe you'd say it's more, you know, data gathering from people who are experiencing the stockouts themselves and seeing what they experience and what they say. [00:05:00]

Julie Faller: Yeah. Well, at the health facilities, I'm really interested if they're open to it, to try to look at the pharmacist record keeping systems and just understand practically, like what is the process by which they keep track of what drugs are in place? Do they have to do this pen and paper? Is there a system? If so, does it seem to be collecting the relevant data? Like, how much room is there for user error and how annoying is it?

Elie Hassenfeld: Yeah. Cool. Yeah, that makes a lot of sense. You know, one of the big goals for this trip is focusing on the effect of aid cuts on the health system, and the health system and how it functions overall, by just like being here and talking to a lot of people. So I know, I hope we learn a lot.
--

Elie Hassenfeld: We started the day meeting with three government officials about the impact they'd seen so far from this year's foreign aid cuts, and also their thoughts on what it could mean for their future. One thing that they mentioned throughout the conversation was that the country's health data indicators didn't yet show the impact of the cuts, because most of the programs that had been terminated were preventative instead of direct treatment. But they were very [00:06:00] worried about the impacts on health in the future.

We asked them if we could look at the data indicators ourselves, and they said that we could, at a nearby health facility where we were going to visit next.

When we arrived at that health facility, we parked in a fairly large lot adjoining, something that looked familiar to us now, which was several long one-story brick buildings that were connected and served as the health facility. We were first welcomed there by a number of hospital staff, and then the GiveWell team split up into different groups to look at parts of the hospital.

I went to the section where they had large wall charts showing data from deliveries of babies in the maternity ward, looking at questions like the proportion of births that had had a major complication, whether any babies had died in the process of giving birth.

And then we also wanted to walk [00:07:00] over to another building where we could see on a computer, data from Malawi's national data system. We sat down with the person in charge of describing the hospital's data, and he walked us through some of the data that they had.

--
Hospital data analyst: Okay, so you can see, this is just the reporting status of a certain program, which is antenatal. In terms of completedness. So you can see . . .

Elie Hassenfeld: Is there a way to see all the indicators that we can look at?

Hospital data analyst: Let's, let's see the indicators.

Elie Hassenfeld: Let's see what else is here.

Hospital data analyst: So we can choose the indicators of the different programs.

Elie Hassenfeld: All right. So reporting completeness. How about like immunizations received or some indicator like that?

Hospital data analyst: Yeah, yeah, yeah. We have them.
--

Elie Hassenfeld: You know, one of the things we found it was fairly difficult in the time that we had to come up with analysis that we could [00:08:00] run on the data that would enable us to see what effects the aid cuts might have had.

We looked at things like the number of children being vaccinated, people's access to family planning, access to HIV treatment, and in all of those cases, we weren't able to see a meaningful effect in the data. There was a lot of movement in the data that was reported, you know, some months with higher levels and lower levels, but in the time that we had, we weren't able to suss out what the driving factors were for those changes over time.
--

Elie Hassenfeld: So how about indicators for accessing family planning? Are there indicators like that?

Hospital data analyst: Yeah. We have some indicators for accessing the family planning. Now you can see here it means, uh, there had been a gap in terms of family planning service provision. So since there, there are just injectables.

Elie Hassenfeld: Can we add [00:09:00] another, like the oral contraceptive to this same chart?

Hospital data analyst: Yeah. Yeah.

Elie Hassenfeld: So do you think when the Depo-Provera is stocked out, people switch to oral?

Hospital data analyst: Yeah. Yeah. They'll choose the method, which is available at that time.

Elie Hassenfeld: Right. Let's see, so this one was injectable, and then where's oral? Is it up . . . this [is] combined oral?

Hospital data analyst: Yeah, combined oral.

Elie Hassenfeld: Right, and then this went way up. And this came way down.

Hospital data analyst: Yeah, exactly.

Elie Hassenfeld: But then over time . . .

Hospital data analyst: Yes, over time, exactly.

Elie Hassenfeld: Pretty flat.
--

Elie Hassenfeld: We were talking about the fact that it was challenging to see effects in the data, and we knew from the day before that we should see significant effects on viral load testing because we understood that funding for viral load testing had been cut off for a significant period of time. And to show us that, the data analyst took us to a [00:10:00] different smaller room, with a different kind of data-capture system.

There he pulled up records of the number of patients who had received viral load testing in early 2024 versus the ones in 2025. And there we saw a really substantial difference. In 2025, a much lower proportion of people had their viral load tested than the previous year.
--

Hospital data analyst: So we have the viral load reports.

Elie Hassenfeld: Oh, so these are just individual people's names we're seeing now.

Hospital data analyst: So these are, that's the individuals, and this is the first name that is.

Elie Hassenfeld: Oh, I see. So you have to take all this, then you would export this?

Hospital data analyst: Yeah. Yeah.

Elie Hassenfeld: And then you could visualize it. But that's all, that's . . .

Hospital data analyst: Yeah, we have the seven pages for this one.

Elie Hassenfeld: So this is seven pages in February of 2024.

Hospital data analyst: [00:11:00] 2024. Yeah. So let's try to look at, let's try to, say, February 2025.
Elie Hassenfeld: I see, okay.

Hospital data analyst: So just did one to one page, this was because the stop order now.

Elie Hassenfeld: So you just stopped immediately?

Hospital data analyst: We stopped. So you see we had seven pages. That was last year.

Elie Hassenfeld: So this is like 25.

Hospital data analyst: Exactly.

Elie Hassenfeld: And then what we saw a year ago was like 140 or something.

Hospital data analyst: Exactly. That is so, because that was due to the stop order.

Elie Hassenfeld: And this is for this facility?

Hospital data analyst: Yes.

Elie Hassenfeld: Is what we're looking at.

Hospital data analyst: So there had been really, with the stop order, there had been an impact with the withdrawal of USAID. There's been really a big impact. Particularly in terms of data management. So there had been activities that were being supported, but they all stopped. Of which now we expect a big drawback in terms of data management. We had activities, [00:12:00] that we had to bring the facilities together to review the data to see how best we can improve the indicators, but all that is stopped, yeah.

Elie Hassenfeld: Well, thank you so much for your time.

Hospital data analyst: You're welcome.

Elie Hassenfeld: And for taking us around. Really appreciate it.
--

Elie Hassenfeld: Hey, this is Elie. I'm sitting here with Julie. We're at the end of day two, where this morning we visited with some officials, and then a couple health facilities. And, you know, just trying to further our understanding of the health system and how it's affected by the cuts from USAID. So Julie, what did you think of that first conversation with the officials?

Julie Faller: Yeah, that was a really interesting conversation, Elie, where we were trying to learn about this USAID program that had been cut. And this morning I thought that the program was mainly about local control and kind of like efficiency and accountability through local [00:13:00] control.

And I think that was part of it. But then something that we learned was that it was just also a massive infusion of resources into the district that they were able to use for a variety of things like hiring people into the health service, like commodities, a bunch of other stuff. And so I came out of that just thinking about the very sudden withdrawal of resources that came along with the cancellation of that program.

Elie Hassenfeld: Yeah. And so like local control, so that's different because normally USAID works through large NGOs and . . .

Julie Faller: Right. Yeah. Normally they would be working with implementing organizations, these large NGOs, and in this case, it's not like they were funding a local NGO. Instead, they were giving transfers directly to the district government, which is this administrative unit. And then the district was using it to make decisions about how to allocate their funds like they would normally in day-to-day operations.

Elie Hassenfeld: Yeah. I mean, two of the things that I thought were interesting and telling about that meeting was, you know, they gave us a lot of figures of how things had changed. So they [00:14:00] had hired, via this program, 83 full-time health staff and a hundred something temporary staff or consultants to do outreach. And they just had immediately laid off all those 120 plus consultants. And so that's just like this direct loss on general outreach. And then they also made this point because they had to lay off those staff, they also, what did they say, they had 51 health facilities of some kind in the district, and then 40 of these mobile clinics. And they maintained the permanent health facilities, but shut down all the 40 clinics doing outreach in the community.

Julie Faller: Right, it was those temporary staff or volunteer staff who weren't officially part of the permanent staff, had been the ones who were doing those mobile clinics. And so they just had nobody to run them anymore.

Elie Hassenfeld: They showed us that, that map where there's like an island in a, I guess like a lake, that was part of the district, and they were pointing out like those are particularly hard to reach people. When we had the temporary staff, we could take a mobile clinic and reach them for immunizations or, you know, other outreach. And now [00:15:00] their expectation is those people are just not going to be getting health services.

Julie Faller: They're not going to transport themselves to be able to come to a facility.

Elie Hassenfeld: And, and yeah, we were sort of trying to ask them where they'd see these problems showing up in health data. And they basically said, we don't expect to see it anytime soon. Because a lot of what the outreach was doing was encouraging people to be tested for HIV, TB case finding, TB testing, and that kind of activity stopping now just means that there'll be health effects in the near future. And that's what they really were talking about, like fearful of the future.

Julie Faller: Right. And another of those activities was like childhood immunizations, which again, not immediately after the baby is vaccinated, but maybe in a year or two you see the effects of them not being vaccinated.

Elie Hassenfeld: Yeah, the way they said something, like at the end, we're fearful of going back to the way it was in 2000, 2001, when I think they said there's almost like a funeral every day. And I [00:16:00] guess that was when the HIV/AIDS epidemic was really raging. And it sounded like they'd made a lot of progress here in the last 25 years. And now they're just very fearful about the idea of going back on all this progress they made.

Julie Faller: Yeah. That was a very moving and tangible image of something that they had lived through and could easily imagine.

Elie Hassenfeld: Yeah, yeah. And so then, we left the officials and we visited a couple health facilities, and one of the things that we talked about earlier was, learning more about supply chains and what we could do to ensure that, you know, is there something GiveWell could do that would ensure that needed health supplies are available when people need them. I know you've talked to a couple people in the health facilities about that and just curious what you're thinking now, what you learned.

Julie Faller: Yeah, I mean, so we were talking at people who were in a sort of primary health center level, sort of first level, and then also at a district hospital level.

And the story that came through to me from speaking with those pharmacists who are responsible for the stock supply was basically like, yeah, there are some issues with the system. You know, they're doing [00:17:00] manual data entry and then they have to add it to an electronic system. So it's inefficient, it takes time.

But basically that whole system kind of works. So maybe you could tighten it up a little bit, maybe you could make it easier, but they kind of feel like it works. The problem is, when they need supplies and they're simply not in the central warehouse because they're not available. One of the pharmacists was like, speculating, maybe it's because of like the forex problems where Malawi's currency has been devalued, and so they have trouble purchasing commodities on the international market, maybe that's contributing to it.

If you credit what we heard today, the implication is this is a funding problem. The funds are not available for the commodities and you should be able to just purchase them and make them available. And they particularly highlighted that the sort of essential medicines category, which includes routine supplies like sutures, as well as certain antibiotics, that that's an area where there are often stockouts, and also family planning. And so I think that those are two sectors that we've heard similar things in the past, and I think it's worth looking further [00:18:00] into.

Elie Hassenfeld: Yeah. One of the things that I found interesting in the facility visits was, you know, people were telling us that when the stop work order happened, the ministry issued guidelines about how they should respond.

And it seems like the guidelines really prioritized in-facility services of critical activities. So you know, they told us that they had been planning to scale up training for the HSAs, these health assistants that go out into communities, to learn more about provision of family planning services. And they had to stop that training because they didn't have the funding. You know, the idea was they're already going out to communities, let's empower them to do more because that would be more cost-effective.

They also had this plan to send other assistants out with a children-under-five kit. You know, so things like blood pressure monitor, I guess that's for pregnant mothers, a scale to weigh children. And that was something else where the funding fell through for the equipment. And so they'd invested a lot in this activity, but they weren't going forward and instead ensuring they could [00:19:00] maintain the in-facility activities, which included things like HIV testing. A wild example that they said, in order to maintain HIV testing, they had staff who had been hired to clean the hospital who they then trained to be HIV testers. And it's like that's how they solve the problem. I think it obviously should raise the question of, is that possible? Like, does that work? But yeah, that was the solution that they described.

Julie, anything else that you are thinking about after today or about halfway through the trip?

Julie Faller: Yeah. I think the other thing that's coming through is just the complexity and the need for all of the pieces of the system to be working together. We saw that yesterday, I think today I was hearing from some of the pharmacists about how they have a system where if within the district one health center needs commodities and another health center has them, they have like [00:20:00] a reporting system so that they can say, hey, take it from health center A to health center B.

But the way that they were actually transporting those commodities was relying on NGO implementing partners who could, you know, actually like put it in the car and drive it across the district. And so when that stop work order happened, they couldn't do that redistribution. It's back up and running now, but if the implementing partner has to drawback again, the ministry does not have the funds to be able to do that kind of ancillary work.

Elie Hassenfeld: Right, right. My big takeaway was like this big question that has been on my mind, on our minds this whole time, you know, the last six months was, with all these cuts, are there things we should be funding that we're not?

And I think being here is just helping bring into focus like why it's so challenging to find those programs. So I mean, you said this to me when we were just driving today, but when we supported malaria prevention programs, well, we know those really well. So we could look at them, understand them, see that they needed funds, be confident that we knew they needed funds, and direct funds.

[00:21:00] But you know, then issues come up like, should we step in because there's not enough funding for viral load testing for HIV or to ensure that the community outreach that the assistants are doing continues to happen? So we just don't know enough to say, and then it's really hard in the short term to see the effects, like the demonstrable health effects of the cuts because they're not here yet. Really, that seemed like very consistent across everyone. Like Malawi made this decision to reduce community outreach, which is heavily preventative, and focus on in-facility work. That’s at least, you can understand their reasoning.

And because of that, it's hard, you know, we don't, we don't know enough to really have a sense of how cost-effective or whether we should be funding the things that they stopped funding. And seeing it closer up, the effect of the cuts makes it easier for me to understand the decisions we made and why and, you know, they seem, at least so far, reasonable.