Elie Hassenfeld: [00:00:00] Hey everyone. I'm Elie Hassenfeld, GiveWell's co-founder and CEO. Today we're gonna talk about an area that's new for GiveWell, which we call family planning, and it's the provision of contraception to people who want it around the world.

One of the reasons we're talking about it today, is because of US government funding cuts that we've been talking about in these conversations over the last few months. But even aside from those cuts, family planning is an area we've been working on pretty seriously for the past year and a half. We've been getting up to speed on the evidence, thinking about how we should model the benefits that come from family planning, speaking with people in the sector, and then beginning to conduct some grant investigations.

We thought this was an area that seemed promising a year and a half ago, and it looks even more promising now because the US government and other governments that fund it are talking about pretty large cuts. We think these cuts could be as large as 70% to existing funding for family planning efforts.

You know, [00:01:00] this is a really big area, and we're only gonna scratch the surface in today's conversation. And if this topic is of interest to you, please let us know. We'd love to hear about that, and we can try to share more information.

But the things that we'd like to talk most about today is to give an overview like, what is this area? And so specifically, what are the programs that we might support? You know, what exactly do they do and how do they work? What some of our open questions are, and then why is this area challenging? Why aren't these gaps filled already? Is it hard to know if programs are working well and why?

And so joining me today is Dilhan Perera, he's a researcher who works at GiveWell. Hey Dilhan, before we dive in, tell us a little bit about your background before coming to GiveWell.

Dilhan Perera: Thanks for having me, Elie. So I've been at GiveWell for two years now. My background is in economics and psychology, which together some people refer to as behavioral science, so the science of human behavior and how that can be applied to improve policy and program outcomes.

Before coming to GiveWell, I was a researcher at a behavioral science [00:02:00] consulting firm, trying to evaluate the sort of effects of interventions or changes to programs that would try to apply an insight from human behavior to improve the outcomes of those programs.

Elie Hassenfeld: Cool. Glad you're at GiveWell and glad you're working on this question of family planning. So, you know, at a high level, family planning is provision of contraception to people who want it. But then maybe just to introduce us, what exactly are the programs that donors could support with charitable dollars? What do organizations actually do?

Dilhan Perera: Yeah, so there's a range of programs in the space. Some of them are basically about delivering contraceptives to people who otherwise wouldn't have access to them. So going out to reach communities with people who have trained to provide both family planning counseling, so the information about different methods and why or why not you might want to use them, as well as the products, the contraceptives to give them on the spot for people who are interested in taking them up. So there's those types of service delivery-oriented [00:03:00] programs.

There are also programs that focus more on what's called the demand side for family planning. So providing information about different methods, trying to correct misinformation or inaccurate perceptions that people have. This can be done through one-on-one counseling sessions, but also things like mass media where you're reaching a lot of people at once.

And then there's other things that are trying to improve provision of family planning through public health facilities as well as private facilities like pharmacies. So basically finding ways to diagnose issues in how those existing systems are already functioning and supporting governments or private providers and pharmacy shop owners to improve their ability to provide contraceptives.

Elie Hassenfeld: And of those areas, so one is service delivery, so providing contraception to people who want it, another is trying to correct misconceptions, a third is improving the existing systems. Would you say that at this point any of these seem more likely to you to be the areas that we focus on, do they [00:04:00] all seem about the same? Like how do you think about their relative priority?

Dilhan Perera: Yeah, so I think because this is a new area for us, I've been focusing on the programs that are sort of easier to understand and to evaluate. That tends to be the more straightforward service delivery program. So where you are trying to identify the people or the communities that have limited access, physically going out to them, and then providing services and the contraceptives closer to where they are, which makes it more likely that the people who want to use it will take them up.

Elie Hassenfeld: Maybe you could give an example of how a program like this works, what we know about it, what we don't know about it, what some of the key questions are, and just how you think through analyzing what you described as the more straightforward service delivery programs.

Dilhan Perera: Yeah. So there are a couple of different approaches that programs use. So one is that they will hire their own health workers and pair them up with a driver in a four-wheel drive with a bunch of products in the back. And they will drive out to communities that they've identified, in consultation with government [00:05:00] officials and community health workers as having both demand for contraceptives, but also limited supply. So they'll just physically drive out there with their own staff and their own products and do some information sessions with the communities, ask people who are interested to come to a clinic where they've set up shop, and then deliver counseling and the services.

An alternative method is to find existing community health workers or nurses who live in communities already and are sort of already working within the health system. And providing them with a bit of additional training, as well as particular types of contraceptives that they're qualified to provide. So they tend to be the ones that are simpler to administer the shorter term contraceptives. And basically providing them with financial support, and the products to go out and find people in their communities who might want to use contraceptives, but for whatever reason aren't already using them.

Elie Hassenfeld: And so you mentioned there's these different kinds of products, what kinds of products are people providing, and which are relatively easier and relatively harder to provide?

Dilhan Perera: [00:06:00] There's a big variety of different methods, but a lot of methods are either short acting contraceptives or long acting, reversible contraceptives. So the short acting contraceptives are things like condoms and oral pills, which you have to use every day or every time you have sex, and provide temporary protection. The long acting methods are things like implants that people can get in their arms or intrauterine devices or IUDs. And they require a provider to perform a clinical operation to insert them into a woman's body. And once they're inserted, they can provide protection from pregnancy for several years. So a lot of people find those methods to be more convenient, because they don't require you to take an action every time or every day. But at the same time, you have less control over when you take them out. You need a skilled provider to take them out for you when you want to stop using them.

Elie Hassenfeld: As you've looked into this, do you have a sense of which area seems more likely to be the one that [00:07:00] GiveWell would focus on? There's this trade off, like the short term one, on one hand it's easier to provide, but requires this behavior to use consistently. And then on the other hand, you have the longer lasting, less user behavior driven approach, but then that requires more of a commitment on the part of the person taking it up, and also some level of medical care to be in a position to implant and remove the device. Given those trade-offs, do you have a sense of where you're gonna focus your attention?

Dilhan Perera: Yeah, I think ideally, most programs these days try to provide a full mix of methods, because different people will have different preferences over the different methods. And some of the methods have particular side effects or aren't suitable for people with particular medical conditions.

So, I think programs in an effort to both, you know, maximize the number of people they can serve and ensure that people have some choice and control over matters of their reproduction, try to provide a full range of methods. Some programs are limited in the range of methods they can provide [00:08:00] just because health workers that are supporting the program aren't qualified to provide all methods. So, in those cases, it would be a consideration about whether the program is still enabling people who want to use a different method, a long-acting method to access that through other channels that might already exist, so referring them to a nearby health facility or a program run by a nother organization.

So I think in general, we're not trying to focus on programs that provide a specific method, but there might be case by case reasons why the range of methods provided isn't the full range, and we'll consider them case by case.

Elie Hassenfeld: Let's talk about this question, which is a really critical one, are you encouraging people who want contraception to take it, which would be good, rather than what would be bad like trying to increase the number of people who utilize contraception regardless of whether they want it. And that does seem like a particularly pressing issue with respect to family planning, in a way that doesn't quite exist with many of the other programs we support. [00:09:00] It's more clear that it's good for a child to get preventative malaria medication during the malaria high season to prevent illness and death. But contraception should be used by people who want it at the time that they're taking it.

And that seems like sort of a somewhat novel challenge with respect to family planning that has not come up as commonly in other programs we've supported. And so I'm really curious how you've thought about that question and the work you've done so far.

Dilhan Perera: Yeah, I totally agree with that. And I think there is a history of within family planning and it's early days, there was less concern for people's preferences and rights, and there was sometimes intent to just maximize the number of people using contraception, or in some cases, in some countries, minimize the number of people using contraception to pursue sort of population level objectives.

I think these days, all programs emphasize voluntary use of contraception and contraception as a way of increasing people's reproductive choices. So I think the main question is to what extent programs in practice implement the sort of [00:10:00] commitment to choice and, you know, providing access to people who want to use contraception, but not pressuring people who don't want to use it into using it.

And so it's a fundamentally difficult thing to monitor, because a lot of the potential for people being pressured into using contraception or using a particular method occurs in this sort of private clinic room interaction between a person and the service provider. Programs try to monitor this using a variety of methods, like interviewing people straight after they've left the clinic to ask them about their experience, providing feedback mechanisms for people to say if they've had a bad experience. A lot of programs do supervision where like someone more senior will come in and observe some clinical sessions to make sure that the provider at least knows what to do, even if they don't always do that. And some programs also try to use these things called mystery client sessions where they will send someone from the community to try to obtain services and report back on how the provider interacted with them.

But none of these methods are [00:11:00] perfect, they've all got various limitations, and so we're also considering whether we should fund more research with people in the communities targeted by these programs to understand how they perceive them. So what have their experiences around this been? Just to get a better sense of how big an issue this is and also, we know that there is a sort of ongoing potential for people to not be fully informed about the method that a provider is encouraging them to take up, or struggling to get long acting methods removed.

But we also need to consider that against the issue that a lot of people who want to use contraception just don't have access. And so, you want to make sure we're sort of striking the right balance and making sure we can support access where it's highly cost effective to do so, and where there's demand and benefits for people who want to use it, while trying to minimize potential for people to be coerced into using contraceptives.

Elie Hassenfeld: You know, one thing that makes me think of as you were talking this through, is it really seems like in a certain sense, it's harder to fund family [00:12:00] planning programs cost effectively. Because just in one small example, the effort needed to monitor this kind of program effectively seems higher than the effort that might be needed to monitor a program like vitamin A supplementation, you know, a very infrequent delivery of a health product to a defined population. And I guess that just means that like other things equal, these programs cost a little bit more and makes it a little bit harder to find the programs that are going to be cost effective by our assessment.

We're looking into it because we think, I guess we don't know for sure, but our sort of hypothesis that we're testing is that notwithstanding those additional costs of monitoring, there still will be large gaps that we expect to fill, and taking all costs into account that's still very great, and we're excited to deliver funds at that level.

Dilhan Perera: I think that's exactly right. The family planning contraceptive products themselves are fairly cheap, but the sort of service delivery and monitoring costs on top of that can be quite a bit more [00:13:00] than say, a bed net campaign.

And I think it's not just the financial costs and the sort of due diligence we need to do to understand whether a program is delivering its services responsibly, you know, there is also an opportunity cost there for us.

One other uncertainty I have, and it's a pretty key uncertainty, is to what extent family planning programs can reach people who otherwise wouldn't have had a way of accessing some form of contraception. And I think the programs that have the best chance of reaching those people are the ones that go out to communities that are more remote, or you know, in lower resource settings where there just aren't many other options or health facilities or providers who can provide a range of methods.

So I think those programs, of the ones that I've considered are the ones that seem more promising, but it's still not clear to what extent they can really target communities that have limited access, because often the information about access and use just isn't that granular. And there's just more costs involved [00:14:00] in hiring your own health workers to go out long distances to reach these communities. So there might be a trade off between to what extent you are reaching people who have limited other options, but also the cost of going out there and reaching them.

But despite all that, I think there is enough potential for these programs to be highly cost effective and to produce a lot of benefits for the people who want them that I think it's worth us continuing to consider them, at least for the short to medium term.

Elie Hassenfeld: So in some ways, like a way of framing the problem is the outcome we want to achieve is enabling people who would otherwise use contraception but don't have access to it to utilize it. And then the challenge, is number one, identification. There are very straightforward pieces of information that we have trouble accessing as GiveWell. For example, in a totally different context, it's hard to know how many cases of malaria occur in a certain village in a certain time. So just like basic data is challenging, and so this is a more complicated piece of data, what do people want? Do they have access to it?

Then the second thing is, other things equal, you'd expect that the [00:15:00] places that meet these criteria are more remote, meaning places where people want but maybe don't have access to contraception or have some misconception. But then, two challenges with that. Number one, there's going to be fewer healthcare facilities to refer people to, so it sort of limits your options, and also, the cost of reaching them is higher.

And I think this dynamic that exists here with family planning seems similar to the challenges we face in other programs too with this basic dynamic. It's like the dollars will go further in more resource constrained environments, but it can be hard to identify them. When you're trying to serve a resource constrained environment, there's less support to build a program on top of, so you have to do more of it yourself. And then that plus transportation just increases the cost. And so in some ways it's like weighing up all the factors together and that leads us to then identify the ones that ultimately seem most cost effective. What else would you add to that, or how would you modify that high level take on the challenge?

Dilhan Perera: Yeah, I think [00:16:00] that's exactly right. In terms of programs that directly deliver family planning services and products, that is exactly the sort of tension that we're grappling with. I wanted to also highlight that in some cases it might be purely the supply of the products that are not there and are limiting uptake or use of contraception by people who want to use contraception. In some low income countries and even middle income countries, there are frequent reports of stockouts at public health facilities of contraceptives.

There are doctors and nurses and other health workers at these facilities who can provide these contraceptives and people who come in seeking family planning services, but the products just aren't there. So there's also a type of program which just tries to improve the supply of products at existing facilities. And there is less of a question of whether the person would've gotten the contraception in the absence of the program. The challenge is more how do you get the products to the facilities where people want them. So it's a slightly different approach to the problem, but I [00:17:00] think it's complimentary to the type of program that goes out beyond where these health facilities are to directly deliver additional services.

Elie Hassenfeld: In that case, in that kind of program, where you're trying to address the lack of supply in an existing health facility for a contraceptive product that someone wants. Does that include the long lasting implants, or is it primarily the short term, or both?

Dilhan Perera: It's both. It's the health facility, so the primary and secondary healthcare clinics and even hospitals where a lot of the capacity to provide the longer acting products exist. So it's where the skilled service providers are. But even at these sort of higher level health facilities, often they just don't have the right amount of stock. It's hard for them to predict what the demand for the stock is, and you often hear about people coming in to access the service, but walking away empty handed because the product's just not there.

Elie Hassenfeld: So we're trying to like do this bird's eye view of family planning programs and how we're thinking about it. One of them is, you fund an organization to go out to a fairly [00:18:00] underserved area, give information, provide products. That has these trade-offs around challenges of identification, challenges of fewer pathways to get services that are needed, potentially higher cost.

Then there's this second approach, which is pure provision of supplies via existing healthcare facilities. And I guess I don't know, that one seems extremely promising. So like what's the challenge there? If you know where there are stockouts, then you provide additional supply, and then that has this huge benefit of cost is lower because you're building on the existing healthcare infrastructure that's going underutilized in a sense. Like someone shows up at a facility, and there's a healthcare worker there, they could purchase the product or obtain the product, and then it's just not there. And you know that when the product is not in stock, that is someone who wants it who's not getting it otherwise, so it solves that challenge too.

So I'm curious if you could say more about that one, 'cause that seems so intuitive. What are the key challenges that come up there, and why are you not just like, oh yeah, we should make sure to do this, immediately.

Dilhan Perera: Yeah, I think it does sound very [00:19:00] intuitive and appealing for the reasons you mentioned. I think the main challenge is you're having to work with the supply chain in a country to get the products from say, a central medical warehouse in the country, to the facilities where people are demanding products and where there's not enough product.

One, it can be difficult to know where the demand and the lack of supply is in terms of facility. So I think one of the existing challenges with these supply chains is that products can just end up misallocated across facilities. So some facilities will just have a very big amount of stock that doesn't actually match the demand, whereas a lot of other facilities just don't get enough stock.

And then the actual distribution and the logistics, so the trucks and the people to transport the products, you know, you can always pay for more trucks and more people to do that work. But I think there's this bigger challenge that a lot of health systems face of making sure those systems are efficient and sustainable and not overly dependent on donor funding, which can be [00:20:00] volatile.

And so I think there's an ongoing question about to what extent government should be owning these supply chain systems, and the challenge for donor funded programs is to figure out how to work with government to make those systems more efficient or more effective without setting up parallel systems that might not persist over the longer run.

Elie Hassenfeld: Right. So this one at face value seems very promising. Simply put, it's providing supplies to clinics that exist and have stockouts. But then there are these two challenges. One is sort of narrow but important, which is just it's hard to know, and this is a very consistent problem that we always face across everything we do, where if you could obtain information that seems like fairly basic, then it would make the work of allocating charitable dollars more effectively much easier.

But it is often the case that very basic information is hard to come by because of the environments where we're supporting programs. And so that means that even knowing which facilities tend to have stockouts, it's just not something that is in most places digitally recorded and easily obtainable. And so that [00:21:00] makes it challenging.

The second challenge is, the charitable dollars are really like on top of this existing system. And so then for it to work, you have to enable the supplies to flow through this existing government supply chain more effectively than they were. That's challenging both because you're like engaging with this larger infrastructure, which is the supply chain. And then also for the reason you said, which is there's this question of, how should governments interact with privately funded components of their health systems?

So basically it's like you can potentially do something very targeted that's very cost effective, but it comes with this additional challenge of having to navigate engagement with the government infrastructure, which itself is more complicated and more challenging, presumably, than just directly providing a service.

Dilhan Perera: Yeah, that's exactly right. And one additional note to your second point is that it's not just about what role should the government be playing with the supply chain, so like the distribution and logistics and products. But to what extent should [00:22:00] governments be funding the products themselves out of their own money versus donors funding the products, 'cause I think one problem here is also just that there isn't enough funding between donors and governments to pay for the products, given what people demand.

Elie Hassenfeld: It's kind of interesting to me just to contrast this program we're talking about, different forms of family planning programs, to other GiveWell programs and just to see like some of the challenges.

So something like seasonal malaria chemoprevention, you know, that's preventative malaria medication. The way that's delivered is monthly, door-to-door to all children under five. Number one, it's addressing this challenge of identification, because they're delivered at fairly large scale to areas with high burden of malaria. It's engaging with the government, but not then relying on this other system. And you can see how that program, which we've directed a lot of funding to historically, has a lot of these simplifying characteristics that make it easier. And I don't know, I would say that's probably one reason why we did so much on something like malaria prevention [00:23:00] in the past, and it remains an area that dollars can be spent really cost effectively.

And then as we've grown our team, we're in a position to take something on like family planning, which is much more complicated, but could offer very cost-effective funding needs that we can fill to help people.

Dilhan Perera: I was thinking about family planning programs compared to things like seasonal malaria chemoprevention, bed net campaigns, vitamin A supplementation campaigns. In all of those cases, at least the target population is pretty well defined. So it's kids of a certain age in a certain region, and you're trying to just get the product to as many of those households or kids as possible within the relevant time period.

With family planning, it's complicated because not all people want to use contraception. People's preferences matter a lot. And their preferences can change over their lifetime, but also as the wider sort of society changes.

And secondly, there are multiple ways to access contraception in a lot of contexts, so you can get them at pharmacies, at public health facilities. There are different types of methods that are available at different places, and so it's less [00:24:00] clear to what extent you're making a difference above that baseline provision, relative to something like seasonal malaria chemo prevention, which is just something that people are very unlikely to buy for themselves a shop.

Elie Hassenfeld: And then I guess someone might be wondering like, well if there's all these challenges for family planning relative to say malaria, then why do family planning at all? Why not just do malaria? And just to explain it, we're not trying to do family planning for the sake of it. As donors, we provide more funding for malaria, there are diminishing returns or diminishing opportunities in the malaria space. And so the question is you know, at the current level of funding, do family planning programs surpass the last malaria program that we would fund? And that's the question we're really trying to answer. And right now we think that's really plausible, which is why we're looking into it.

So I just wanted to close with your thoughts on how we're gonna move this ahead. You know, we've talked about a couple of types of programs, we've talked about a lot of the questions we need to answer and a lot of the challenges. And then I imagine, at a very high level, the approach that we tend to take is, there's no way to answer these questions globally for all [00:25:00] programs, they tend to come down to the specific programs, the specific locations, specific organizations, and what information they have can provide what they do, how it works. Can you just say more about where we go from here and how we'll make progress on answering some of these questions so we're in a position to know how cost effective the programs are and then direct support their way.

Dilhan Perera: So we are currently considering a handful of funding opportunities in each of the buckets that we spoke about. So programs that go out into communities to deliver services that otherwise wouldn't be available. Programs that train healthcare workers to provide products to people in their own communities, as well as programs that try to increase the supply of contraceptives at public health facilities. We're actively considering whether to fund those programs, and we think that if we do fund those programs, either through following along with the data that they generate, as well as potentially funding additional data collection to happen, we might learn through those initial grants about some of these open questions we have, and then be in a more informed position in the future to [00:26:00] decide whether to continue funding those programs or expand our funding or contract it.

And one other thing we didn't mention is that evidence generation in this space could also be really helpful. So more things like randomized trials to understand what is the additional effect of these programs on contraceptive access and use, and trying to use those also to monitor these questions of coercion and unwanted contraception. We think that could also be a pretty good use of funds. So we're considering some opportunities to fund more of that type of evidence generation to help us and other funders make better decisions in the future.

Elie Hassenfeld: Got it. And so this is a case where in some ways the questions we need to answer are so concrete about specific programs and opportunities. One of the best ways that we may be able to do that is to direct funds to programs and organizations, learn from that fund direction, potentially also fund research. We just think there's a limit to how far we can get assessing information that already exists without providing some funding. And so we're likely to do that to enable us to learn more.

Dilhan Perera: Yeah, that's exactly right.

Elie Hassenfeld: Anything [00:27:00] else you wanna add about this that we didn't cover that is really important to know before we wrap up?

Dilhan Perera: Yeah, so I think there's two things I'd wanna mention. First, our conversation mostly focused on what I'd call the supply side of family planning or the availability of contraceptive products and services. But I think work on the demand side can be just as important. And this includes things like mass media campaigns all the way down to one-on-one counseling to help people and their families figure out how family planning fits into their lives and their existing values. The demand side work can be particularly tricky because people's values and preferences around having kids are complex and dynamic. But I think without trying to do the sort of demand side work and counseling to figure out whether there's underlying demand for better birth control, there might just be a limit to how much programs can make a difference by just focusing on the products and the services.

And second, I think a lot of the value of family planning programs is tied to how much they actually increase [00:28:00] people's control over their sexual and reproductive lives. And I think the evidence on this is fairly limited. There are a handful of randomized trials, some of which show effects on pregnancy rates, for example, but others which don't. Whether or not programs have these more sort of downstream effects really depends on how well the program is reaching people who have limited options to control their births to begin with.

And so I think over time I'd hope to see more evidence about how specific family planning programs affect not just contraceptive use, but also the more downstream outcomes that contraceptive use is meant to lead to, like a reduction in unintended pregnancies and improvements in different types of wellbeing.

Elie Hassenfeld: Yeah, that's great. Alright, thanks, Dilhan.

Hey everyone, it's Elie again. One of the things that I think is most interesting about this discussion that Dilhan and I had is just how challenging it [00:29:00] is to find cost effective programs to fund in family planning. It actually makes a lot of sense because in GiveWell's history, you know, we really have always tried to start with the areas that we felt like we had the best ability to assess, given our team's capacity and capabilities.

And so, you know, GiveWell started out looking at the most measurable global health and development programs. And then over time slowly expanded further into more complicated areas. With an area like family planning, you can really see how complicated the challenges are. You know, these programs are, in my opinion, far more difficult to identify and deliver in an effective and cost effective way than some of the programs GiveWell has supported in the past.

But as our team has grown and we have increased our capabilities and we're in a position where we can look at new areas. And like I said in the conversation, we're not looking at these newer areas for the sake of expanding our reach, that breadth itself is not the goal, but you know, we wanna find the most cost-effective [00:30:00] ways to help people no matter what programs they're in. We want to be able to evaluate as wide a scope of programs as we can, so we'll make better decisions if we're comparing family planning programs against all the other programs we're considering.

Family planning is new to us. So if it's an area that you're interested in supporting you can do that via donations to our All Grants Fund, which gives us flexibility to respond to any need, even new ones that are relatively riskier to us that we know less about.

Of course you can also always support our Top Charities Fund, those groups will have a lot of need too. We're appreciative of any support, especially during a time when the needs for people in low income countries are as high as they've been in a long time.

Thanks again for listening, for supporting GiveWell, and for caring and helping us do the most good we can together. We really appreciate it.