Alliance for International Medical Action (ALIMA) — Malnutrition Treatment in Chad (May 2021) and in Niger and Nigeria (November 2022)

Note: This page summarizes the rationale behind two GiveWell grants to ALIMA. This page discusses our rationale at the time we recommended the grants. ALIMA staff reviewed this page prior to publication.

Published: March 2024

Summary

What we thought these grants would do

With these grants, we expected ALIMA to support government-run health facilities to increase the number of children who receive treatment for malnutrition and to improve the effectiveness of that treatment. This would be done through activities such as providing staff and staff training and making infrastructure improvements. In all locations except N'Djamena, we also expected ALIMA to support a broader set of pediatric care interventions. (More)

Why we recommended these grants

• We thought that ALIMA's program in these locations was highly cost-effective.

  In simple terms, we thought these grants would be cost-effective because:

  • Children with moderate acute malnutrition (MAM) and severe acute malnutrition (SAM) are at greatly elevated risk of dying compared to non-malnourished children. Depending on location, we estimated that the mortality rates of children with untreated malnutrition were 3% to 19%. Our estimates were based on Institutes for Health Metrics and Evaluation and Demographic and Health Surveys Program estimates of under-5 mortality rates, local malnutrition prevalence surveys, and a pooled analysis of historical data that estimates the risk of excess mortality associated with untreated MAM and SAM. (More)
  • Children who are malnourished are significantly less likely to die if they receive treatment. We estimated that MAM treatment reduces the relative risk of mortality over the next year by about 40% and SAM treatment reduces it by about 70%. This is based on the pooled analysis of historical data referenced in the previous bullet point. Together with our estimate of the mortality rates of malnourished children, this implied that malnutrition treatment reduces mortality by 1 to 14 percentage points, depending on the program location and the severity of malnutrition. (More)
  • ALIMA's support for malnutrition treatment programs in government-run health facilities leads to more children being treated for malnutrition. We estimated that ALIMA's program would increase the number of children who receive malnutrition treatment by ~60,000 over three years in Chad, and ~190,000 over three years in Niger and Nigeria. This is based on government data provided by ALIMA and historical caseloads (in some locations) and population and malnutrition incidence estimates (in others), and ALIMA’s best guess of the percentage of children that would be treated without their support. (More)
  • ALIMA's support for other pediatric care leads to significant benefits. We expected most benefits to come from vaccinations. We estimated that this work made up 7% to 30% of the program's total benefits, depending on location, in Dakoro and Mirriah, Kaita, and Ngouri. (More)

• We expected these grants would provide information that would affect future grants. In particular, we expected ALIMA's monitoring data from Ngouri to improve our estimate of the number of additional children treated as a result of ALIMA’s work. This is because there were no NGOs supporting malnutrition treatment programs in Ngouri at the time this grant started. As a result, we expected to learn about baseline coverage in an area not receiving NGO support, and how coverage changes with ALIMA's support.
  We expected ALIMA mortality data from Dakoro, Mirriah, and Kaita to improve our estimate of mortality rates for malnourished children, one of the major drivers of the grants’ cost-effectiveness. (More)

Our main reservations

• What’s the risk of death among malnourished children and what effect does malnutrition treatment have on mortality? There exists very limited evidence on mortality rates among malnourished children and the effect of community-based management of acute malnutrition (CMAM) on mortality. For the 2021 grant, we heavily relied on Olofin et al. 2013, a widely-used meta-analysis of observational studies conducted in the 1980s and 1990s that estimates the ratio between mortality rates of malnourished and non-malnourished children. Since we did not feel we had a solid understanding of the methodology employed in the paper, we hired an external consultant to reanalyze the meta-analysis.
  
  Before recommending the 2022 grant, we significantly updated our cost-effectiveness model for malnutrition treatment programs (more). In particular, we:
  • Incorporated the result of the reanalysis of Olofin et al. 2013.
  • Estimated a maximum plausible mortality rate of malnourished children, based on malnutrition incidence and overall mortality rates in the targeted areas. We applied this estimate as a ceiling to our best guess.

  We had substantial uncertainty about our updated estimates. In some contexts, the ceiling mortality rates were substantially lower than those we estimated in our main cost-effectiveness analysis. We were unsure what explains this discrepancy. We were also uncertain about our estimate of the effect malnutrition treatment has on malnutrition. In particular, that’s because we assumed that the effect of malnutrition treatment on mortality was captured by the correlation between weight-for-height Z score (WHZ), a measure of nutritional status, and mortality. This might not be true for a number of reasons. Moreover, ALIMA identifies malnourished children by using mid-upper-arm circumference (MUAC) and/or edema, rather than WHZ. Our understanding, based on the reanalysis of Olofin et al. 2013, is that using MUAC implies similar mortality ratios as using WHZ. However, ALIMA believes this might lead us to underestimate the mortality risk of children treated in their programs. (More)

  • How many additional children would receive malnutrition treatment as a result of ALIMA's support? We estimate this based on our estimates of the number of children who would be treated with ALIMA's support and without ALIMA's support.
    • We were uncertain about the number of children who would be treated with ALIMA's support. In some cases, our estimates relied on historical caseload data, which might not be representative of future caseload due to fluctuations in malnutrition prevalence. In other cases, our estimates were based on a formula using malnutrition incidence, for which we relied on ALIMA's best guesses.
    • Our estimate of the number of children who would be treated without ALIMA's support were also based on ALIMA's best guesses. We were uncertain about ALIMA’s estimates, since we did not triangulate them and we did not know how well-calibrated they might be. (More)
  • How large would the benefits of pediatric care be? The benefits from vaccinations, malaria treatment, and other pediatric care made up a large percentage of the benefits we model in Ngouri and Dakoro and Mirriah (20% to 30%). However, we only investigated pediatric benefits at a shallow level. (More)
  • How much would we be able to learn from these grants? We were unsure of the extent to which we could extrapolate from program data to make predictions about future years and additional geographies. (More)

  After we recommended the 2021 grant in Ngouri and N'Djamena, we significantly updated our cost-effectiveness model for malnutrition treatment programs. Using our updated model and the information we had at the time of recommending the grant, ALIMA's program in Ngouri and N'Djamena still met the cost-effectiveness threshold we used at the time of recommending the grants (eight times as cost-effective as unconditional cash transfers, the benchmark we use for cost-effectiveness). (More)

  The 2021 grant was funded by Open Philanthropy and individual donors. The 2022 grant was funded through the GiveWell All Grants Fund.

  The organization

  ALIMA supports inpatient and outpatient emergency care and long-term clinical care in 13 countries, mainly in the Sahel region of Africa.1 In January 2021, GiveWell recommended a grant to ALIMA to support its program providing treatment for malnutrition and pediatric emergencies in Mirriah, Niger.

  The intervention

  ALIMA supports the treatment of acute malnutrition in government facilities.2 Acute malnutrition refers to excessive thinness for one's height, low middle upper arm circumference, and/or the presence of nutritional edema, which is swelling caused by excess fluid retention in tissues.3 Acute malnutrition is believed to raise the risks of developmental delays and death from infectious disease.4

  Malnutrition treatment is provided by identifying children experiencing malnutrition and providing them with therapeutic food in an inpatient or outpatient setting. Children experiencing severe acute malnutrition (SAM) are also given a course of antibiotics to reduce infections.5

  ALIMA aims to increase the coverage and quality of malnutrition treatment by supporting the governments in the countries where it works. It does this by training and mentoring government clinic staff, providing medical supplies, refurbishing facilities, providing additional staff, and providing incentives for staff.6 ALIMA also trains caregivers to recognize cases of malnutrition through measurements of mid-upper-arm circumference (MUAC).7

  In all locations we supported except N'Djamena, ALIMA also supports the broader pediatric care provided in the government-run health facilities it works in through a similar set of activities. This work includes providing routine childhood vaccinations (except in Ngouri), diagnosing and treating malaria, treating diarrhea, and other pediatric care.8

  For more detail on malnutrition treatment, see our intervention report.

  The grants

  Ngouri and N'Djamena, Chad (2021 grant)

  Grant activities

  This three-year grant (July 2021 to June 2024) supported malnutrition treatment for children with MAM or SAM, including inpatient treatment for children with complications. The grant also supported inpatient and outpatient treatment for pediatric emergencies in Ngouri.9 Additionally, the grant included funding for learning activities, including annual coverage surveys and a database for tracking individual patients from admission through recovery.10

  Budget

  The grant budget was $7,968,073 over three years.11 We expected approximately $6.1 million to support activities in Ngouri and approximately $1.9 million to support activities in N'Djamena.12

  Dakoro and Mirriah, Niger, and Kaita, Nigeria (2022 grant)

  Grant activities

  This grant supported malnutrition treatment for children with MAM or SAM in Dakoro and Mirriah, Niger and Kaita, Nigeria.13 It provided three years of support for each location between January 2023 and December 2026.

  With this grant, we also expected ALIMA to work to increase vaccination coverage with activities including supporting health workers in checking vaccination status of children, providing logistical support for vaccine transport, and paying incentives to staff who were spending more time in the health center to deliver vaccinations.14 We also expected ALIMA to work to improve pediatric care, such as routine immunizations and testing and treating for malaria.

  This grant included support for learning activities, including annual coverage surveys and mortality surveys.15

  Budget

  The grant budget was $13,455,542.16 We expected approximately $7.9 million to support activities in Dakoro and Mirriah, and approximately $5.6 million to support activities in Kaita.17

  The case for the grants

  Cost-effectiveness

  At the time of recommending the 2021 grant to support ALIMA's program in Chad, we estimated that the cost-effectiveness ranged from 14 to 20 times as cost-effective as unconditional cash transfers ("14x to 20x cash"), the benchmark we use for cost-effectiveness.18

  We made significant updates to our cost-effectiveness model for malnutrition treatment programs between 2021 and 2022. Using the new model structure and the inputs we had at the time of recommending the grant, our updated estimate of the cost-effectiveness of the 2021 Chad grant ranged from 9x to 18x cash. Our cost-effectiveness threshold for funding grants in 2021 was 8x cash, so these updates did not change our assessment that ALIMA was above our funding bar at the time of the recommendation.

  Below, we describe the key inputs into our cost-effectiveness model for the 2021 and 2022 grants. We only reference the revised, 2022 model below. We do not plan to publish the model we used at the time we recommended the 2021 grant, as it is out of date.

  At the time of recommending the 2022 grant to Niger and Nigeria, we estimated that the cost-effectiveness ranged from 11x to 18x cash. Our cost-effectiveness threshold for funding grants was 10x cash.

  See our full CEA for ALIMA's programs in Dakoro and Mirriah, Niger and Kaita, Nigeria here, and our CEA for ALIMA's programs in Ngouri and N'Djamena, Chad here.

  Key inputs into our cost-effectiveness model

  Mortality rates among targeted children

  We estimated that the mortality rates of children with untreated malnutrition were 3% to 19%, depending on location.19

  We calculated these mortality rates in two steps:
  • First, we formulated an initial estimate based on:
    • Mortality ratios that our consultant David Roodman calculated on the basis of observational data from the 1980s and 1990s.20 In these locations, we estimated that children with untreated MAM are ~2.5 times more likely to die than children without malnutrition over a one-year period following measurement of their weight and height.21 We estimated that children with untreated SAM are ~7 times more likely to die.22 We describe the method we use for calculating these ratios in more detail in our malnutrition intervention report.
    • The local prevalence of untreated SAM and MAM. In the locations we support, we estimated this to be 1% to 2% for SAM and 5% to 14% for MAM. We estimated this on the basis of the most recent available prevalence surveys.23
    • Local population-wide mortality rates for 6-59 month-old children. In the locations we support, we estimated these to be 4% to 5%, based on data from the Institutes for Health Metrics and Evaluation and the Demographic and Health Surveys (DHS) Program.
  • Then, we developed a “ceiling analysis” to check whether this initial estimate passed a plausibility check. If it didn't, we applied a discount to account for that. The intuition behind the plausibility check was that there is a constrained relationship between total under-5 mortality rates, malnutrition mortality rates, and malnutrition incidence that is not reflected in our initial estimate. This meant that our model could return malnutrition mortality rates that are unrealistically high, given the amount of malnutrition in a population and its overall mortality rate. Our ceiling analysis accounted for this constrained relationship. This analysis implied that our initial estimate of mortality rates was indeed unrealistically high in the locations targeted by these grants. To address this, we discounted the initial mortality rates by approximately 40% to 70%, depending on location.24 We describe the method we use for calculating this ceiling in more detail in our malnutrition intervention report.

  Malnutrition treatment effect on mortality

  Our estimate of the impact of malnutrition treatment on mortality in our model relied on the pooled analysis of observational studies from the 1980s and 1990s referenced above. We used that analysis to generate a mortality ratio that represents the comparison of average WHZ before and after malnutrition treatment.25 WHZ inputs come from a literature review of the impact of malnutrition treatment programs on WHZ.26 We estimated that MAM treatment reduces the relative risk of mortality over the next year by about 40% and SAM treatment reduces it by about 70%.27 Together with our estimate of the mortality rates of malnourished children, this implied that malnutrition treatment reduces mortality by 1 to 14 percentage points, depending on the program location and the severity of malnutrition.28 We share more detail in our malnutrition intervention report.

  Additional number of children treated as a result of ALIMA's work

  We estimated that ALIMA's program will increase the number of children who receive malnutrition treatment by ~60,000 over three years in Chad and ~190,000 over three years in Niger and Nigeria.29

  We estimated this on the basis of our:

  • Estimate of the number of children treated with ALIMA's support:
    • For Ngouri, N'Djamena, and SAM cases in Kaita, we estimated ALIMA's future caseload based on government data provided by ALIMA (for Chad) and ALIMA's historical caseload (for Kaita).30
    • For Dakoro and Mirriah and MAM cases in Kaita, we lacked reliable historical data. Instead, we relied on population and malnutrition incidence estimates to estimate future caseload. Our estimates of malnutrition incidence are based on ALIMA’s best guesses, which we have not triangulated.31
  • Estimate of the number of children treated without ALIMA's support:
    • We relied on ALIMA's estimates, which we have not triangulated.

  Leveraging malnutrition work to support additional pediatric care

  In Ngouri, Dakoro, and Mirriah, pediatric care made up 20% to 30% of the total benefits we estimated.32 We estimated that pediatric care made up 7% of total benefits in Kaita.33 ALIMA does not support pediatric care in N'Djamena.

  Our estimate of the benefits in Ngouri was based on a subjective guess that additional pediatric care made up 20% of total program benefits.34

  In Dakoro, Mirriah, and Kaita, we modeled benefits coming from increased vaccinations and malaria treatment, and estimated the majority of benefits to come from the former. Our estimate of the benefits from vaccinations was based on our previous research on vaccination benefits and our estimate of ALIMA's effect on the coverage and timing of vaccination. Our estimate of the coverage effect was based on two surveys commissioned by ALIMA that measured vaccination rates in areas supported by an ALIMA program and areas not supported by an ALIMA program.35 On the timing of vaccination, our best guess is that ALIMA causes vaccinations to happen earlier than they otherwise would.36 However, we were not aware of any evidence of the effect that earlier vaccinations have on mortality. We estimated ALIMA's impact based on a subjective guess.

  Learning value

  We expected these grants would help us learn information that would affect future grants.

  In particular, we expected to learn about additional children receiving malnutrition treatment as a result of ALIMA's support, which was one of our key uncertainties about this grant. We expected to learn about the number of children treated with ALIMA on the basis of ALIMA program data. We expected the Ngouri coverage surveys to be particularly useful in informing our understanding because NGOs were not supporting malnutrition treatment programs there at the time this grant started.37 As a result, we expected the baseline survey to inform our understanding of baseline coverage in an area that is not currently receiving NGO support, and subsequent coverage surveys to improve our understanding of how coverage changes with ALIMA's support.

  We also expected to improve our estimates of the mortality rates for malnourished children, one of the major drivers of the grants' cost-effectiveness. ALIMA plans to collect mortality data for children who are 6-59 months old in Ngouri, N’Djamena, Dakoro, Mirriah, and Kaita.

  Risks and reservations

  We use our simple cost-effectiveness model of the program in Dakoro and Mirriah (above) as an illustrative example to report model sensitivities in this section.

  What’s the risk of death among malnourished children and what effect does malnutrition treatment have on mortality?

  We describe how we calculated the risk of death among malnourished children and the effect that malnutrition treatment has on mortality above (here and here).

  Risk of death among malnourished children

  We calculated the mortality rates of children with untreated malnutrition in two steps, described in more detail above. First, we formulated an initial estimate, based on the observational studies conducted in the 1980s and 1990s. Then, we developed a ceiling analysis to estimate the maximum plausible mortality rates for malnourished children.38

  Our ceiling analysis implied that our initial estimate substantially overestimated mortality rates. We added an adjustment to our model account for this discrepancy. However, we remain unsure what explains this discrepancy.

  For ALIMA's malnutrition treatment program in Dakoro and Mirriah, Niger, our 25th to 75th percentile confidence interval for mortality rates of untreated malnourished children is 3% to 9%, which implies a cost-effectiveness of 6x to 17x cash.

  Effect of malnutrition treatment on mortality

  We do not have any direct evidence of the effect that malnutrition treatment has on mortality. This is different from most other child health programs we support, where we have direct estimates of the programs' effectiveness, usually from a number of randomized controlled studies.

  To estimate the effect of malnutrition treatment, we used the observational data mentioned above to estimate mortality rates associated with average WHZ before and after malnutrition treatment, and we then calculated the ratio between the two. We used that ratio to estimate mortality rates in treated and untreated children today.

  We have a number of uncertainties about this method. Importantly, it assumes that the effect of malnutrition treatment on mortality is captured by the correlation between WHZ and mortality. This might not be true for a number of reasons. For example, the mortality risk associated with malnutrition is partially a result of confounding by socioeconomic conditions. While we adjusted our estimate for this concern, we were unsure about the size of the adjustment we should make.

  Moreover, ALIMA identifies malnourished children by using mid-upper-arm circumference (MUAC) and/or edema, rather than WHZ. Our understanding, based on the reanalysis of Olofin et al. 2013, is that using MUAC implies similar mortality ratios as using WHZ.39 However, ALIMA believes this might lead us to underestimate the mortality risk of children treated in their programs.40

  For ALIMA's malnutrition treatment program in Dakoro and Mirriah, Niger, our 25th to 75th percentile confidence interval for the effect of malnutrition treatment on mortality is a 25% to 70% reduction in mortality, which implies a cost-effectiveness of 7x to 16x.

  We describe these uncertainties in more detail in our malnutrition intervention report.

  How many additional children would receive malnutrition treatment as a result of ALIMA's support?

  We describe how we calculated the number of additional children that receive malnutrition treatment as a result of ALIMA's support above.
  Our estimate is based on calculating the number of children treated with ALIMA's support and the number of children treated without ALIMA's support. We then subtracted the latter from the former.

  However, we had the following uncertainties about the number of children treated with ALIMA's support:

  • For estimates of children relying on historical caseload (Ngouri, N'Djamena, and SAM cases in Kaita), the main limitation was that malnutrition prevalence and population can change significantly across years, so we were unsure of whether historical caseload could reliably predict future caseload.
  • For estimates relying on population, and malnutrition incidence (Dakoro, Mirriah and MAM cases in Kaita), we were especially uncertain about malnutrition incidence. Our estimates heavily rely on ALIMA's best guesses based on its experience operating in these locations, which we have not triangulated against other data sources. We were also concerned that past observations might not be a reliable source for estimating future caseload, since malnutrition incidence can change significantly year to year.

  We were also uncertain about the number of children treated without ALIMA's support:

  • We estimated this on the basis of ALIMA staff best guesses, which we had not triangulated.

  For ALIMA's malnutrition treatment program in Dakoro and Mirriah, Niger, our 25th to 75th percentile range for the number of additional children who receive malnutrition treatment with ALIMA support is 150,000 to 250,000, which implies a cost-effectiveness of 8x to 13x.

  We expected to learn more about this during the grant.

  How large are the benefits of pediatric care?

  The benefits from vaccinations, malaria treatment, and other pediatric care made up a large percentage of the benefits we modeled in Ngouri and Niger (20% to 30%). We describe how we estimated this above.

  • For Ngouri, we made a subjective guess that additional pediatric care made up 20% of total benefits. We were very unsure of the size of this adjustment.41
  • In Niger, vaccinations made up 26% of the overall benefits we modeled. However, we had only investigated this at a shallow level. In particular:
    • Our estimate of the effect ALIMA has on vaccination coverage was based on two surveys commissioned by ALIMA that measured vaccination rates in areas supported by an ALIMA program and areas not supported by an ALIMA program. We were unsure of the extent to which these areas are comparable.42
    • Our estimate of the extent to which ALIMA makes vaccinations happen earlier than they otherwise would was based on a subjective guess about which we were highly uncertain.43

  How much would we be able to learn from these grants?

  We were unsure about the extent to which the number of children treated with ALIMA support during the grant would be a good proxy for the number of children treated with ALIMA support in future years. This is because the number of children treated is affected by malnutrition prevalence, and, based on conversations with experts, our understanding was that malnutrition prevalence can be highly variable along with economic, security, and climate conditions.

  We were also unsure whether changes in coverage resulting from this grant would be a good indication of the effect NGOs would have on coverage in other geographies and in future years. This is because non-programmatic factors (for example, security challenges) could make it more difficult to assess the impact of programmatic factors on coverage.

  Updates to our cost-effectiveness model between 2021 and 2022 grants

  The cost-effectiveness model we used to recommend the 2021 grant to support ALIMA's work in Chad was different from the model we describe above. In particular:

  • It relied on mortality ratios estimated in Olofin et al. 2013, rather than the reanalysis run by David Roodman.
  • It did not include a ceiling analysis to estimate the maximum plausible mortality rates of malnourished children.

  Using our updated model (along with the information we had at the time of recommending the grant), our estimates of ALIMA's cost-effectiveness in Ngouri and N'Djamena decreased. However, our estimates still exceeded the cost-effectiveness threshold we used at the time of recommending the grants (8x cash).

  	2022 estimate	2021 estimate44
  Ngouri	9.0x cash	14.2x cash
  N'Djamena	18.4x cash	20.4x cash

  Plans for follow-up with ALIMA

  We planned to have periodic conversations with ALIMA throughout this grant and to receive information on coverage survey implementation and results.

  Forecasts

  We made the following predictions for the grant in Chad in 2021:

  Confidence	Prediction	By time
  85%	ALIMA will obtain permission to implement the OptiMA protocol for malnutrition treatment45 in Ngouri and N'Djamena	End of October 2021
  80%	Baseline coverage surveys will take place in both Ngouri and N'Djamena	End of September 2021
  70%	"Post Year 1" coverage surveys will take place in both Ngouri and N'Djamena	End of September 2022
  60%	"Post Year 2" coverage surveys will take place in both Ngouri and N'Djamena	End of September 2023
  50%	Year 2 coverage surveys will show an increase in coverage compared to Y1 for both MAM and SAM in N'Djamena	End of December 2023
  55%	Year 2 coverage surveys will show an increase in coverage compared to Y1 for both MAM and SAM in Ngouri	End of December 2023

  In October 2021, ALIMA obtained permission to implement OptiMA in Ngouri and N'Djamena.46 Baseline coverage surveys took place from November 2021 to January 2022.47 "Post Year 1" coverage surveys were completed in Ngouri in January/February 2023 and in N'Djamena in March/April 2023.48

  We made the following predictions for the grant in Niger and Nigeria in 2022:

  Confidence	Prediction	By time
  80%	ALIMA will obtain permission to implement OptiMA in Niger	End of January 2024
  80%	ALIMA will obtain permission to implement OptiMA in Nigeria	End of July 2023

  ALIMA received permission to implement OptiMA in Katsina state, Nigeria, where Kaita is located, in May 2023.49 ALIMA obtained permission to implement OptiMA in Mirriah, Niger, in April 2023 and in Dakoro, Niger, in December 2023.50

  Questions for further investigation

  We list our key questions for further investigation in our malnutrition intervention report.

  Our process

  GiveWell recommended the 2021 Chad grant as part of a broad investigation into treatment of malnutrition as a potentially cost-effective intervention. Prior to recommending this grant, we:

  • Had more than 30 conversations related to malnutrition, including five conversations with ALIMA and five conversations with other funders of ALIMA.
  • Built a cost-effectiveness model, which was peer-reviewed by several GiveWell staff members.
  • Reviewed ALIMA's coverage surveys.

  Between the 2021 Chad grant and the 2022 Niger/Nigeria grant, we conducted significant additional research to inform our cost-effectiveness model for malnutrition treatment. That work is described in our malnutrition intervention report.

  In addition to updating our cost-effectiveness model, prior to recommending the 2022 Niger and Nigeria grant, we:

  • Requested information about the budget and activities that ALIMA would conduct with grant funding in Niger and Nigeria.
  • Received information about expected caseload from ALIMA.
  • Had several conversations with ALIMA about this grant.
  • ALIMA also reviewed our updated malnutrition cost-effectiveness model and provided input.

  Sources

  Document	Source
  ALIMA, "What we do"	Source (archive)
  ALIMA, "Where we work"	Source (archive)
  ALIMA, "Who we are"	Source (archive)
  ALIMA, 2023-2025 Niger budget proposal (including 10% of RUTF costs)	Source
  ALIMA, 2023-2025 Nigeria budget proposal (excluding RUTF costs)	Source
  ALIMA, Chad 2021 final budget	Source
  ALIMA, Niger vaccination survey, 2016	Source
  ALIMA, Notes for GiveWell call, 17 May 2023	Source
  ALIMA, OptiMA Chad Projections	Source
  ALIMA, Summary Coverage Surveys - Ngouri, 2023	Source
  David Roodman, On the association between anthropometry and mortality in children, 2022	Source
  Frison, Checchi, and Kerac 2015	Source (archive)
  GiveWell and ALIMA compiled data, 2021	Source
  GiveWell, Ceiling analysis for untreated malnourished children mortality ratios (ALIMA Chad 2021 with updated model)	Source
  GiveWell, Ceiling analysis for untreated malnourished children mortality ratios (ALIMA combined protocol, Niger and Nigeria)	Source
  GiveWell, Impact of CMAM on WHZ	Source
  GiveWell, Malnutrition treatment CEA (ALIMA Chad 2021 with updated model)	Source
  GiveWell, Malnutrition treatment CEA (combined protocol ALIMA, Niger and Nigeria)	Source
  GiveWell, Requested reporting from ALIMA's nutrition programs in Chad	Source
  GiveWell's non-verbatim summary of a conversation with ALIMA, June 9, 2020 (unpublished)	Source
  GiveWell's non-verbatim summary of a conversation with ALIMA, March 5th, 2021	Source
  GiveWell/ALIMA compiled data	Source
  HealthDirect, "Fluid retention"	Source (archive)
  Kevin Phelan, email to GiveWell, May 11, 2023 (unpublished)	Source
  Kevin Phelan, email to GiveWell, October 11, 2022	Source
  Kevin Phelan, email to GiveWell, October 14, 2021	Source
  Olofin et al. 2013	Source
  Phelan 2019	Source
  Susan Shepherd, email to GiveWell, November 2, 2022	Source
  Susan Shepherd, email to GiveWell, October 17, 2022	Source
  UNICEF, WHO, World Bank, "Joint child malnutrition estimates — levels and trends," 2020	Source
  WHO, "Guideline: updates on the management of severe acute malnutrition in infants and children," 2013	Source
  WHO, "Supplementary foods for the management of moderate acute malnutrition in children aged 6–59 months," 2019	Source

  • 1
    • "ALIMA provides emergency medical care in conflict zones and difficult-to-reach areas that may be suffering from high mortality rates. We are also implementing longer-term clinical and operational research projects to develop new treatments and offer innovative solutions in order to transform humanitarian medicine. ALIMA, "What we do"
    • In addition to directly employing staff, ALIMA provides financial support to several local implementing partners. "ALIMA works hand-in-hand with our local partners, an integral part of our governance model. Our local partners look to their vast experience to create solutions adapted to local or regional contexts. Departing from conventional humanitarian protocols, a network of partner organizations increases the impact of our medical emergency programs, especially in areas of conflict with high mortality rates. In addition, the majority of our staff are African, representing 98% of all ALIMA employees, at all levels, from country directors to our operational headquarters in Dakar." ALIMA, "Who we are"
    • ALIMA, "Where we work"

  • 2

    "ALIMA works within existing Ministry of Health systems, which include outpatient care as well as inpatient care with pediatric, surgical, and maternity wards." GiveWell's non-verbatim summary of a conversation with ALIMA, March 5th, 2021

  • 3
    • "Wasting refers to a child who is too thin for his or her height. Wasting is the result of recent rapid weight loss or the failure to gain weight." UNICEF, WHO, World Bank, "Joint child malnutrition estimates — levels and trends," 2020, p. 2
    • “Acute malnutrition is a major public health issue in low-income countries. It includes both wasting and edematous malnutrition, but the terms wasting and acute malnutrition are often used interchangeably." Frison, Checchi, and Kerac 2015, Abstract.
    • “Fluid retention is also called oedema or water retention. It occurs when parts of the body swell due to a build-up of trapped fluid. The fluid gets trapped and makes the area swollen or puffy.” HealthDirect, "Fluid retention"
    • "In children aged 6–59 months, moderate acute malnutrition is defined as moderate wasting (i.e. weight-for-height between –3 and –2 Z-scores of the WHO Child Growth Standards median) and/or mid-upper-arm circumference (MUAC) greater or equal to 115 mm and less than 125 mm." WHO, "Supplementary foods for the management of moderate acute malnutrition in children aged 6–59 months," 2019.

  • 4
    • "Restricted growth as a result of inadequate nutrition and infections is an important cause of morbidity and mortality in infants and children worldwide. . . . Several prospective studies have shown associations of undernutrition with increased risk of various disease outcomes, and reduced survival, in children." Olofin et al. 2013, Introduction.
    • "All degrees of underweight, stunting and wasting were associated with significantly higher mortality. The strength of association increased monotonically as Z scores decreased. Pooled mortality HR was 1.52 (95% Confidence Interval 1.28, 1.81) for mild underweight; 2.63 (2.20, 3.14) for moderate underweight; and 9.40 (8.02, 11.03) for severe underweight. Wasting was a stronger determinant of mortality than stunting or underweight." Olofin et al. 2013, Abstract.

  • 5

    "Children with uncomplicated severe acute malnutrition, not requiring to be admitted and who are managed as outpatients, should be given a course of oral antibiotic such as amoxicillin." WHO, "Guideline: updates on the management of severe acute malnutrition in infants and children," 2013, p. 29.

  • 6

    ALIMA confirmed via review of this grant page prior to publication.

  • 7

    "Under this approach [family MUAC], [caregivers] are trained by community health workers to diagnose malnutrition and edema in their own children." GiveWell's non-verbatim summary of a conversation with ALIMA June 9, 2020 (unpublished)

  • 8

    ALIMA confirmed via review of this grant page prior to publication.

  • 9

    ALIMA confirmed via review of this grant page prior to publication.

  • 10

    See row "Monitoring and learning," ALIMA, Chad 2021 final budget.

    • GiveWell, Requested reporting from ALIMA's nutrition programs in Chad (unpublished)

  • 11

    The grant budget was presented as an annual budget in euros; the final amount is based on a conversion to USD at the time the grant agreement was drafted. See here for the full one-year budget.

  • 12

    The grant budget is ~500,000 euros for N'Djamena and ~1.6 million euros for Ngouri per year for three years. The budget is significantly lower in N'Djamena because of co-financing in that location. GiveWell and ALIMA compiled data, 2021 (unpublished). See here in our CEA for these annual costs in USD.

  • 13

    This grant funds operations from:

    • January 2023 through December 2025 in Mirriah, Niger;
    • January 2024 through December 2026 in Dakoro, Niger; and
    • June 2023 through May 2026 in Kaita, Nigeria.

    This grant supported malnutrition treatment for children with MAM or SAM as defined by low MUAC or the presence of nutritional edema. ALIMA does not use weight-for-height-Z score in determining which children to treat.
    We assessed the grant opportunity in Dakoro and Mirriah as a single opportunity because ALIMA initially presented these to us together. We did not expect the cost-effectiveness of these two districts to be significantly different, as the caseloads and budgets presented were similar.

  • 14
    • "But for the most part, creating a single consultation circuit and checking vaccination status of all kids coming for a consultation (not simply malnourished kids) and updating vaccinations as needed is what we feel will have the biggest impact on coverage. Support to EPI activities as well (fuel for and management support of the cold chain, supply management and supervision to avoid stock outs, etc will also be an important contributor.)" Kevin Phelan, Nutrition Advisor, ALIMA, email to GiveWell, October 11, 2022 (unpublished)
    • "This is explained by indirect effects of logistical support for vaccine transport and cold chain; incentives paid to health center staff increasing time present in the health center; participation in fuel supply enabling more vaccine outreach." Susan Shepherd, Senior Advisor, ALIMA, email to GiveWell, October 17, 2022 (unpublished)

  • 15

    Susan Shepherd, Senior Advisor, ALIMA, email to GiveWell, November 2, 2022 (unpublished)

  • 16

    Note: The total budgets shared by ALIMA differ from the final grant amount due to adjustments GiveWell made to account for ready-to-use therapeutic food (RUTF) and survey costs. See ALIMA, 2023-2025 Niger budget proposal (including 10% of RUTF costs) and ALIMA, 2023-2025 Nigeria budget proposal (excluding RUTF costs).

  • 17

    A full breakdown of the grant budget is here in the "Costs" tab of our cost-effectiveness analysis. To calculate the total grant amount, we started with the total three-year budget for each location, excluding ALIMA’s proposed ready-to-use therapeutic food (RUTF) costs, survey costs, and indirect costs. Then, we added costs for annual coverage and mortality surveys ($25,085 per location, per year), a buffer for RUTF needs (10% for Dakoro and Mirriah, and 20% for Kaita), and then added 7%, ALIMA’s indirect cost rate. We did not fund ALIMA’s full RUTF buffer needs because we believed it would be able to obtain funding from other sources.

  • 18

    We learned through ALIMA's review of this grant page that it is not supporting vaccinations in Ngouri, which we considered a significant portion of the pediatric care benefits we modeled. If we were updating our model, we would expect reduced pediatric care benefits, which would lead to lower cost-effectiveness.

  • 19

    See our estimated mortality rates in our CEA for Niger and Nigeria here, and our estimated mortality rates in our CEA for Chad here.

  • 20

    See David Roodman, On the association between anthropometry and mortality in children, 2022, for discussion of methodology.

  • 21

    See David Roodman, On the association between anthropometry and mortality in children, 2022, Table 9, p. 45

    • Ngouri and N'Djamena: Malnutrition treatment CEA (ALIMA Chad 2021 with updated model), "Untreated mortality" sheet, "Mortality rate of untreated children by nutritional status" section.
    • Dakoro, Mirriah, and Kaita: Malnutrition treatment CEA (combined protocol ALIMA), "Untreated mortality" sheet, "Mortality rate of untreated children by nutritional status" section.

  • 22
    • Ngouri and N'Djamena: Malnutrition treatment CEA (ALIMA Chad 2021 with updated model), "Untreated mortality" sheet, "Mortality rate of untreated children by nutritional status" section.
    • Dakoro, Mirriah, and Kaita: Malnutrition treatment CEA (combined protocol ALIMA), "Untreated mortality" sheet, "Mortality rate of untreated children by nutritional status" section.

    The mortality ratio we estimate for untreated versus treated SAM is 3.7 for Chad, 3.4 for Nigeria, and 3.2 for Niger.

  • 23

    See here in our CEA for Niger and Nigeria and here in our CEA for Chad for details.

  • 24

    The plausibility adjustments from our ceiling analyses were 41% for Niger, 37% for Nigeria, 43% for Ngouri, Chad, and 72% for N'Djamena, Chad. See our ceiling analysis for Niger and Nigeria here and our ceiling analysis for Chad here.

  • 25

    See the "Treatment effect" sheet of our CEA for Niger and Nigeria here and for Chad here.

  • 26

    See GiveWell, Impact of CMAM on WHZ for this analysis.

  • 27

    Percent mortality reduction is calculated by taking the inverse of the mortality ratios in table 10 of Roodman 2022 and subtracting them from 1. Using Niger as an example, the mortality ratio for NGO-supported malnutrition treatment vs. no treatment is 1.71 for MAM and 3.23 for SAM:

    • Mortality reduction from MAM treatment: 1 - (1 / 1.71) = 0.42 (0.58 relative risk of mortality with MAM treatment)
    • Mortality reduction from SAM treatment: 1 - (1 / 3.23) = 0.69 (0.31 relative risk of mortality with SAM)
    • Our estimates for Nigeria and Chad are similar to those for Niger.

    David Roodman, On the association between anthropometry and mortality in children, 2022, table 10, p. 45.

  • 28

    See here in our CEA for Chad and here in our CEA for Niger and Nigeria.

  • 29
    • Chad: We estimated that ALIMA will annually cover about 15,000 children with MAM and 6,000 children with SAM who would not have been treated without the program.
    • Niger and Nigeria: We estimated that ALIMA will annually cover about 36,000 children with MAM and 27,000 children with SAM who would not have been treated without the program.

  • 30
    • See our caseload estimates for Chad in our CEA here.
    • See our caseload calculations for Kaita in our CEA here.

  • 31
    • See our caseload calculations for Dakoro and Mirriah in our CEA here.
    • See our caseload calculations for Kaita in our CEA here.

  • 32

    In Dakoro and Mirriah, vaccination accounts for 26% of total benefits, and malaria testing and treatment accounts for 6% of total benefits. As noted below, in Ngouri we roughly estimated a 20% adjustment for benefits from other pediatric care.

  • 33

    In Kaita, vaccination accounts for 6% of total benefits, and malaria testing and treatment accounts for 1% of total benefits.

  • 34

    See here in our CEA for Chad. We learned through ALIMA's review of this grant page that it is not supporting vaccinations in Ngouri, which we considered a significant portion of pediatric care benefits. If we were updating our model, we would expect pediatric care to make up a lower proportion of program benefits for Ngouri.

  • 35

    ALIMA, Niger vaccination survey, 2016, Tableau 18, p. 32 (in French). See here in our CEA for our analysis of this data. The second survey provided by ALIMA, which is unpublished, provides evidence that in areas of Niger where it implements malnutrition treatment programs, the relative rate of vaccination in children under two is about twice as high as in neighboring areas where ALIMA does not work.

  • 36

    ALIMA has provided unpublished evidence from Niger supporting the idea that the intervention increases cumulative vaccination rates and implies that infants are also likely to be vaccinated closer to recommended ages in ALIMA areas. The evidence we have does not allow us to quantify how much earlier this program causes children to be vaccinated on average, so we rely on a subjective guess of four months. See our calculations in our CEA here.

  • 37

    "No other treatment is being provided in this area (no humanitarian partner, no specific treatment in place)." GiveWell/ALIMA compiled data (unpublished)

  • 38

    See our ceiling analysis for Niger and Nigeria here and our ceiling analysis for Chad here.

  • 39

    David Roodman, On the association between anthropometry and mortality in children, 2022

  • 40

    ALIMA, comment on draft of this grant page, March 1, 2024.

  • 41

    We learned through ALIMA's review of this grant page that it is not supporting vaccinations in Ngouri, which we considered a significant portion of pediatric care benefits. If we were updating our model, we would expect pediatric care to make up a lower proportion of program benefits for Ngouri.

  • 42

    See this section of our CEA for Niger for the vaccination survey data.

  • 43

    See this section of our CEA for our rough calculations on the benefits of moving vaccination earlier.

  • 44

    Note that we are not publishing the 2021 version of our CEA, as it contains information that is now out of date.

  • 45

    The OptiMA protocol is a simplified treatment for malnutrition.

    "ALIMA’s Optimizing treatment for acute MAlnutrition (OptiMA) is one such strategy, proposing three main changes to current protocols:

    • "Earlier detection by training mothers and caregivers how to use mid-upper arm circumference (MUAC) bands to screen children regularly for malnutrition in the home (i.e., family MUAC.)
    • "Simplification and easier management by using only one anthropometric measure (MUAC <125 mm (and/or oedema)) for admissions and one product (RUTF) for treatment.
    • "More intelligent use of the costliest input (RUTF) by gradually reducing the dosage based on a child’s MUAC status and weight to increase the number of children with access to treatment at no extra or similar cost." Phelan 2019, p. 40.

  • 46

    Kevin Phelan, Nutrition Advisor, ALIMA, email to GiveWell, October 14, 2021 (unpublished)

  • 47

    "From late November 2021 to early January 2022, ALIMA and its partner Alerte Santé conducted two coverage surveys in Chad, one in an urban area and one in a rural area." ALIMA, Summary Coverage Surveys - Ngouri, 2023 (unpublished)

  • 48
    • Ngouri: ALIMA, Summary Coverage Surveys - Ngouri, 2023 (unpublished)
    • N'Djamena: ALIMA, Notes for GiveWell call, May 17, 2023 (unpublished)

  • 49

    Kevin Phelan, Nutrition Advisor, ALIMA, email to GiveWell, May 11, 2023 (unpublished)

  • 50

    ALIMA confirmed this as part of their review of this page.