Why malnutrition treatment is one of our top research priorities

Added October 2022: From 2020 to 2022 we used the name “Maximum Impact Fund” to refer to the fund used to support the highest-priority funding needs among our top charities each quarter. In September 2022, we changed the name of this fund to the “Top Charities Fund” to better describe what opportunities this fund supports; more information here.

We believe malnutrition is a very promising area for charitable funding in the future. In 2021, we directed nearly $30 million to two organizations—The Alliance for International Medical Action (ALIMA) and International Rescue Committee (IRC)—working on malnutrition, and we expect to direct more funding to malnutrition programs in the future. (We have published a write-up about one of these grants here and will publish write-ups about the other grants in the near future.)[1] To give a sense of what we expect, we would not be surprised if GiveWell directs as much funding to malnutrition in the future as we have to malaria programs in recent years. We haven’t written much about this cause, so we thought it was important to remedy that. In this post we will share: • What malnutrition is and the scope of the problem. While we have remaining uncertainties, we estimate that 45 to 210 million children experience malnutrition each year and that malnutrition increases their chance of death by two to seven times relative to children who aren’t malnourished. More below. • How NGOs and governments support treatment. NGOs and governments implement relatively simple programs to treat malnutrition. • Why we believe these programs are promising. In sub-Saharan Africa, which accounts for nearly 25% of global cases and where we have the best understanding, we estimate that the total funding gap for malnutrition programs is between$350 million and $13 billion, at a cost of$2,000 to $18,000 per death averted. • What our remaining open questions are. We have open questions about the evidence, what it means for the likely effect size of the programs we’re supporting, and challenges to support organizations growing significantly to meet the large global need. We’re not recommending that donors give to malnutrition programs at this time. We’ve filled the cost-effective funding gaps that we’ve identified. We’re investigating further spending opportunities and expect to direct additional funding to malnutrition programs in the future, but we don’t have specific recommendations for donors today. Now, we continue to recommend that most donors give to our Maximum Impact Fund. A note on the estimates below As you’ll note, most of our estimates involve very wide ranges. This includes our estimate for the number of children globally who experience severe malnutrition, the cost to reach them, and the cost-effectiveness of programs we might fund. We expect that some programs will be on the more cost-effective end of the range and some on the less cost-effective end, and much of our future work will be focused on finding programs that meet or surpass our current cost-effectiveness bar for funding recommendations. What is malnutrition, and how big is its burden? Globally, we estimate that between 45 million and 210 million[2] children experience malnutrition at some point during a year . Acute malnutrition refers to “wasting” (a significant weight loss resulting in an abnormally low middle-upper arm circumference, or low weight for height ratio) and/or the presence of nutritional edema (swelling caused by excess fluid retention).[3] Malnutrition weakens children’s immune systems, which can make them more susceptible to and more likely to die from illnesses such as diarrhea, pneumonia, and malaria. We estimate that acute malnutrition increases the chance of death by two to seven times compared to non-malnourished children.[4] It also affects children’s growth and development.[5] There are known solutions NGOs and government health systems know how to treat acute malnutrition. Treatment can take place in an in- or out-patient setting, depending on the severity of malnutrition.​​[6] In both settings it generally involves providing ready-to-use therapeutic food (RUTF), a nonperishable, calorie- and micronutrient-dense food designed for treating malnutrition, until children meet criteria for discharge.​​[7] Treatment for the most severely malnourished also includes a standard set of medical treatments to address any underlying complicating illnesses. For example, children with severe malnutrition are also given a course of antibiotics, since they may be more susceptible to infection.[8] Based on our investigations to date, our understanding is that NGOs often support capacity-constrained government health systems to deliver high-quality treatment to children that need it by: • Supporting the identification of malnutrition in the community. Parents often don’t seek treatment for malnutrition—possibly because malnutrition is common in their communities, so they perceive their children as typical. NGOs aim to make it simple for caregivers to identify malnutrition episodes by providing them with color-coded bands they can use to check their children’s upper-arm size and, if it’s in the red or yellow zone of the band, to seek medical treatment.[9] • Providing training, mentoring, and supervision of health workers, or hiring additional staff when capacity is constrained. Health facilities can be understaffed, and existing staff often lack training on the latest protocols to effectively deliver malnutrition treatment. Hiring additional staff ensures there is enough capacity to meet demand (especially during “peak” malnutrition periods).[10] Ongoing mentoring ensures workers stay supported and motivated to deliver high-quality services. • Bringing treatment to difficult-to-reach communities through strategies like mobile treatment vans or covering caretaker transportation costs. • Ensuring that RUTF and other needed medical supplies are in stock. Depending on the context, this can include technical assistance to better predict demand, assistance with procuring supplies, and/or logistical support to get supplies to their destination.[11] Why we think malnutrition programs are promising: cost-effective with large unmet funding needs Malnutrition has a huge global funding gap. In sub-Saharan Africa alone (where we have the best understanding), we estimate that 10 million children are experiencing acute malnutrition (wasting) at any given time.[12] That implies that somewhere between 10 million and 46 million[13] children experience acute malnutrition each year, with 7 million to 38 million[14] of these children being untreated. It costs between$35 and $805 to treat an episode of acute malnutrition.[15] This wide range indicates the substantial variation in program intensity and context. Some cost differences can be explained by illness severity (severe acute malnutrition is more expensive to treat than moderate acute malnutrition), but other factors like the density of cases and government capacity can matter just as much: on one hand, due to high fixed costs, the cost per treatment is driven by the number of children who need treatment in the targeted area (the higher the number of children who need treatment, the lower the cost per child); on the other hand, the strength of government services determines how many additional resources need to be mobilized to provide malnutrition treatment (as strong health systems have more existing capacity to treat malnutrition).[16] When we crudely model these inputs, it implies that it would cost between$350 million and $13 billion[17] to treat all otherwise untreated episodes of malnutrition in sub-Saharan Africa. If all of these children were treated, it could avert between 200,000 and 800,000 deaths annually, for a cost of about$2,000 to $18,000 per death averted.[18] This is roughly 2 to 17 times more effective than GiveDirectly’s cash transfer program, and places malnutrition treatment within the range of cost-effectiveness of top charities we have directed significant amounts of funding to.[19] What are our major open questions? Malnutrition is a new area of research so there are many uncertainties relative to recommendations we’ve made and refined over a longer period of time (e.g., malaria). In particular we would like to better understand: • The mortality rate for children who have acute malnutrition but do not receive treatment. Our estimates of mortality rates rely on historical observational data and on an analysis we have not vetted in detail.[20] Moreover, while we discount the initial estimates to account for limitations of the evidence and differences in contexts between historical studies and the programs we fund, we are very uncertain about the size of those adjustments. We are re-analyzing the data we currently use and hope this will help us refine our approach. • The extent to which NGOs increase treatment. In many areas, malnutrition treatment is available through government-supported facilities. We believe that NGO support causes more children to be treated than would have been by the government alone, but we are unsure about the size of the effect. Our grants support coverage surveys that will help us make progress on this question. Beyond these key questions about the program’s benefit, a major challenge will be finding places that we’re confident in to fund. Two of the largest uncertainties we have on this front are: • Organizational capacity. The groups we’ve spoken with that deliver malnutrition programs have been subject to erratic funding streams that are often well below the needs of affected populations. Because of that, they have tended to scale programs up and down quickly in response to funding flows rather than invest in developing a long-term, stable foundation in any given place. • Identifying the most promising opportunities. Not every location is cost-effective, so it’s a journey working with grantee partners to find the places that thread the needle of lower costs, higher burden, etc. Although our rough calculation of the global funding need is very high, the portion of that gap that would be cost-effective enough to fund with GiveWell grants remains an open question. Based on our investigation so far, we would guess there are many cost-effective opportunities, though we remain uncertain about the precise number and size of such opportunities. Conclusion Overall, we see malnutrition as a very promising area for funding and further research. It potentially offers$1 billion or more in funding opportunities at cost-effectiveness levels that are consistent with our top charities.

We have many open questions and a long road to go to answer them. We are currently investing significant energy into addressing our uncertainties, and we look forward to sharing more updates about our progress.

References

Footnotes for this post may be found here.

• Hi, when you say treating an episode of malnutrition costs $35-$805, but averting a death from malnutrition costs $1,600 to$18,000, are you making an assumption about how many times a single individual would need to be treated throughout their lifetime or only an assumption about the mortality rate of malnutrition? If you avert an episode by providing RUTF until the patient is discharged, what is the likelihood they return in a few months, a year, 5 years? At essence is the question: if the root causes of malnutrition (lack of arable land, drought/severe weather, government corruption, inefficient farming practices, etc.) are not addressed, are we just throwing money down a never-ending drain in trying to treat the episodes?

• Oscar Delaney on November 22, 2021 at 2:52 am said:

Like Paul, I am wondering about the role of advocacy or other up-stream, higher-leverage interventions in malnutrition. For malaria it makes sense that among the best ways to reduce the malaria burden is to prevent malaria with ITNs or treat it with SMC. However for malnutrition, my intuition would be addressing underlying causes could be the more important long-term approach. This is perhaps particularly suitable given you are already looking to expand to health-policy work beyond just more quantifiable interventions, I believe.

• Lucas Lewit-Mendes on November 22, 2021 at 8:42 pm said:

Hi,
Thanks for the great write-up. Would it be possible to make the calculation links (in the footnotes) available for viewing?
Cheers,
Lucas

• Maggie (GiveWell) on November 23, 2021 at 6:20 pm said:

Hi Lucas,

Thanks for catching this! We have adjusted the permissions and you should be able to see those spreadsheets now. Please let us know if you run into any trouble.

• Lucas Lewit-Mendes on November 24, 2021 at 1:30 am said:

Thanks Maggie!
Re “for a cost of $1,600 to$18,000 per death averted” – as far as I can tell, the link in footnote 18 implies $1,900 to$18,000 for SAM/MAM. Is that a typo or am I missing something?
Cheers,
Lucas

• Maggie (GiveWell) on November 29, 2021 at 3:25 pm said:

Many thanks for picking this up, Lucas! That is indeed a mistake on our end. We have corrected the post above to reflect the cost per death averted in the spreadsheet (rounding up the lower estimate to \$2,000). We also noticed another error in the post and have changed the upper estimate of the mortality burden from 1,000,000 to 800,000 to match our spreadsheet calculations.

• Lucas Lewit-Mendes on December 1, 2021 at 3:14 am said:

Thanks Maggie!

• Maggie (GiveWell) on December 1, 2021 at 9:33 am said:

Hi Paul and Oscar,

Thanks for your questions. The cost of averting a death is calculated based on the mortality rate of malnutrition. We do account for the risk of relapse (and other factors that might lead us to overestimate the program’s effect) in our adjustment, but we are very uncertain about the size of this adjustment. For more details, see row 47 of this spreadsheet Rate of relapse is an open question that we hope to learn more about as we investigate this area further.

As for addressing the root causes of malnutrition, there’s a lot we still don’t know. We haven’t looked into food provision programs yet, and it’s possible those could be very cost-effective and worthy of funding. However, we think that malnutrition treatment appears to be a very cost-effective program for saving lives. We think it’s important to do work on this now while we figure out whether there are additional effective pathways for us to recommend and direct funding toward.

• OFHSoupKitchen on February 10, 2022 at 4:43 am said:

Malnutrition globally is really one of the problems that need attention. Hope the government will also focus on providing funds or programs to treat this.