Seasonal Malaria Chemoprevention

In a nutshell

Seasonal malaria chemoprevention (SMC) involves giving children monthly courses of antimalarial medicines during the high malaria season (in places where malaria is seasonal). We estimate that it costs approximately $2,000 to $7,000 (depending on the location) to avert a death in areas where GiveWell supports SMC. We think SMC is cost-effective because we think:

  • SMC reaches a high proportion (around 80% or more) of targeted children (who would not otherwise receive SMC).
  • There is strong evidence that SMC reduces malaria.
  • SMC probably provides substantial additional benefits like increased later-life income.

Overall, we are confident that the case for SMC is strong, and we have relatively few significant reservations (even compared to GiveWell’s other top charities). But we are still uncertain about factors including:

  • whether SMC campaigns would be funded by other malaria funders in GiveWell’s absence (which could mean our funding is not increasing the overall number of children reached with SMC as much as we think)
  • observational studies of large-scale SMC programs finding smaller effects than randomized controlled trials
  • the impact of drug resistance on SMC, and how far the scale-up of SMC will increase drug resistance in the future

GiveWell’s current top recommended organization for SMC is Malaria Consortium. Our page for Malaria Consortium’s SMC program is available here, and our cost-effectiveness analysis is here.

Table of Contents

Published: January 2024 (October 2018 version, November 2017 version)

Summary

Basics

Seasonal malaria chemoprevention (SMC) involves giving children monthly courses of antimalarial medicines, usually for four or five months per year, in locations where malaria is highly seasonal (i.e., a high proportion of cases occur in a relatively short period each year).

GiveWell funding to Malaria Consortium supports SMC campaigns. Malaria Consortium uses the funding to (among other things) buy the medicines used in SMC, train community distributors to deliver the medicines door-to-door, and conduct monitoring to understand what proportion of children are reached. (More)

How cost-effective is it?

As of December 2023, we estimate that it costs approximately $2,000 to $7,000 (varying by location)1 to avert a death through SMC in areas where GiveWell supports SMC programs. This equates to being 10 to 28 times as effective as spending on unconditional cash transfers (GiveWell’s benchmark for comparing different programs).

We think that SMC is cost-effective because:

  • Malaria is a major cause of child mortality in areas where GiveWell funds SMC. We estimate that the chance that a 3-59 month old child (the main target group for SMC) will die as a result of malaria is around 0.3% to 0.7% per year, depending on the location. We rely primarily on malaria-specific mortality estimates from the Global Burden of Disease (GBD) Model, which we think are broadly in line with World Health Organization (WHO) estimates, another widely used source. (More) We then adjust these estimates upward because we think malaria indirectly causes 0.75 deaths from other causes for every direct malaria death. This is based on evidence that malaria control programs often have larger impacts on mortality than would be expected from their impact on malaria alone. (More)
  • SMC is effective at preventing malaria. We estimate that receiving SMC reduces malaria mortality by 79% during the period when it is delivered. This is based on a meta-analysis finding that SMC reduces malaria cases by 75%, to which we apply some small adjustments (more). We also assume that reduced malaria cases translate 1:1 into reduced malaria deaths (more). Our impression is that SMC is widely viewed in the global health community as an effective program, strengthening our confidence in these estimates. (More)
  • SMC is a highly targeted program. It is delivered to a group at high risk of malaria (young children) in the high transmission season where most malaria occurs. We currently estimate that 70% of malaria occurs in this period in most countries in our analysis (more). We think this targeted program design contributes to its cost-effectiveness.
  • It is relatively cheap to reach children with SMC. We estimate that it costs Malaria Consortium approximately $5 to $6 per year (depending on the location) to reach a child with all their recommended cycles of SMC. This is roughly in line with the cost of other mass campaign programs we have seen (e.g., insecticide-treated net campaigns). Intuitively, reasons for SMC being relatively cheap include the drugs used being affordable and Malaria Consortium’s door-to-door campaign approach reaching a high proportion of targeted children. (More)
  • SMC probably provides significant benefits beyond averting child mortality. In particular, we think that by averting malaria during a sensitive period of childhood development, it could improve children’s income in later life. This is based on two studies that find historical malaria eradication campaigns led to long-term increases in income. We use a combined estimate from these studies and discount this by 70% to reflect our uncertainty about the quality of this evidence. Even with this discount, we estimate that approximately 15% to 35% of the total modeled benefits of SMC come from increased income rather than averted deaths. (More)
  • GiveWell funding increases the proportion of people protected by SMC. Without campaigns like the ones GiveWell funds, we estimate that virtually no children would get SMC from other sources. This is because campaigns are the only form of SMC delivery we have heard about, and our understanding is that WHO recommends that national governments prohibit the private sale of the SMC drugs in areas implementing SMC (more). We think it’s more likely that another funder would replace GiveWell’s funding for SMC campaigns in our absence (we roughly estimate a 10% to 65% likelihood depending on the location), since we think SMC is a high priority for funders and national malaria programs. Our adjustment for “diverting other actors’ spending away from SMC” lowers cost-effectiveness by 7% to 51%, depending on the location. But we think SMC remains cost-effective even after attempting to account for this possibility. (More)

We quantify this using a cost-effectiveness analysis, which allows us to compare across different programs. Here is a sketch, using estimates for Burkina Faso as an example.

What we are estimating Best guess (rounded) Confidence intervals
(25th - 75th percentile)
Implied cost-effectiveness
Donation to Malaria Consortium (arbitrary value) $1,000,000
Cost per child reached with SMC (more) $5.69 $5.30 - $6.50 19x - 16x
Number of children receiving SMC (more) 176,000
Percent of children who would have received SMC without mass distribution (more) 0%
Annual mortality rate from malaria and associated causes among children who do not receive SMC (more) 0.67% 0.31% - 0.95% 8x - 25x
Proportion of malaria mortality occurring in SMC season (more) 70% 62% - 79% 16x - 20x
Reduction in malaria mortality from receiving SMC (more) 79% 67% - 85% 15x - 19x
Initial cost-effectiveness estimate
Cost per death averted (child mortality only) ~$1,500
Moral weight for each death averted 116
Initial cost-effectiveness estimate 23x
Summary of primary benefits (% of modeled benefits)
Reduced child mortality 80%
Reduced mortality among older children and adults (more) 4% 2% - 8% 17x - 19x
Income increases in later life (more) 16% 9% - 22% 17x - 19x
Additional adjustments
Adjustment for additional program benefits and downsides (more) 19% 5% - 34% 16x - 20x
Adjustment for grantee-level factors (more) -8% -17% - -4% 16x - 19x
Adjustment for diverting other actors’ spending into SMC (“leverage”) (more) -0.3%
Adjustment for diverting other actors’ spending away from SMC (“funging”) (more) -43% -64% - -23% 11x - 24x
Overall cost-effectiveness (multiples of cash transfers) 18x

We’ve also considered other perspectives that might not be captured explicitly in these cost-effectiveness estimates (e.g., whether experts see SMC as a good investment). Overall, we don’t think any of these alternative perspectives undermine the case for SMC. Factors we’ve considered:

  • SMC seems to have widespread support from the global malaria community.
  • SMC appears to have relatively few unintended negative consequences.
  • There is evidence that SMC and the RTS,S vaccine delivered together are more effective than either intervention alone. We see this as important because we expect the vaccine rollout to play a major role in the malaria landscape in the future.

However, we have spent considerably less time on these questions than we have on our main cost-effectiveness model. We hope to engage more with these questions in the future. (More)

How could we be wrong?

Overall, we are confident that the case for SMC is strong, and we have relatively few significant reservations (even compared to GiveWell’s other top charities). But we are still uncertain about questions including:

  • Are we crowding out other funders? We think there is a high chance (10% to 65% depending on the location) that SMC campaigns would be funded by other malaria funders in GiveWell’s absence. If this were the case, it would imply that our funding for SMC is less impactful than we think because it is not leading to an increase in the overall number of children treated with SMC. We attempt to adjust for this based on our analysis of malaria funding trends and conversations with national malaria programs and other funders, but our guesses about this are necessarily speculative. We are also uncertain about whether our funding will create a future expectation of continued SMC funding and reduce other funders’ future spending on SMC.

    Our current adjustment for diverting other actors’ funding away from SMC, which attempts to account for this factor, decreases cost-effectiveness by about 43% in Burkina Faso, and our 25th-75th percentile confidence interval is -64% to -23%. This implies that SMC is 11x to 24 times as cost-effective as direct cash transfers (“11x-24x”). (More) We hope to improve our estimates on this question in the future by investigating whether funding gaps that we didn’t fill were covered by other funders, but we haven’t conducted this analysis systematically yet. (More)

  • Is there evidence that large-scale SMC programs lead to reductions in malaria? Although we generally rely on randomized controlled trials (RCTs), we are more confident if evidence from real-world programs shows an impact similar to that of studies conducted in experimental conditions. We have seen one large study of SMC in a real-world context that finds significantly smaller impacts (reductions in malaria cases of about 25% to 55%) than our analysis would suggest. Although we still see the experimental results as a more reliable guide to the impact of SMC, we see this as a negative update. We’re not sure what’s driving this difference, and we plan to investigate this in more detail in the future. (More)
  • How much of a threat is drug resistance? It is likely that widespread delivery of SMC will lead to increased resistance to the antimalarial drugs used in SMC. We include a -4% adjustment to account for this, but we have not reviewed data on this question since 2020, and it is possible we are underestimating the growth of resistance over time. This adjustment also only accounts for current drug resistance to SMC, and not for further development of resistance in the future. We are planning to spend more time analyzing resistance data and planning how to account for this in the future. (More)
  • Will SMC remain impactful in the future? We’re unsure how changes in the malaria landscape will affect SMC. One trend we plan to monitor is the rollout of the malaria vaccine. This could lead us to estimate that SMC is either more cost-effective (because there is an opportunity to layer both interventions) or less cost-effective (because overall malaria burden could be lower). We’re planning to investigate this question in more detail in the future. (More)
  • How reliable are the malaria mortality estimates we rely on? Record keeping on the causes of mortality in malaria-endemic countries is not reliable. Our analysis relies on malaria mortality estimates from the Global Burden of Disease Project, and we’re unsure how accurate they are (more). We also adjust these estimates by assuming that malaria contributes to 0.75 deaths from other causes for each death attributed directly to malaria. This is a highly uncertain best guess (more). Our 25th-75th percentile confidence interval for mortality caused by malaria in Burkina Faso is 0.31% to 0.95%, which implies a cost-effectiveness of 8x to 25x.
  • Would other ways of supporting SMC be more effective? To date, GiveWell has only supported SMC through direct funding for mass campaigns. It might be that we’re missing other ways of funding SMC that would be more effective (e.g., an advocacy campaign to encourage other funders to prioritize SMC). As of December 2023, we have only just begun thinking about this question, and we’re not sure whether or how it will affect our grant-making. (More)
  • How accurate was our analysis of SMC in hindsight? GiveWell’s cost-effectiveness analyses are “forward-looking” and aim to predict the impact a program will have at the time we make a grant decision. We have paid less attention to backwards checks to understand how accurate the predictions in our SMC grants were. This is a weakness in our approach and something we aim to improve in the future. (More)
Cost-effectiveness analysis accompanying this report: Link

Note: The figures in this report are from our December 2023 cost-effectiveness analysis. Our estimates change over time as we gather new information and update our analysis, and so the numbers in this report may not exactly match those in our most recent cost-effectiveness analysis (available here).

GiveWell’s current top recommended organization for SMC is Malaria Consortium. Our page on Malaria Consortium’s SMC program is available here.

1. The basics of the program

1.1 What is malaria and what are its impacts?

Malaria is a disease caused by Plasmodium2 parasites that are transmitted to people through the bites of infected mosquitoes.3 Symptomatic cases involve flu-like symptoms including fever, which can progress to severe illness or death.4 Some groups are at particularly high risk of severe symptoms including infants, children under five years of age, pregnant women, travelers, and people with HIV or AIDS.5

Globally, malaria deaths fell from an estimated 897,000 in 2000 to 619,000 in 2021.6 There were an estimated 249 million malaria cases in 2022.7 The World Health Organization (WHO) African Region carries a disproportionately high share of the malaria burden. In 2022, the region accounted for approximately 94% of all malaria cases and 95% of deaths.8

1.2 What is seasonal malaria chemoprevention (SMC)?

Seasonal malaria chemoprevention (SMC) involves giving children monthly courses of antimalarial medicines in locations where malaria is highly seasonal (i.e., a high proportion of cases occur during a relatively short period each year). SMC is delivered to all children in a given location9 (exceptions in footnote).10 The antimalarial medicines used are sulfadoxine–pyrimethamine (SP) and amodiaquine (AQ).11 SMC has scaled up rapidly in recent years. An estimated 2.6 million children were reached with SMC in 2014, rising to approximately 49 million in 2022.12

WHO has recommended SMC for deployment since 2012.13 The original WHO recommendation was for up to four monthly cycles of medicine for children aged 3–59 months in areas where more than 60% of the annual incidence of malaria occurs within four months and where resistance to the medicines used in SMC was low.14

In recent years, some countries began to expand their use of SMC beyond the original WHO recommendation. These changes included delivering five cycles rather than four in some locations in Nigeria and Burkina Faso beginning in 202115 and pilots of SMC in Mozambique and Uganda in 2020-2021.16 In 2022, WHO updated its policy recommendation for SMC in line with these developments. The key changes included:17

  • dropping the requirement that SMC be delivered only in places where resistance is low
  • allowing for variability in the number of cycles delivered
  • more flexibility on the age range for targeted children (although in most countries, WHO states that children under five will still be the highest priority)18

As a result, SMC—which has historically only been delivered at large scale in countries in the Sahel region of West Africa, where drug resistance is low—is expanding in other areas with highly seasonal malaria transmission. Countries may also decide to deliver greater or fewer cycles of SMC, in line with the length of their peak malaria transmission seasons.19

GiveWell has funded a number of pilots and research projects to test the feasibility, acceptability, and impact of SMC outside the Sahel.20 At the time of writing (December 2023) we have not fully completed or published our analysis of SMC outside the Sahel. Our analysis in this report therefore focuses on analyzing the impact of SMC in the Sahel.

1.3 How do SMC campaigns work?

The organization GiveWell has funded to date for SMC is Malaria Consortium. Key features of Malaria Consortium’s SMC program include:21

  • SMC is primarily delivered door-to-door by trained health workers ("community distributors") to eligible children.
  • The antimalarial medicines are administered over three days. The community distributor administers the first day’s doses of SMC (one tablet of SP and one tablet of AQ) and gives the remaining doses (two tablets of AQ) to caregivers to administer over the next two days.

SMC campaigns are implemented under the leadership of each country’s national malaria program. Malaria Consortium collaborates with national malaria programs and other partners to support these campaigns. Its role includes providing the funding for the campaigns,22 procuring the antimalarial medications, training community distributors, and monitoring and evaluation.23

Our review of Malaria Consortium’s SMC program has more information.

2. How GiveWell estimates cost-effectiveness

GiveWell recommends programs that we believe save or improve lives as much as possible for as little money as possible. To estimate this, we produce a cost-effectiveness analysis (“CEA”) which aims to produce a best guess of the overall impact of a program per dollar donated.

We use moral weights to quantify the benefits of different impacts (e.g., increased income or reduced deaths). We benchmark to a value of 1, which we define as the value of doubling someone’s consumption for one year. The main moral weights we use for our analysis of SMC are in the table below.

Moral weight
(units of value
per outcome)
Doubling consumption for one person for one year 1
Value of increased income through averting a case of malaria in a person under age 15 with SMC 0.23
Averting the death of a child under five from malaria 116
Averting the death of someone aged over five from malaria 73

For more about how GiveWell thinks about cost-effectiveness, see our discussion on this page.

As of December 2023, our SMC cost-effectiveness analysis contains 18 locations in seven countries.24 In this report, our analysis focuses on Malaria Consortium’s SMC program in Burkina Faso, Togo, and various states in Nigeria only. We largely exclude the other locations from our discussion (details in footnote).25

This report and accompanying cost-effectiveness analysis include 25th-75th percentile confidence intervals for specific parameters. See the summary table above and this sheet of our cost-effectiveness analysis. These intervals are based on GiveWell staff members’ subjective levels of uncertainty for each parameter (see footnote for more details on our method).26

3. How many people does SMC reach?

3.1 Summary

The starting point for our analysis is an estimate of the number of children protected with SMC per $1 million spent.27 As of December 2023, we estimate that $1 million protects approximately 160,000 to 200,000 children per year with SMC (varying by location).28 A summary of our calculations is below, using one country (Burkina Faso) as an example:

What we are estimating Value (rounded)
Total number of SMC cycles delivered in Malaria Consortium-supported campaigns in Burkina Faso, 2018-2021, adjusted for non-adherence (more) ~21 million29
SMC spending by Malaria Consortium in Burkina Faso, 2018-2021 (more) ~$27.5 million
Costs incurred by other NGOs on Malaria Consortium-supported SMC campaigns in Burkina Faso, 2018-2021 (more) $250,000
Subtotal: Average cost per SMC cycle delivered, excluding in-kind government costs $1.32
Donation to Malaria Consortium (arbitrary value) $1,000,000
Average number of cycles delivered per year in Burkina Faso (varies by district) (more) 4.3
Subtotal: Cost per year of SMC delivered $5.69
Total (children treated per year with SMC) ~176,000

The main uncertainties in our estimates are (more below):

  • Our analysis relies on target population estimates (the number of 3-59 month old children in each location). We think these are drawn from administrative data, and we’re unsure how reliable they are.
  • We include other philanthropic actors’ financial contributions in our analysis, but we do not have any information on these actors’ costs in Togo. We therefore use a weighted average of other countries’ cost figures in Togo, which we would expect to be less accurate.

3.2 Cost per cycle of SMC

We calculate the cost per cycle of SMC delivered in each country based on three main estimates:

  • the number of SMC cycles delivered by previous Malaria Consortium-supported SMC programs
  • costs incurred by Malaria Consortium
  • costs incurred by other actors

We divide total costs by total cycles delivered to obtain the estimated cost per cycle in each country. This spreadsheet contains our full calculations.

Number of cycles delivered

We estimate the number of cycles delivered in SMC campaigns between 2018 and 202130 in this sheet using the following sources of data:

  • Target populations (i.e., the number of children aged 3 to 59 months) in districts with Malaria Consortium-supported campaigns.31 These figures are provided by Malaria Consortium, and our understanding is that they are based on administrative data from national governments. A key uncertainty in our analysis (discussed below) is that these may be based on outdated census data, updated with estimates of population growth.32
  • SMC coverage (i.e., the proportion of targeted children who receive SMC). Malaria Consortium conducts surveys after campaigns to understand what proportion of children in the target population received SMC. These surveys are discussed in detail in our separate review of Malaria Consortium. In summary:
    • Malaria Consortium conducts separate surveys after each cycle of SMC (i.e., every month during the SMC season) and each SMC round (i.e., after the SMC season is complete, four or five months depending on the country).33 We use coverage estimates from the post-round surveys, which we think are likely to be more accurate (details in footnote).34
    • In the most recent program years, post-round surveys measured average coverage across cycles at 92% in Burkina Faso (2018-2020), 85% in Chad (2018-2021), 78% in Nigeria (2018-2021), and 87% in Togo (2020-2021).35
    • Overall, we think that these surveys are high-quality and provide evidence that a high proportion of target children were reached in Malaria Consortium campaigns. Our main reservations are that the surveys rely on caregiver self-reports about whether their children received SMC (which we would guess somewhat inflate the true proportion of children reached) and discrepancies between post-cycle and post-round surveys in some locations.36 We account for the risk of bias these issues introduce with a rough -2% adjustment elsewhere in our analysis.37

For each country, we multiply estimates of SMC coverage38 by the target population to estimate the total number of cycles delivered39 in each SMC campaign.40

Finally, we adjust our estimates of the number of cycles delivered to account for "non-adherence." Community distributors give children the first day of medication for each SMC cycle directly, but caregivers give their children the remaining doses on days two and three.41 We would expect caregivers to somewhat over-report how frequently they adhere to this protocol because of social desirability bias (the tendency of survey participants to over-report "good" behaviors).

As of December 2023, we account for this with an adjustment of 93 to 94% (i.e., -7% to -6%), varying by country (details on our approach in footnote).42

Malaria Consortium’s costs

Our estimates of Malaria Consortium’s cost per cycle are based on figures reported by Malaria Consortium to GiveWell on its spending on SMC campaigns between 2018 and 2021.43 Overall, these figures show that Malaria Consortium spent around $27.5m in Burkina Faso, $19.6m in Chad and $66.9m in Nigeria in these years.44

Other actors’ costs

Other actors also contribute resources to the SMC campaigns that Malaria Consortium supports. The costs we include in our analysis are:

  • Spending by other philanthropic actors on Malaria Consortium-supported campaigns. The campaigns that Malaria Consortium supports sometimes receive additional funding from other NGOs or multilateral organizations (e.g., UNICEF). We estimate that these costs are approximately $3m in Nigeria and $250,000 in Burkina Faso between 2018 and 2021 (details on our approach in footnote).45 In Togo, Malaria Consortium co-funds SMC in the districts it supports with UNICEF and the Global Fund,46 but we do not have information on these organizations’ costs. We therefore exclude Togo from our analysis and use an estimate that the cost per SMC cycle in Togo is the weighted average of all other countries' cost per cycle.47
  • In-kind contributions from national governments. The campaigns that Malaria Consortium supports are ultimately managed by national governments’ ministries of health. We expect that Malaria Consortium’s funding diverts some ministry of health resources (e.g., staff time, office space, etc.) toward SMC campaigns that might otherwise have gone toward other activities. We estimate that these in-kind contributions account for 10-11% of total costs (varying by country).48 This figure is based on a costing study conducted as part of ACCESS-SMC, a program in which Malaria Consortium was one participant, in 2015 and 2016.49

We use the cost per supplement excluding in-kind government contributions when estimating the number of children reached per $1m spent by Malaria Consortium (summarized in the table above).50 We consider these contributions (but not financial contributions from other philanthropic actors) that we include in our analysis to be “leveraged” (i.e., Malaria Consortium’s spending results in more of these resources being used for SMC and less for other activities that we think are less cost-effective).51 We account for this in our adjustment for other actors’ spending:

  • We exclude these costs from the cost side of the cost-effectiveness equation in the main part of our analysis.
  • However, the benefit of these costs is already incorporated in our initial impact calculations. To account for them on the benefits side, we adjust the impact of the program downward to account for those funds not being spent on something else.
  • We think this approach is the best way to account for situations where GiveWell funding diverts other actors’ spending from a less cost-effective to a more cost-effective use. See this blog post for more.

Shortcomings and uncertainties

Our biggest uncertainties (ways in which our estimates could be wrong and which could affect our bottom line the most) are:

  • Target population data. Our estimates of the number of children reached are based on data on the number of children aged 3 to 59 months in districts where Malaria Consortium works. These figures are based on administrative data from national governments, often old census data that has been adjusted to account for population growth. These may be inaccurate for various reasons (e.g., differential population growth at the local level, migration, or internal displacement).52
  • Lack of information on other actors’ spending in Togo. We use a weighted average figure for other countries ($1.50) in place of a cost-per-cycle estimate in Togo. This is because Malaria Consortium co-funds its program in Togo with UNICEF and the Global Fund,53 but until recently we did not have information on these other actors’ costs.54 We would therefore expect our estimate in Togo to be less accurate than in other locations.
  • Not accounting for older children being reached. Our current estimates assume that Malaria Consortium only delivers SMC to children under age five. In fact, Malaria Consortium’s surveys find that some older children are reached too (e.g., because a caregiver misreports a child’s age or distributors aren’t strict in enforcing age cut-offs).55 By not accounting for these children, we may be underestimating the number of children reached and the benefits of SMC. We have conducted some initial analysis suggesting that accounting for this could increase cost-effectiveness by roughly 5% to 10%, but this is a rough estimate and we have not yet fully vetted or published this work.
  • Lack of cross checks. We have not systematically cross referenced our estimates against independent data sources (e.g., costing studies conducted by independent research groups). We have conducted a rough internal analysis of nine SMC costing study estimates, indicating our estimates were slightly lower than but in roughly the same range as other published data (details in footnote).56 However, we have not investigated this in detail (e.g., to check whether the programs evaluated were similar to Malaria Consortium’s SMC program or to evaluate the costing methodology used), or published our analysis. We may do more work on this in the future.
  • Not adjusting for survey biases. Our estimates of the number of SMC cycles delivered rely on coverage surveys that do not account for survey biases.57 We have conducted some rough internal analysis suggesting that we may be modestly overestimating coverage by not accounting for social desirability bias (survey participants tending to overreport “good” behaviors), but we have not yet deeply investigated this question or published this work. We plan to consider this in more detail in the future.

3.3 The number of cycles delivered per year

WHO recommends that SMC campaigns should deliver between three and five cycles of SMC per year at 28-day intervals, with the number of cycles depending on the local context (e.g., the length of the high-transmission season).58 The campaigns that Malaria Consortium supports deliver either four or five cycles, depending on the location. In Burkina Faso, where some districts receive four cycles of SMC and some receive five,59 we use an average figure of 4.3. This is because Malaria Consortium estimates that 70% of the target population lives in locations that receive four cycles and 30% lives in locations that receive five cycles.60

We multiply the number of cycles per SMC campaign by the cost per cycle (excluding in-kind government costs) to estimate the overall cost per child per year (approximately $5 to around $6.30, varying by location).61

4. What impact does SMC have?

4.1 Summary

Our cost-effectiveness analysis models three main benefits from SMC:

  1. Reduced mortality for children under age five (more).
  2. Reduced mortality for older children and adults, from reduced malaria transmission in the community (more).
  3. Increased long-run income, from improved health in a sensitive developmental window of childhood (more).

A summary of the contributions of each type of benefit to our estimate of the modeled value of the program is below, using Burkina Faso as an example.62

What we are estimating % modeled benefits
Reduced mortality for children under five 80%
Reduced mortality for older children and adults (age five+) 4%
Long-term income increases 16%

We also include a number of supplemental adjustments to account for additional benefits and offsetting impacts. Rather than explicitly modeling these, we have applied percentage adjustments based on our best guesses.

  • We divide these into intervention-level factors (e.g., costs averted from treatment of illness), which increase our cost-effectiveness estimate by 19% to 34% overall (19% in Burkina Faso) (more), and grantee-level factors (relating to the implementation of the program), which reduce our estimate by 8% (more).
  • We also include an adjustment to incorporate the impact of Malaria Consortium’s funding on other actors’ spending. This reduces our estimate by 8% to 51% (43% in Burkina Faso) (more).
  • After factoring in all these impacts, we estimate that it costs approximately $2,000 to $7,000 (varying by location)63 to avert a death through SMC. This equates to being approximately 10 - 28x as effective as spending on unconditional cash transfers.64

4.2 Reduced mortality for young children

What is the impact of SMC on malaria?

Summary

We estimate that SMC reduces malaria cases among children receiving SMC by 79% during the SMC season.65 We also estimate a 1:1 ratio between reduced malaria cases and reduced malaria deaths,66 implying that SMC reduces malaria deaths among children receiving SMC by 79% in the SMC season. A summary of our calculations is below:

What we are estimating Value
The impact of being targeted for SMC, adapted from a published meta-analysis (more) 75%
Adjustment for internal validity (study quality) (more) -5%
Proportion of children who received SMC in studies in the meta-analysis (used to convert to the impact of receiving SMC) (more) 90%
Ratio of reduced malaria cases to reduced malaria deaths (more) 100%
Total (reduction in malaria deaths among children receiving SMC) 79%
The impact of being targeted for SMC

To estimate the impact of SMC, we rely on a systematic review and Cochrane meta-analysis67 , Meremikwu et al. 2012. This paper summarizes seven randomized controlled trials (RCTs) of SMC (then known as IPTc, intermittent preventive treatment for malaria in children).68

The review’s main finding is that being targeted for SMC reduces malaria cases (clinical episodes) by approximately three-quarters (74%).69 We incorporate this figure into our cost-effectiveness analysis after applying several small adjustments.

Overview of the Cochrane meta-analysis

Meremikwu et al. 2012 summarizes seven RCTs involving 12,589 participants.70

The review’s main finding is that SMC reduces malaria cases by 74% (rate ratio 0.26, 95% CI71 0.17 to 0.38, data from six studies).72 Secondary analyses included:

  • Severe malaria: SMC reduced severe malaria cases by 73% (rate ratio 0.27, 95% CI 0.10 to 0.76, two studies).73
  • Co-delivery with insecticide-treated nets: Two studies distributed nets to both intervention and control groups. SMC retained high efficacy in these studies against malaria cases (rate ratio 0.22, 95% CI 0.13 to 0.38).74
  • All-cause mortality: There was no statistically significant impact of SMC on all-cause mortality (risk ratio 0.66, 95% CI 0.31 to 1.39), but the total number of deaths across all studies was very low and the studies were underpowered to detect an impact on all-cause mortality.75 The review concludes that SMC “probably produces a small reduction in all-cause mortality consistent with the effect on severe malaria.”76

The review authors assessed each study for its risk of bias and determined that the evidence for malaria cases, severe malaria, and co-delivery with insecticide-treated nets was high quality.77 The evidence for all-cause mortality was assessed as moderate quality.78 We have not independently assessed these studies’ risk of bias.

Adjustments to the meta-analysis

We conducted an updated version of the Cochrane meta-analysis of SMC on malaria cases:

  • We identified two additional RCTs (Tagbor et al. 2016, Matangila et al. 2015) published after the Cochrane review. Both found smaller effect sizes than the Cochrane meta-analysis.79 We added Tagbor et al. 2016 to our updated meta-analysis, but did not add Matangila et al. 2015 because courses were delivered at three-month (rather than one-month) intervals, and so are unlikely to have achieved similar reductions in malaria incidence to Malaria Consortium’s SMC program.80
  • One of the RCTs (Dicko et al. 2008) included in the meta-analysis treated participants every eight weeks (rather than monthly, as in Malaria Consortium’s SMC program).81 We removed this study from our version of the meta-analysis.

This leaves six studies in the updated meta-analysis that measured the impact of SMC on malaria cases.82 After applying these updates, we estimate a very similar impact of SMC to the original Cochrane analysis (75% reduction; rate ratio 0.25; 95% CI 0.18 to 0.37).83

A World Health Organization report claims that SMC offers a high level of protection for about four weeks after the last course of medication, after which protection appears to decline.84 We have not independently vetted this claim. We note that some of the trials continued to collect malaria incidence up to six weeks after the last course was administered, which could underestimate the level of protection over a four-week period.85 In 2023, we received feedback from Professor Paul Milligan, Professor of Epidemiology and Medical Statistics at the London School of Hygiene and Tropical Medicine, that using the main estimate from Meremikwu et al. 2012 to estimate the impact of SMC programs could mean we’re underestimating the impact of SMC on malaria (details in footnote).86 We have not yet deeply investigated this claim or understood how we could update our analysis to account for this.

The Cochrane review also notes there is substantial heterogeneity between the program results, meaning the variance between results of different trials is unlikely to have occurred by chance.87 The authors note there are insufficient trials to make meaningful conclusions about the causes of this heterogeneity.88 We reviewed two of the studies that find substantially lower reductions in malaria incidence (Dicko et al. 2008; Tagbor et al. 2016) and note possible causes in this footnote.89

Comparing the meta-analysis studies to Malaria Consortium’s SMC program

We examined the six individual RCTs in our updated meta-analysis to understand whether there are any major additional differences between the programs they evaluated and Malaria Consortium’s SMC program.90 Our findings were:

  • All six delivered SMC cycles on a monthly basis, as in Malaria Consortium's program.91
  • Four of the six trials treated children with a combination of sulfadoxine-pyrimethamine (SP) and amodiaquine (AQ), as in Malaria Consortium's program.92
  • All six trials were conducted in West Africa, where most of Malaria Consortium’s programs operate.93
  • All six trials only delivered SMC to children under 59 months of age, as in Malaria Consortium's program.94
  • The number of courses ranged between three and five. Malaria Consortium’s program delivers four or five courses.95
  • Coverage of insecticide treated nets (ITNs) varied widely between <1% and 93%.96

We conclude that the programs evaluated in these studies are reasonably similar to Malaria Consortium's SMC program, and therefore there is no need for an additional adjustment to account for differences between the programs. In our cost-effectiveness analysis, we therefore set our adjustment for external validity (the generalizability of the underlying studies to contemporary contexts) to 0% (i.e., no downward adjustment).97

Adjustment for internal validity (study quality)

We apply an adjustment of 95% (i.e., -5%) to the estimate from the meta-analysis to account for possible problems with the quality of the underlying studies (their "internal validity").98 Our reasoning is:

  • Our analysis suggests that the studies in the Cochrane review and our updated meta-analysis appear to be high-quality overall. Our best guess is therefore that the meta-analysis estimate is close to the true impact of SMC in the populations being studied.
  • Even though we don't have specific, serious concerns about study quality, we use a default value below 100% based on our expectation that published studies are more likely to overstate an intervention's efficacy than understate it (e.g., because of publication bias).99
The impact of receiving SMC

We estimate that the reduction in malaria cases among children who received SMC in the studies in the Cochrane meta-analysis was 79%100 :

  • We get to this figure by dividing the estimated effect of SMC on malaria cases (75%), adjusted for internal validity (95%), by an estimate of the proportion of children in the studies who received SMC (~90%); ((75% x 95%) / 90%) = ~79%).101
  • The estimate of the proportion of children who received SMC is a weighted average of reported coverage rates from the six studies in our meta-analysis. See this spreadsheet for our calculations.

Our uncertainties about this adjustment are:

  • Spillover effects: We think that SMC provides indirect benefits to people who don’t receive SMC by reducing malaria transmission in the wider community (more below). Our current adjustment does not take these "spillover effects" into account. This could affect our analysis in two ways. Our best guess is that these factors lead us to underestimate the effect of SMC on mortality.
    • Spillovers to the control group, which could lead us to underestimate the effect of SMC. All the studies we rely on in our meta-analysis randomized individuals to receive SMC, rather than whole communities.102 Our best guess is that some protection accrued to children in the control group who did not receive SMC (assuming they lived in the same communities as the children who did receive SMC). Our current analysis implicitly assumes that children in the control group got no benefit. This implies we could be underestimating the benefits of SMC on children who receive SMC.
    • Spillovers to children who did not receive SMC in the treatment group, which could lead us to over- or underestimate the effect of SMC. Our conversion implicitly assumes that children in the treatment group who did not receive SMC received no benefit, and all the benefits of SMC were concentrated among the children who did receive SMC. In fact, children who didn’t receive SMC are likely to have gotten some benefit from reduced malaria transmission. We would expect this not to bias our results if coverage of SMC was similar between trial contexts and contemporary SMC programs (because the conversion doesn’t affect the overall benefit of SMC, just how it is allocated between children who do and don’t receive SMC). But the conversion could yield inaccurate estimates of the impact of SMC if coverage rates differed significantly between trial contexts and contemporary SMC campaigns (due to varying transmission patterns). We’re unsure whether this would positively or negatively update our analysis. Our concern about this is mitigated because we believe that coverage in contemporary SMC programs is relatively high (see our discussion above for more detail).
  • Coverage measurement: We have not reviewed in detail how coverage was measured in these studies. We would guess that much of this data was based on self-reports from caregivers. This kind of data is at risk of being inflated by social desirability bias (the tendency for survey participants to overreport “good” behaviors). This implies we could be overestimating coverage (and therefore underestimating the impact of SMC on children who receive it).
The impact of SMC on malaria deaths

The trials included in the Cochrane review were underpowered to detect a significant effect on all-cause mortality.103 However, to convert our estimates of the impact of SMC on malaria cases into an estimate of their impact on mortality, we need to estimate the ratio between malaria case reductions and malaria mortality reductions.

We assume that the reduction in malaria mortality from SMC is proportional to the reduction in malaria incidence (e.g., if SMC reduces malaria cases by 79%, it also reduces malaria mortality by 79%).104 Our reasoning for this assumption is:

  • We would expect that a reduction in clinical malaria from SMC results in a similar reduction in malaria mortality, unless there was evidence that SMC was disproportionately likely to prevent non-severe cases of malaria.
  • The Cochrane review found that SMC prevents a similar proportion of severe malaria episodes (73%, rate ratio 0.27, 95% CI 0.10 to 0.76) and all clinical malaria episodes, though only two trials measured effects on severe malaria.105
  • The Cochrane review found that all-cause mortality in the underlying studies was lower in the treatment groups than the control groups, although this difference did not reach statistical significance. The review authors conclude that “IPTc probably produces a small reduction in all-cause mortality consistent with the effect on severe malaria, but the trials were underpowered to reach statistical significance.”106 We have not yet tested the reduction in all-cause mortality implied by our primary analysis against this conclusion, but may do so in the future.

What is the mortality rate among children who do not receive SMC?

Summary

We estimate that there is around a 0.3%-0.7% risk (varying by location) that a child not receiving SMC will die from malaria or associated causes each year.107 We also assume that 70% of malaria cases occur during the season when SMC is delivered (and that these are the only deaths that could potentially be averted by SMC).108 A summary of our calculations for one country, Burkina Faso, is below as an example:

What we are estimating Value
Annual risk of death for malaria for children not receiving SMC (more) 0.38%
Number of deaths indirectly caused by malaria for every one malaria death directly attributed to malaria (more) 0.75
Subtotal: Annual risk of death attributable (directly and indirectly) to malaria for children not receiving SMC 0.67%
Proportion of malaria mortality occurring during the SMC season (more) 70%
Total (risk of death from malaria or associated causes during the SMC season each year
for children not receiving SMC)109
0.47%
Baseline malaria mortality

Our estimates of malaria mortality among children targeted for SMC are drawn from the Institute of Health Metrics and Evaluation (IHME) Global Burden of Disease (GBD) project. We use 2019 (the most recent published version of GBD at the time of writing) national-level estimates for malaria mortality among children 3-59 months old110 with the exception of Nigeria, where we use state-level estimates.111 In Nigeria (but not other locations), we also apply a rough adjustment of -5 to -35% (varying by state) to the GBD estimates. This is because we think the GBD estimates might not account for a recent increase in insecticide-treated net campaigns in some states in Nigeria, and could therefore overestimate malaria burden (details in footnote).112

This results in estimates of annual malaria mortality rates ranging from 0.16% (Togo) to 0.38% (Burkina Faso).113

These estimates are a source of significant uncertainty in our analysis. Our understanding is that the GBD estimates rely on a number of modeling assumptions,114 in part because raw data on malaria from health surveillance systems is relatively unreliable in many low-income countries.115 We have not investigated all the modeling assumptions underlying these estimates in detail and we’re not sure how accurately GBD can attribute deaths to specific causes.

An additional uncertainty is how the GBD estimates compare to other sources. In 2023, GiveWell commissioned Rethink Priorities to conduct a research project comparing the GBD estimates to widely-used estimates of malaria malaria mortality from WHO.116 Our understanding had been that the WHO estimates were generally lower than the IHME estimates. We were concerned that we could be systematically overestimating malaria mortality by using GBD estimates only.

Overall, the report reduced our concern about this issue. The main findings of the report were:

  • WHO’s estimates were historically significantly lower than GBD estimates (17% lower in 2019), but went through a significant methodology change in 2021 that has narrowed the gap.117
  • The gap between sources is largest in areas outside sub-Saharan Africa, where the malaria disease burden is lower (and where GiveWell does not fund SMC).118
  • GBD estimates tend to record lower deaths among children under five, and higher deaths among older children and adults.119 This might be the result of GBD using verbal autopsy reports to inform its estimates, whereas WHO (and some other researchers) believe that these can’t distinguish between malaria and other febrile illnesses.120

While we’re still uncertain about the reliability of each source, and the verbal autopsy question, we opted not to change data sources following this project. The benefits of SMC are concentrated in children under five (where the gap between IHME and GBD is smaller), and so we think that switching would not not lead to major changes in our bottom line.

Non-malaria deaths indirectly averted

We estimate that for each death directly caused by malaria, an additional 0.75 deaths are indirectly caused by malaria but attributed to other causes.121 Our reasoning is:

  • Deaths may have several contributing factors while only being attributed to one cause. For example, malaria may increase the likelihood of death from malnutrition or other infectious diseases.122
  • Malaria control interventions often have a larger effect on all-cause mortality than would be expected exclusively from declines in malaria-specific mortality, which we interpret as evidence that averting malaria may also avert deaths attributed to other causes.123

We are highly uncertain about what the exact value for this effect should be. Our rough estimate of 0.75 is based on triangulating three different sources of information:

  • We have spoken with malaria experts who told us that it is widely accepted there are roughly 0.5 to 1.0 indirect malaria deaths for every direct malaria death.124
  • In our analysis of insecticide-treated nets (another malaria intervention), we estimate a ratio of up to 1.5 indirect deaths for every direct malaria death. While there are reasons to think this is an overestimate, this updates us toward thinking that high ratios are plausible (details in footnote).125
  • Our analysis of water chlorination programs, another intervention that reduces child mortality by averting infectious diseases, suggests a ratio of 2.7 deaths indirectly averted for every death directly averted from enteric infection.126 This also updates us toward thinking that a high ratio of indirect to direct deaths is plausible.

Taking all three sources of information into account, we use a value of 0.75 as our best guess for the ratio of indirect malaria deaths to direct malaria deaths.127

There are some additional lines of research we could use to check our estimate, but which we haven’t prioritized yet:

  • An analysis of what indirect deaths ratio would be implied by the estimates of reduced malaria and all-cause mortality in the Cochrane analysis (details in footnote).128
  • We have seen one paper, Uyoga et. al. 2019, that uses the presence of the HbAS “sickle cell trait” (which is associated with high protection against malaria) to estimate the impact of malaria on other infectious diseases. The paper finds that the sickle cell trait is associated with strong protection against other conditions, implying additional evidence that malaria worsens other health problems.129 We have not fully looked into this paper or understood its methodology in depth, but we may do so in the future.
How seasonal is malaria mortality?

Across countries, we estimate that 70% of malaria deaths take place in the high-transmission season when SMC is delivered.130 Our analysis is sensitive to this estimate because we assume that SMC only averts malaria during the period when it is delivered.131

Our reasoning for the 70% assumption is:

  • Until 2022, WHO guidelines recommended that SMC should be deployed in areas where 60% or more of annual malaria incidence takes place within four months.132 Following the 2022 revisions to the guidelines, areas with 60% or more seasonal malaria remain the “priority target areas” for SMC.133 We therefore think of 60% as a reasonable minimum bound for this estimate.
  • We have reviewed an estimate from a single 2012 paper, which finds that the median proportion of malaria incidence in SMC-eligible sites occurring in the high-transmission season was 77%, and the mean proportion was 75.7%.134 We then adjust this modestly downward to 70% because our best guess is that, on average, the locations where Malaria Consortium currently delivers SMC have a less seasonal pattern of malaria transmission than those included in the 2012 paper.135
  • Some locations where Malaria Consortium delivers SMC have a less seasonal pattern of rainfall, and our understanding is that rainfall seasonality is correlated with malaria transmission seasonality.136 We would therefore expect these locations to have a lower proportion of malaria cases over an equivalent period of time. However, because of their less seasonal transmission, these locations typically have a five-month SMC delivery period rather than a four-month period.137 We therefore assume that the overall proportion of malaria that could be averted by SMC is the same as in locations with more seasonal transmission (70%).

Our main uncertainty about this is that 70% is a rough best guess. We have not deeply investigated malaria transmission seasonality across locations. We believe that our cost-effectiveness model would benefit from additional research to inform this parameter, particularly data that would enable us to use location-specific estimates.

4.3 Reduced mortality for older children and adults

Summary

We estimate that SMC reduces malaria deaths by approximately 11% among older children and adults.138 This accounts for 2% to 14% of the total modeled benefits of SMC (varying by location).139 In Burkina Faso, our 25th-75th confidence interval is that this accounts for 2% to 8% of the total modeled benefits of SMC, with a best guess of 4%.140

This is because delivering SMC to children under five disrupts malaria transmission, reducing malaria in the wider (untreated) population. A summary of our calculations for one country (Burkina Faso) are below.

What we are estimating Value
Reduction in malaria cases among children receiving SMC, according to one study of impact on the untreated population (Cissé et. al. 2016) (more) 60%
Reduction in malaria cases among untreated older children and adults in Cissé et. al. 2016 (more) 26%
Subtotal: How much did SMC reduce malaria among older children and adults, relative to children under 5, in Cissé et. al. 2016? 43%
Adjustment for differences between Cissé et. al. 2016 and other SMC programs (more) -69%
Reduction in mortality for children under five receiving SMC (discussed above) 79%
Total (expected reduction in malaria mortality among people over age 5) 11%

Impact of SMC on the untreated population

Our analysis relies on a single published RCT we have reviewed, Cissé et. al. 2016. This study was not included in the 2012 Cochrane review. The study’s key features were:

  • It tested SMC in Senegal between 2008 and 2010 using a stepped-wedge study design.141 The sample size was large (approximately 14,000 children in year one, rising to approximately 160,000 in year three).142
  • The intervention was similar to the program implemented by Malaria Consortium, involving monthly distribution of sulfadoxine-pyrimethamine (SP) and amodiaquine (AQ) during the high-transmission season.143 It differed from Malaria Consortium’s program in that children up to age 10 received SMC.144
  • The study found a 60% reduction (95% CI 55% - 67%) in malaria incidence (measured by rapid diagnostic test) in children living in areas that received SMC in a particular transmission season, relative to children living in areas not receiving SMC that transmission season.145
  • Malaria cases in the untreated population (older children and adults) in areas receiving SMC in a given transmission season fell by 26% (95% CI 18% - 33%) compared to people in areas not receiving SMC.146
  • The study did not detect an effect on all-cause mortality.147 It was initially designed to have sufficient power to detect a decline in all-cause mortality,148 but mortality was lower than expected in both the treatment and control areas.149

Overall, we interpret Cissé et. al. 2016 as evidence that SMC reduces malaria among older children and adults who do not receive SMC. We use the 26% estimated reduction in malaria cases among this group as the starting point in our analysis. We divide this figure by the impact of SMC on the treated population in Cissé et. al. 2016 (60%) to estimate the ratio of SMC’s impact in the untreated vs the treated population (26% / 60% = 43%).150

Adjustment for differences in contemporary SMC programs

We account for differences between the Cissé et. al. 2016 study context and the contexts in which Malaria Consortium works today with a -69% adjustment.151 This incorporates two factors:

  • Different age groups. Malaria Consortium’s program delivers SMC to children between the ages of 3 months and 5 years old (rather than up to 10 years old, as in Cissé et. al. 2016).152 We expect that because Malaria Consortium’s program treats a smaller proportion of the population, there would also be smaller effects on the untreated population. We account for this with a 47% (i.e., -53%) adjustment (details in footnote).153
  • Higher malaria transmission rates. We think that Malaria Consortium’s SMC program operates in areas with higher average malaria transmission rates than the areas in Senegal targeted in Cissé et. al. 2016. Our understanding (from an unpublished modeling study) is that this is likely to mean the indirect effects on the untreated population are proportionally lower. We account for this with a 65% (i.e., -35%) adjustment (details in footnote).154

We multiply these factors together to produce our overall -69% adjustment ((47% x 65%) - 100% = -69%).155

Shortcomings and uncertainties

Our adjustment to account for differences between Cissé et. al. 2016 and Malaria Consortium’s program relies on estimates from an unpublished modeling study. We have not vetted the transmission model that the study relies on. One particular point of uncertainty is that the modeling study relies on Zongo et al. 2015, an RCT of SMC using a different drug regimen from the one used in Malaria Consortium’s program.156 We are unsure how this might affect our estimates.

4.4 Long-term income increases

Summary

Our best guess is that averting childhood malaria cases leads to small income gains in adulthood. This contributes a substantial share of our estimated benefits for malaria programs. As of December 2023, we estimate that these benefits account for approximately 15% to 35% of the total modeled benefits of SMC (varying by location).157 In Burkina Faso, our 25th - 75th confidence interval is that this accounts for 9% - 22% of the total modeled benefits of SMC, with a best guess of 16%.158

To create our best guess on the effect of reductions in malaria on later-life income, we rely on evidence from two historical quasi-experiments, Bleakley 2010 and Cutler et al. 2010, that measure the effect of eradicating malaria on later-life income in the Americas and India. Based on this evidence, we estimate that each case of childhood malaria averted increases adult income by 0.6%, and that these benefits persist over 40 years of an individual’s working life. A summary of our calculations is below, using one country, Burkina Faso, as an example:

What we are estimating Value (rounded)
Arbitrary donation to Malaria Consortium $1,000,000
Total children protected by SMC (aged 3 - 59 months) (discussed above) ~176,000
Number of children aged 5 - 14 in the national population for every child aged <5 (more) 2
Subtotal: Number of 5 - 14 year olds exposed to SMC spillovers per $1m spent ~353,000
Increase in annual income per childhood malaria case averted (more) 0.6%
Baseline counterfactual malaria incidence (Age <5) (more) 60%
Baseline counterfactual malaria incidence (Age 5-14) (more) 25%
Reduction in malaria incidence for children aged <5 treated with SMC (discussed above) 79%
Reduction in malaria incidence for children aged 5 - 14 from spillovers (discussed above) 11%
Proportion of malaria cases occurring during the SMC season (discussed above) 70%
Malaria cases averted by SMC per $1m spent (age <5) ~59,000
Malaria cases averted by SMC per $1m spent (age 5 - 14) ~6,500
Years until long-run income benefits start (more) 12.5
Years of long-term benefits (more) 40
Discount rate on future economic benefits (more) 4%
Multiplier for resource sharing within households (more) 2
Totals
Total children receiving a 0.6% long-term income increase ~65,000
Moral weight on a 0.6% increase in adult income for a child benefitting from SMC 0.23
Total units of value from income increases per year ~15,000
Total units of value from deaths averted ~81,000
% of total modeled benefits from long-term income increases 16%

The full reasoning for our estimates of the long-term impacts of malaria on income is in our separate report on insecticide-treated nets. A short summary of the evidence we use and how we apply it to SMC is below.

What is the evidence that malaria reduces income?

Our analysis of income effects relies on two studies of historical malaria eradication campaigns: Bleakley 2010159 and Cutler et. al. 2010.160 These studies are summarized in detail in our separate report on insecticide-treated nets. In summary:

  • Both studies analyze the impact of malaria eradication campaigns (Cutler et. al. 2010 in India in the 1950s161 and Bleakley 2010 in the US in the 1920s and Colombia, Brazil, and Mexico in the 1950s).162
  • Both find that the malaria eradication campaigns were successful at reducing malaria rates. People who benefitted from reduced exposure also saw faster income gains in adulthood relative to their peers in non-malarious areas.163 This suggests that averting malaria in childhood can increase income in later life.
  • Bleakley and Cutler are both retrospective, non-randomized studies that we believe are at elevated risk of publication bias. We are generally hesitant to rely on this kind of evidence in our analysis.164

Our overall estimate from combining these studies’ results is that averting one case of malaria for a child aged 0 to 14 raises their later life income by about 2.2%, on average.165 We then apply a replicability adjustment of -70% to this figure to account for our uncertainties about the evidence base (details in footnote).166 This results in an overall estimate of ~0.6% increased income per malaria case averted.167 Our calculations are available in this spreadsheet.

Our approach

We apply this estimate to the expected reduction in malaria cases from SMC in each country in our analysis. Our analysis uses estimates of malaria incidence, the number of children we think will receive SMC per $1m, and the impact of SMC on malaria (details in footnote).168

We then apply various adjustments to transform our estimates into the total benefit over the course of an individual's (and their household’s) lifetime:

  • We roughly guess that, on average, it takes 12.5 years for the impact of SMC on income to materialize for children who benefitted from SMC (details in footnote).169
  • We use an annual discount rate of 4%. A discount rate represents the difference between how much we value increases in economic consumption now and projected economic benefits in the future. For more details on the analysis underlying this rate, see here.
  • We assume that each beneficiary will see benefits every year for 40 years of their life. This represents a rough guess about the average length of an individual’s working life.170
  • The benefits of increased income via averted malaria are ultimately valuable for how much they increase a person's ability to consume goods and services. Increasing a household’s income may increase the consumption among multiple members of the household (for example, if a household has one income earner whose income increases by 10%, each member of the household may be able to consume 10% more than before). We make some rough assumptions about the likely number of household members over time and the number of wage earners in each household. We estimate that this effect doubles the overall benefit of the income gains achieved via SMC campaigns (a "household multiplier" of 2). See this spreadsheet for our calculations.171

Shortcomings and uncertainties

While we think it is plausible that averting malaria would lead to increased income in later life, we are fairly uncertain about our analysis. In particular, the historical evidence we rely on has a number of shortcomings (e.g., possible publication bias), and we’re unsure how far it generalizes to contemporary SMC programs. We discuss our uncertainties about the long-term impact of malaria on income in more detail in our report on insecticide-treated nets.

4.5 Additional benefits and downsides

Summary

Our cost-effectiveness analysis includes a number of additional benefits and downward adjustments that we have opted not to explicitly model.172 Instead, we incorporate them as rough percentage best guesses.

As of December 2023, we sum these adjustments together to increase our estimate of the impact of SMC by 19% (Burkina Faso) to 34% (Nigeria).173 See the table below for a summary.

What we are estimating Value
Reduced malaria morbidity 9%
Short-term anemia effects 9%
Investment of income increases 5%
Treatment costs averted from prevention 20%
Adjustments for subnational malaria rates 3%
Rebound effects / decreased immunity development -4%
Development of drug resistance -4%
Marginal funding goes to lower priority areas 0% to -15%174
Serious side effects -1%
Failure to ingest first day’s doses of SMC -4%
Total (adjustment for additional benefits and offsetting impacts) 19% to 34%

We are particularly uncertain about these adjustments (details on our method in footnote)175 and they should be thought of as rough best guesses. See this spreadsheet for our full calculations.

Additional benefits

As of December 2023, we include the following additional benefits in our analysis:176

  • Reduced malaria morbidity (+9%): SMC results in averted morbidity by reducing malaria cases. This includes averted morbidity from cerebral malaria, which can involve significant neurological and cognitive impairments, as well as epilepsy.177 Our current adjustment for averted morbidity is 9%.178 In 2023, we conducted some rough internal (unpublished) analysis suggesting that we may be overestimating this adjustment (and the adjustment below for anemia) and that malaria morbidity is a lower share of benefits than we had previously assumed. We are planning to investigate this question in more detail in the future, but haven’t prioritized this yet.
  • Short-term anemia effects (+9%). Malaria infection is a risk factor for anemia, and there is evidence that malaria control interventions reduce anemia.179 We account for this effect in our analysis with an adjustment of 9%.180
  • Investment of income increases (+5%). We think that SMC might increase people’s incomes in later life (see above). If this is correct, some recipients might invest a portion of their increased income. We include a 5% adjustment to account for this. Including this benefit is consistent with our approach for other programs that we think lead to income increases, although the specific adjustment used varies case-by-case.181
  • Treatment costs averted from prevention (+20%). By reducing childhood mortality and morbidity, SMC may also avert costs that would have been incurred to seek and receive treatment for disease. These savings include the direct costs of treating malaria (incurred by households or the medical system), as well as indirect costs (e.g., caregivers taking time off work to care for sick children). We account for these savings with a 20% upward adjustment.182 This adjustment is consistent across all GiveWell’s top charities focused on improving child health.183 We chose to use a consistent figure because our model for estimating the value of costs of illness averted was very similar across these interventions (all around 20%),184 and we thought that explicitly modeling this benefit stream across countries would not be worth the added complexity. For more detail on our reasoning, see this summary.
  • Subnational adjustments (+3%). Our analysis uses national-level data on malaria burden for most countries.185 We expect that, in general, SMC campaigns will be targeted toward higher burden areas than average. To account for this, we include a 3% upward adjustment.

Offsetting impacts

  • Rebound effects (-4%). Some researchers have argued that another malaria program, insecticide-treated nets, may reduce opportunities for children to develop immunity and make them more susceptible to malaria over the long term (by averting cases in the short term).186 This could also apply to SMC. Three of the studies in the meta-analysis we rely on to estimate the impact of SMC (Meremikwu et al. 2012) investigated this question and found no evidence of a rebound effect in the high-transmission season after the SMC programs ended.187 However, we have not seen any evidence about the long-term risk of rebound effects. We include a small (-4%) adjustment in our analysis to account for the risk of rebound effects.
  • Development of drug resistance (-4%). SMC may increase the development of resistance to SP and AQ, the antimalarial drugs used in SMC. To understand more about this, we spoke with Professor Sir Brian Greenwood (Professor of Clinical Tropical Medicine at London School of Hygiene & Tropical Medicine) in 2017 about timelines for possible development of drug resistance. He told us:
    • It is very likely that, if SP and AQ are used on a wide scale for SMC, resistance will develop eventually.188
    • Timelines are uncertain, but he believes based on historical experience that resistance to SP/AQ is unlikely to develop in the next 5 to 10 years.189
    • Some resistance to SP has emerged in East Africa,190 but AQ has relatively little history of resistance.191
    • If resistance to SP/AQ did develop, dihydroartemisinin-piperarquine (DHAPQ) may be a viable alternative, although we have not carefully reviewed the evidence for DHAPQ.192

    We revisited this question in 2020. Our analysis was based on a study of genetic resistance markers conducted before and after the first two seasons of mass SMC administration during ACCESS-SMC (in 2016 and 2018).193 The study found low prevalence of resistance to SP and AQ overall. However, the prevalence of drug resistance markers associated with SP grew from 0.4% to 0.7% during the study among children under five, and 0.2% to 1% among people aged 10 to 30 years.194 AQ resistance did not increase.195 At the time, we interpreted these results to be evidence of increasing resistance, but at a low enough level that it would be accounted for by our existing -4% adjustment.

    We have not revisited our analysis on drug resistance in the Sahel since 2020 (we have since done additional analysis of drug resistance outside the Sahel, details in footnote).196 This is a significant uncertainty in our analysis and a possible way in which we’re overestimating the impact of SMC, as resistance is likely to have developed further since that time. We expect to update our analysis of resistance using more recent data in the future. We are also planning to spend time thinking about how we can monitor future resistance trends more effectively (e.g., by funding monitoring studies).

  • Marginal funding goes to lower priority areas (0 to -15%): If there is not enough funding available to fully reach the SMC-eligible population, our guess is that some countries will make prioritization decisions to exclude some lower burden areas. In effect, providing additional funding for SMC allows these lower burden areas to be covered. This would imply we could be overestimating cost-effectiveness by using malaria burden data at the national level. We account for this with a downward adjustment (0 - 15%, depending on the location). Our estimates are based on a best guess about how each country's government would allocate SMC if we do not provide additional funding and what the impact of this on cost-effectiveness would be. To determine this, we consider:
    • What prioritization the national government has done in the past (if any),
    • The variation in malaria prevalence across the country, and
    • The proportion of SMC funding in the country provided by GiveWell (as a proxy for scope to prioritize areas in the absence of GiveWell funding).

    On this basis, we use a subjective -15% adjustment for Burkina Faso and -5% for Togo (details in footnote).197 We use a value of 0% for Nigeria, where Malaria Consortium supports SMC programs at the state level198 (and we therefore use state-level estimates of malaria burden, meaning an additional adjustment would be double-counting).

  • Serious side effects (-1%): We include a -1% adjustment to account for the impact of serious side effects from SMC. Severe adverse effects associated with SP/AQ appear to be very rare. In 2019, we lightly reviewed (in this document) two large studies that tracked adverse events in recent SMC distributions in sub-Saharan Africa. found very low rates of serious adverse events and no deaths that were clearly attributable to SMC (although we did not deeply vet the study methodologies).199 We did a rough calculation using very pessimistic assumptions that suggests severe adverse events from SMC are very small in comparison to the magnitude of the benefits.200

    We revisited this question in 2023 based on a deeper (unpublished) review of a substudy in one of the earlier papers we reviewed. We would expect this substudy to have a stronger method for detecting adverse events than the other studies we previously reviewed (details in footnote).201 Using data only from this study, we calculated a worst-case estimate (using very pessimistic assumptions) implying that deaths caused by SMC are ~1.4% of the mortality reduction from SMC.202 We think this suggests that the true rate of mortality from adverse events, if it occurs, is probably between 0 and 3 children per 100,000 treated, whereas SMC averts 213 deaths per 100,000 children treated. Because this was in line with our current adjustment, we opted not to make a quantitative adjustment.

  • Failure to ingest first day’s doses of SMC (-4%). Our estimates of the number of children reached with SMC per dollar donated rely on our analysis of the cost to deliver each cycle of SMC, which in turn relies on coverage surveys conducted by Malaria Consortium.203 Our understanding is that Malaria Consortium's coverage surveys are designed to assess whether children received the first day of SMC, but that they are not designed to measure the degree to which the medicines are fully ingested by program participants.204 We have heard of cases of children vomiting drugs back up, and these children may be recorded as receiving the first day’s doses even if they have not experienced the benefits. We account for the risk that this inflates our estimates of the number of children reached with SMC with a -4% adjustment.

Factors we have excluded

  • Possible interaction with insecticide treated nets (ITNs). It is possible that, in areas where children are protected by both insecticide treated nets and SMC, the impact of SMC could be lower (e.g., if deaths that would have been averted by SMC are already being averted by nets). Two of the RCTs in the Cochrane review promoted the use of nets to both the intervention and control group in addition to SMC, and found similar proportional reductions in malaria to studies that did not promote nets.205 Our analysis attempts to account for the impact of ITNs on malaria burden in each location where we fund SMC (more above). However, in light of this evidence we don’t make an additional downward adjustment for SMC leading to a lower proportional reduction in malaria than observed in the RCTs.
  • Drug quality and dosage. Malaria Consortium told us that its policy is to only procure products from suppliers that meet WHO guidelines for pre-qualification quality assurance standards,206 and that these products are approved and quality-assured by the governments of countries where Malaria Consortium works.207 We have not yet asked Malaria Consortium for the details of these processes. If there were issues with drug quality or dosage, it could reduce the effectiveness of the intervention and lead to more rapid development of drug resistance.

4.6 Grantee-level adjustments

Summary

Our cost-effectiveness analyses also include adjustments relating to the specific organizations we recommend rather than the intervention itself. These reflect aspects of the organization’s delivery of a program that might have an impact on cost-effectiveness. Rather than explicitly model these, we apply them as rough percentage best guesses.

As of December 2023, our estimate is that these factors reduce the cost-effectiveness of Malaria Consortium’s SMC program by 8% on net.208 A summary of our calculations is below:

What we are estimating Value
Risk of wastage 0%
Quality of monitoring and evaluation -2%
Confidence in funds being used for intended purpose -6%
Total (adjustment for grantee-level factors) -8%

Adjustments in detail

Risk of wastage

The main adjustment that we have considered here is "double treatment" (i.e., a scenario where SMC is delivered to children who don't benefit or minimally benefit because they have recently received it elsewhere).209 We do not currently account for this risk in our analysis (i.e., we use a 0% adjustment). Our reasoning for this assumption is:

  • Our understanding is that SMC is only delivered through mass campaigns like those supported by Malaria Consortium. SMC is a relatively recent intervention, and it has been rapidly scaled up in the countries supported by GiveWell.210 We therefore believe that we have a relatively comprehensive understanding of the SMC landscape, and we believe it is unlikely that there are other forms of SMC delivery we are not aware of.
  • The World Health Organization recommends that national governments prohibit the private sale of SP or AQ in areas implementing SMC. Our understanding is that this is because of the risk of toxicity from repeat doses, and to minimize the risk of increasing resistance to SP and AQ.211 This strengthens our confidence that children are very unlikely to receive SMC through other sources.
Quality of monitoring and evaluation

We use a downward adjustment of -2% to account for the quality of Malaria Consortium’s monitoring and evaluation.212 This adjustment reflects our best guess at the extent to which methodological aspects of Malaria Consortium’s monitoring could inflate its estimates of the number of children reached with SMC.213

Malaria Consortium conducts monitoring surveys after the campaigns it supports to estimate the proportion of children who received SMC. We believe these surveys are high-quality overall and provide evidence that a high proportion of target children were reached. Our main reservations are that the surveys rely on caregiver self-reports about whether their children received SMC (which we would guess somewhat inflate the true proportion of children reached) and that there are discrepancies between different types of surveys (those conducted after each cycle of SMC and those conducted after each year’s SMC season) in some locations.214 Because our level of confidence in Malaria Consortium’s monitoring is high overall, we use a relatively low (-2%) value for this adjustment.

Confidence in funds being used for intended purpose

We use a downward adjustment of -6% to account for the risk that GiveWell-directed funding for Malaria Consortium’s SMC program will not be used for its intended purpose. This figure incorporates two factors:

  • Change of priorities (-1%). This refers to a scenario where Malaria Consortium uses GiveWell-directed funding for other work that Malaria Consortium sees as related to the program, but that we don't find valuable. This adjustment is small because Malaria Consortium used a small amount of funding provided by GiveWell in 2017 in unexpected ways (primarily supplemental monitoring), but has not done so since then.215
  • Within-organization fungibility (-5%). This refers to a scenario where GiveWell-directed funding for Malaria Consortium’s SMC program frees up some Malaria Consortium funding and fundraising efforts that would have gone to SMC for other uses. Our best guess is that these other uses will be less cost-effective overall than SMC. This adjustment is relatively small because we think Malaria Consortium has limited unrestricted funding relative to the size of its SMC program. However, we know of one example where Malaria Consortium had the option of applying for funding for SMC and chose to apply for a different program because SMC is relatively well funded.216

5. How does the program affect other actors’ spending?

5.1 Summary

Part of our cost-effectiveness analysis involves asking what impact funding a program has on other actors’ spending. Additional funding for SMC may lead other organizations or governments to spend more (we refer to this as "leveraging" funding, or “crowding in”) or less (we refer to this as "funging," from “fungibility,” or “crowding out”) on SMC than they otherwise would.

We include a “leverage and funging” adjustment in our cost-effectiveness analysis to account for this. As of December 2023, our leverage and funging adjustment is -8% to -51%, varying by location (-43% in Burkina Faso). A summary of calculations is below, using Burkina Faso as an example:

What we are estimating Value
Grant size (arbitrary value) $1,000,000
Value of Malaria Consortium spending on SMC in Burkina Faso (more) 0.105
Total units of value generated by Malaria Consortium spending ~105,000
Costs covered by other actors per $1m spent by Malaria Consortium (more)
Burkina Faso government ~$118,000
Other philanthropic actors (UNICEF) ~$9,000
What would happen if we did not fund the program (more)
The Global Fund and/or PMI costs would replace philanthropic costs 50% probability
Nobody would replace fund the program 50% probability
Estimated value of activities that would be funded by other actors instead of SMC
Activities funded by the Burkina Faso government (more) 0.005217
Activities funded by the Global Fund and/or PMI (more) 0.015218
Change in value under different scenarios
Nobody would fund the program (leverage) ~ -300
The Global Fund and/or PMI would fund the program (funging) ~ -45,000
Final adjustments
Adjustment for diverting other actors’ spending into SMC (leverage) -0.3%
Adjustment for diverting other actors’ away from SMC (funging) -43%
Total: Adjustment for leverage and funging -43%

We think of these adjustments as particularly uncertain inputs in our analysis. Our biggest areas of uncertainty are:

  • Our analysis of how other actors will behave if we do not make a grant are necessarily subjective guesses, as we can only speculate about their future priorities and decisions.
  • Our adjustments rely on estimates of the value of other programs that other actors might fund instead of SMC. We outline the estimates and our reasoning for them below, but these are rough estimates based on limited information. We have also invested considerably less time in producing these estimates than we have in our main cost-effectiveness analyses, and they should be thought of as rougher guesses.

5.2 Leverage

Leverage refers to extra funding for SMC causing other actors to spend more on SMC than they otherwise would. We already account for the main part of this effect in our cost per SMC cycle calculations (where we exclude in-kind government resources that we think our funding leverages). This means the benefit of these resources is already baked into our main calculations. To account for these resources, we make a small negative adjustment here reflecting that they will not be used for other programs. We estimate that this effect reduces our initial cost-effectiveness estimate by a negligible amount (0.3%219 in Burkina Faso).

We use the following reasoning:

  • We think that each $1m spent by Malaria Consortium causes the Burkina Faso government to incur ~$118,000 of in-kind costs (e.g., staff time).220
  • We exclude these costs from the cost side of the equation when we estimate the number of children reached by SMC and deaths averted in the main part of our cost-effectiveness analysis (more above). But the benefit of these resources is already incorporated in our initial impact calculations. We account for them by deducting the value of the programs we think they would have been spent on otherwise (see this blog post for more on why we use this approach).
  • Our best guess is that if these resources were not used for SMC, the Burkina Faso government would have used them for something ~5% as cost-effective.221 In total, diverting these funds away from other programs “costs” about 600 units of value (calculation in footnote, more on units of value here).222
  • We think there’s approximately a 50% chance that nobody would replace GiveWell’s SMC funding in Burkina Faso in GiveWell’s absence. This implies that the funding causes the government to divert its resources away from other programs into SMC. For more context, see below.
  • Our final leverage adjustment involves multiplying ~600 units of value by 50%, and deducting the total (~300) from our estimate of the total value generated by Malaria Consortium’s spending.223 This equates to a -0.3% adjustment.224
  • Intuitively, the reasons this adjustment is small are (i) that the Burkina Faso government’s resources constitute a relatively small proportion of the total cost of the program, and (ii) that we think the other activities the Burkina Faso government might use its resources on are considerably less cost-effective than SMC. This means the value lost from diverting these funds away from other activities is relatively small.

5.3 Funging

Funging refers to extra funding for SMC causing other actors to spend less on SMC than they otherwise would. We estimate that this effect reduces cost-effectiveness by about 43% in Burkina Faso.225

Our reasoning is:

  • At the time we made our most recent grant, we thought that there was a 50% chance that other malaria funders (specifically the Global Fund and/or PMI) would replace GiveWell’s SMC funding in Burkina Faso in GiveWell’s absence226 (10% - 65% in other locations).227 This was a rough guess, based on our understanding of the SMC funding landscape in Burkina Faso. See below for more details on our reasoning.
  • If other actors were to replace GiveWell’s spending, this implies that the true impact of the spending is to free up their resources for other activities.228 Our best guess is that the activities other funders might fund instead are about 1/7 as cost-effective as SMC in Burkina Faso,229 and therefore we would lose ~90,000 units of value, relative to our initial estimate of the total value generated by Malaria Consortium’s spending (calculation in footnote).230
  • We think that there’s a 50% chance that other actors would replace GiveWell’s SMC spending. For our final funging adjustment, we therefore multiply by 50% and deduct the total (~45,000 units of value) from our estimate of the total value generated by Malaria Consortium’s spending. This equates to a -43% adjustment.231
  • Intuitively, the reason this adjustment is relatively large is that we think there’s a substantial (50%) chance that the real impact of GiveWell’s funding for SMC in Burkina Faso is simply to free up other actors’ funding for other activities that we think are probably less cost-effective.

5.4 Breakdown of our analysis

Percentage of costs paid by different actors

We estimate the proportion of program costs paid for by different actors. These estimates are based on our cost analysis (discussed above). Overall, we think that for each $1m spent by Malaria Consortium on SMC:232

  • The Burkina Faso government incurs about $118,000 in in-kind costs (e.g., staff time).
  • Other philanthropic actors (e.g., in Burkina Faso, UNICEF has contributed a small amount of funding) incur approximately $9,000 in costs for delivering SMC.

What would happen if GiveWell did not fund SMC?

We make guesses about what would happen to other actors’ spending on SMC if GiveWell did not provide Malaria Consortium funding to deliver the campaign. For Burkina Faso, we most recently estimated these probabilities as part of a January 2023 grant. We guessed that if GiveWell had not provided Malaria Consortium funding for the campaign:233

  • There was a 50% chance that the Global Fund and/or PMI would replace Malaria Consortium’s costs (Scenario 2).
  • There was a 50% chance that nobody would replace Malaria Consortium’s costs (Scenario 4).

These guesses were based on our analysis of the malaria funding landscape (globally and in Burkina Faso). Some of the points we considered (more details on our grant page) were:

  • SMC has been fully funded in Burkina Faso since at least 2019.234 (This includes funding from GiveWell donors.) All else equal, we thought this increased the likelihood that other actors would plug funding gaps in GiveWell’s absence.
  • We thought it was plausible that, if GiveWell reduced its support for SMC, the Global Fund would reallocate funding to cover some or all of the gap. We knew of one previous occasion where the Global Fund increased its support for Burkina Faso's SMC program using reallocated funding.235
  • The Global Fund, which has historically been a major funder of SMC in Burkina Faso,236 appeared to have plateauing funding for malaria at the time we made the grant (details in footnote).237 By contrast, we expected funding needs for malaria control to increase over the years covered by the grant (e.g., because of population growth, inflation, and the introduction of new interventions). We thought this made it less likely that the Global Fund would cover the full funding gap for SMC than we had previously guessed.238

How valuable is SMC, compared to the activities that other actors might fund instead?

Our leverage and funging adjustments estimate the impact of shifting funding to or from SMC campaigns in Burkina Faso, relative to other activities that governments and other funders might fund instead. This means that we need to estimate the value of these other activities.

To do this, we compare activities in terms of "units of value," an arbitrary unit GiveWell uses to compare the moral value of different types of outcomes (e.g., increased income vs reduced deaths). We benchmark the value of each benefit to a value of 1, which we define as the value of doubling someone’s consumption for one year. See this document for more details on how we think about comparing value across different interventions.

Our analysis of leverage and funging for SMC in Burkina Faso involves three specific estimates:

  • Malaria Consortium spending on SMC in Burkina Faso (before leverage and funging): 0.105 units of value per $ (more)
  • Activities that domestic governments might fund instead of SMC: 0.005 units of value per $ (more)
  • Activities that the Global Fund or PMI might support instead of SMC: 0.015 units of value per $ (more)

Our estimates of the other activities that might be funded instead of SMC are based on considerably less work than our main cost-effectiveness analysis, and so we think of them as rougher guesses.

Malaria Consortium spending on SMC in Burkina Faso

We estimate that each dollar spent by Malaria Consortium on SMC in Burkina Faso generates 0.105 units of value.239 This figure is the final output generated by our cost-effectiveness analysis, after factoring in all adjustments except leverage and funging.

Activities that domestic governments might fund instead of SMC

We estimate that each dollar of in-kind resources that the Burkina Faso government contributes to SMC would generate 0.005 units of value if used for other activities.240 This is around 1/20 as valuable as our estimate of spending on SMC.241

In summary, our approach is:

  • We estimate that 80% of these resources would be used on health programs, 10% would be used for education programs and 10% would be used on social security programs.242 This is a very rough best guess, for which we have not done any in-depth research.
  • Next, we estimate the value of spending in each category (details on our approach in footnote).243 This results in the following estimates:
    • Health: 0.0056 units of value per $ spent
    • Education: 0.0028 units of value per $ spent
    • Social security: 0.0026 units of value per $ spent
  • Finally, we take a weighted average of the value of each type of spending, with our guesses about how the spending would be allocated (80% health / 10% education / 10% social security) as the weights. This results in an overall estimate of 0.005 units of value per dollar spent.244
Activities that the Global Fund might fund instead of SMC

We estimate that each dollar that the Global Fund spends on other activities rather than SMC in Burkina Faso generates 0.015 units of value.245 This is around 1/7 as valuable as our estimate of spending on SMC in Burkina Faso.246

Our calculations for this estimate are available in this spreadsheet. In summary, our approach is:

  • The Global Fund funds HIV, tuberculosis (TB), and malaria programs. We assume that all funding that the Global Fund could use for SMC in Burkina Faso but actually uses for other activities is spent on programs combating these diseases.
  • We estimate the value of spending on each type of program (details on our approach in footnote).247 This results in estimates (here) that the Global Fund’s spending on HIV programs is 1.1x as cost-effective as direct cash transfers (GiveWell’s benchmark for comparing the value of different programs), and its spending on TB programs is 3.2x as cost-effective.
    • For malaria, we conduct a more detailed analysis, breaking down Global Fund spending by seven types of malaria program that it may fund instead of SMC in Burkina Faso (varying in cost-effectiveness from 0.5 to 13.6x, depending on the program).248
  • Finally, we estimate the overall value of the Global Fund’s spending on other activities. We calculate a weighted average of each of the programs discussed above in proportion to how likely we think it is to spend reallocated funding on that activity. These estimates are based on 2017-2019 data on how funding reallocated within the Global Fund portfolio in that period was used.249 We do not have permission to publish this data, and so our full calculations are available on a separate spreadsheet (only available to GiveWell staff). This weighted average produces our overall estimate of 0.0154 units of value per $ spent (equivalent to 4.6x as cost-effective as direct cash transfers).

6. Additional perspectives beyond our cost-effectiveness model

6.1 Summary

In theory, our cost-effectiveness analysis intends to capture the total impact of a program per dollar spent. But we recognize that our cost-effectiveness calculations are not able to capture every factor that could make a program more or less impactful. Focusing only on our cost-effectiveness model may mean we’re missing things that are difficult to quantify.

As a result, we think it’s helpful to look at other perspectives and types of evidence that may not be captured in our bottom line cost-effectiveness number. Some additional questions we have considered are:

  • Is there evidence that large-scale SMC campaigns lead to reductions in malaria?
  • Should GiveWell be promoting SMC in other ways?
  • Do experts and practitioners see SMC as a good program?
  • Is GiveWell’s funding of SMC crowding out other funders over the long term?
  • Is it intuitively plausible that SMC is cost-effective?
  • How does our cost-effectiveness model compare to others?
  • Does SMC have unintended negative consequences?
  • How accurate was our analysis of SMC in hindsight?
  • Will SMC remain impactful in the future?

We see asking these questions as a type of “cluster thinking,” or considering a program from multiple perspectives. The more additional perspectives we’ve considered, in general the more confident we are in our funding recommendations.

Overall, the additional perspectives we’ve considered give us some additional confidence that SMC is a good investment. In particular, we find it encouraging that:

  • SMC has scaled up rapidly over time (indicating widespread support from the global malaria community. (More)
  • SMC appears to have relatively few unintended negative consequences. (More)
  • There is evidence that SMC and the RTS,S vaccine delivered together are more effective than either intervention alone. We see this as important because we expect the vaccine rollout to play a major role in the malaria landscape in the future. (More)

But we have spent considerably less time and effort engaging with these questions than we have on our main cost-effectiveness model. Major gaps in our analysis include:

  • Over the long term, GiveWell’s funding for SMC could create an expectation of future funding. This could mean we crowd out funding from other funders (i.e., displace funding that would have gone to SMC into investments that may be less cost-effective). We have found it challenging to find evidence either way on whether this is happening. (More)
  • We have seen two other cost-effectiveness studies that find higher cost-per-death-averted estimates than ours, although we haven’t investigated them in detail. (More)
  • We haven’t conducted much “backwards-looking” analysis to understand if our projections about SMC were correct in hindsight (e.g., would other funders fill funding gaps in our absence?). (More)

6.2 Is there evidence that large-scale SMC programs lead to reductions in malaria?

Why is this important? We generally rely on randomized controlled trials (RCTs) for evidence about the causal impact of a program. But all else equal, it strengthens our confidence if there’s evidence that large-scale programs show similar impact to studies conducted in experimental conditions. There might be a number of reasons why these programs might not be delivered to the same quality at scale.

How we’ve accounted for this

Our bottom line

A large observational study of the ACCESS-SMC program found a smaller impact of SMC (~25% to ~55% reduction in outpatient malaria cases) than would be suggested by RCTs. We see this study as a small downward update to our best guess about the impact of SMC. We do not currently make an adjustment to account for this, but we are actively considering this question and may update our analysis in the future.

In more detail

We have seen several studies evaluating the impact of large-scale SMC programs. These include an observational study of the ACCESS-SMC project (a large multi-organization program in seven countries in which Malaria Consortium led delivery of SMC in three: Chad, Burkina Faso, and Nigeria)250 (ACCESS-SMC Partnership 2020), case-control studies designed to measure the efficacy of SMC from five countries in the ACCESS-SMC program,251 and an assessment of the impact of SMC on malaria rates in Burkina Faso and Chad from 2013 to 2018 and Nigeria from 2017 to 2018 based on administrative data on malaria cases in areas with and without SMC.252

We have paid most attention to investigating the analysis of malaria incidence and mortality from the ACCESS-SMC study, since we’d expect this to closely resemble the impact of SMC on malaria in the programs GiveWell currently funds. The study estimated the impact of SMC by comparing trends in case rates and deaths among children under five (who receive SMC) with people over age five (who don’t).253 In two countries, data on case rates and deaths were collected from national health databases (DHIS2), and in the remaining five countries data on confirmed outpatient cases were collected from outpatient clinics using parasitological confirmation.254

The headline findings from the analysis are that SMC was associated with reductions of ~25% to ~55% in outpatient malaria cases, depending on the country.255 In the two countries estimating mortality, SMC was associated with a 42% reduction in malaria hospital deaths in Burkina Faso and 57% in The Gambia.256

These results are considerably lower than the ~75% reduction in malaria cases and mortality we assume in our analysis (more above). We see this as concerning, since it indicates that large-scale SMC programs might not achieve the results implied by studies conducted under tightly controlled experimental conditions. We think there are some reasons to put weight on this study:

  • It’s a large, multi-country study that aims to look at the impact of SMC under real-world conditions. Several of the SMC programs evaluated were supported directly by Malaria Consortium, so its results are especially relevant to our analysis.
  • The results point in a reasonably consistent direction across countries.
  • It uses a strategy (over-5s as a control group) to separate out the impact of SMC from other factors that might have affected malaria cases that we think is reasonably strong (although less strong than a well-controlled experiment).

But there are also good reasons to downweight the study in our analysis:

  • Some children over age 5 (i.e., those in the control group) also received SMC (53% of 6- to 7-year-olds surveyed in 2015 received SMC at least once, 62% in 2016).257 We would expect this to bias the observed findings downward.
  • The study measured SMC cases and deaths in hospitals and outpatient clinics. Our understanding is that most malaria deaths in malaria-endemic countries happen outside hospitals.258
  • Reported coverage of SMC was lower (76% in 2015 and 74% in 2016)259 both than the earlier SMC RCTs (~90%)260 and the coverage rates reported by Malaria Consortium in programs since then (~80% to ~90%, varying by country, more above). This means we would expect the impact to be lower.
  • It seems possible that the methods used to collect data on malaria cases and deaths (clinic records and national health databases) are unreliable, although we haven’t investigated this in detail, and this could bias the results in either direction.
  • There may be other factors affecting malaria burden among children under-5 and people over-5 that could have affected the analysis.

Overall, we see this study as a small downward update to our best guess about the impact of SMC. We do not currently make an adjustment to account for this, but we are actively considering this question and may update our analysis in the future.

We have seen several other observational studies of large-scale SMC programs. None of these significantly change our perspective on SMC:

  • Case-control studies designed to measure the efficacy of SMC from five countries in the ACCESS-SMC program.261 These found that SMC provided a high level of protection, similar to the levels estimated in the Cochrane RCTs (details in footnotes).262 This analysis was a separate part of the ACCESS-SMC paper discussed above. We’re not sure what’s driving the difference in results.
  • Kirakoya-Samadoulougou et al. 2022, which estimates trends in malaria in SMC and non-SMC health districts in Burkina Faso. The paper finds SMC was associated with a reduction in malaria incidence of 73% (very similar to the experimental estimates).263 We have only skimmed this paper.
  • Issiaka et al. 2020, which uses a similar methodology to Kirakoya-Samadoulougou et al. and finds that SMC was associated with a substantial impact on all-cause mortality (risk ratio of 0.44), considerably higher than we would expect.264 We have only skimmed this paper.
  • An assessment of administrative data conducted by Malaria Consortium, in Burkina Faso and Chad from 2013 to 2018 and Nigeria from 2017 to 2018. This assessment found no evidence of impact. Malaria Consortium attributes the evidence of no impact in the Burkina Faso, Chad, and Nigeria assessments in part to variable and low-quality data.265 While we believe that there is a risk that data quality limitations will be emphasized more where non-positive trends are found than where positive trends are found, independently assessing the quality of HMIS data would be a large project for us, and we have not prioritized that work.

Finally, we conducted an internal analysis in 2023 attempting to identify the impact of SMC through changes in malaria seasonality patterns. If SMC was effective at reducing malaria in real-world contexts, we’d expect data on malaria cases to become less seasonal among young children over time. We conducted a shallow investigation, looking for monthly or weekly malaria case estimates for countries in the Sahel, with data from before and after SMC was implemented.

Overall, none of the evidence we found allowed us to estimate the impact of SMC on malaria seasonality in children under 5 with confidence. We found two papers indicating that malaria case rates in Burkina Faso have not become less seasonal over time, and slightly increased in the 2013-2020 period.266 We would see this as concerning, but our best guess is that this data is not particularly informative because:

  • We think it’s likely that the observed increases reflect improvements in recording and health systems or changes in the effectiveness of vector control tools such as insecticide-treated nets, rather than a genuine change in malaria cases.267
  • The case rate data is for people of all ages, not just young children.268

6.3 Should GiveWell be promoting SMC in other ways?

Why is this important? To date, GiveWell has only funded SMC through mass campaigns. It might be that we’re missing other ways of funding SMC that would be more effective.

How we’ve accounted for this

  • We haven’t yet considered other ways we could fund SMC except mass campaigns in detail. Our understanding is that SMC is currently exclusively delivered via mass campaigns in the countries where GiveWell funds SMC. We have heard of some pilots of SMC being integrated into routine health services in countries where GiveWell does not fund SMC (examples in footnote),269 but have not investigated these in detail.
  • We are actively considering whether it would be possible for GiveWell to promote coverage of SMC through advocacy to increase global funding for SMC. As of December 2023, that work is just beginning and we’re not sure whether it will change our decision-making.

6.4 Do experts and practitioners see SMC as a good program?

Why is this important? We’re more confident in programs which have wide support from experts in malaria-endemic countries and the global health community more widely.

How we’ve accounted for this

  • We have not investigated this question systematically, e.g., by analyzing or conducting polls of global health experts or national malaria program representatives.
  • Through our work on SMC, we have built up an impression over time that SMC commands widespread support in the global health community. Reasons to think this include:
    • Our impression is that national malaria programs in countries delivering SMC consider it to be one of the highest-priority interventions to fund with the donor funding they receive. This is based on multiple conversations with national malaria programs over time.
    • WHO has recommended SMC for deployment since 2012.270 In 2022, WHO updated its recommendation to allow for more flexibility in where and how SMC should be delivered (e.g., number of cycles, level of resistance and age of targeted children), more above. This change followed a number of cases where national malaria programs in malaria-endemic countries began to expand their use of SMC beyond the original WHO recommendation.271 We interpret this as evidence that national malaria programs are generally enthusiastic about SMC and eager to see it expanded.
    • SMC has scaled up rapidly since it was first recommended by the WHO. An estimated 2.6 million children were reached with SMC in 2014, rising to approximately 49 million in 2022.272 While GiveWell funding accounts for part of this scale-up, we see this as indicating widespread support from other actors too.
  • In 2023, we received feedback on our SMC research from Dr. André Tchouatieu, Director, Access & Product Management at Medicines for Malaria Venture.273 Dr. Tchouatieu told us that he believed SMC was a very cost-effective intervention, but raised concerns about equity (since children above age 5 and children in areas that don’t deliver SMC receive considerably less protection than children receiving SMC).274 Perennial malaria chemoprevention (PMC) is recommended for children up to age 2 in some non-SMC areas, but is not as widely implemented as SMC.275 While we are also concerned about this issue, we do not see it as a negative reflection on SMC (but a reflection of insufficient funding for malaria control in general).

While we are relatively confident that SMC commands overall support in the global health community, we may be missing downsides by not investigating this question more systematically. We may prioritize engaging with critics of SMC in the future.

6.5 Is GiveWell’s funding of SMC crowding out other funders over the long term?

Why is this important? Our adjustments for the impact of GiveWell funding on other actors’ spending mainly consider the probability that another actor would fund SMC if we didn’t in the short term. They do not fully account for the long-term impact. For example, it might be that GiveWell’s funding could create an expectation of future funding, and displace other actors’ funding that would have gone to SMC into other programs.

How we’ve accounted for this

  • We have seen some evidence that GiveWell’s SMC funding has crowded out funding from other sources. For example, we have seen cases of national malaria programs subtracting the amount they expect to receive for SMC from Malaria Consortium from their Global Fund funding applications. This evidence is discussed here.
  • We’re unsure how widespread this is, and how much it should affect our grantmaking. We attempt to incorporate this into our analysis of GiveWell’s impact on other actors’ spending here (+20% increase in the likelihood that other funders would replace GiveWell’s spending in our absence), but this is a very uncertain guess.
  • We aim to mitigate this effect in our conversations with national malaria programs and other funders by emphasizing that, to the extent possible, our goal is for the funding we direct to SMC to add to the total pool of funding available, rather than to replace funding that would otherwise have been in that pool.

6.6 Is it intuitively plausible that SMC is cost-effective?

Why is this important? In general, we are more confident in our funding recommendations if we can easily explain the intuitive case for why a program is cost-effective. Thinking through the intuition behind a program can help us reveal where we might have made mistakes.

How we’ve accounted for this

  • Overall, we think that the intuitive case for SMC (set out in detail in the report summary) is very strong.
  • In their most simple terms, GiveWell’s cost-effective analyses can be boiled down to the following components: (a) burden (does the intervention tackle a big problem), (b) impact (does the intervention have a significant impact on the problem), (c) cost (is the intervention cheap to deliver), and (d) impact of our funding the program on intervention uptake (does our funding lead to more people receiving the intervention). We believe that SMC looks like an effective investment on all these criteria, since:
    • Malaria is a major cause of child deaths in sub Saharan Africa (more above)
    • SMC is targeted to a group at high risk of malaria (young children) and during the period where most malaria transmission occurs. We expect this would make it more cost-effective, all else equal.
    • SMC is effective at reducing deaths due to malaria, based on randomized controlled trials, opinions from experts, and a clear mechanism (more above)
    • SMC is reasonably cheap to deliver (roughly $5 to $6 per child per year, more above).
    • We expect children would not be able to access SMC through sources other than campaigns (more), although we think there’s a substantial chance these campaigns would have been funded in our absence (more).

6.7 How does our cost-effectiveness model compare to others?

Why is this important? GiveWell’s analysis is only one attempt to model the cost-effectiveness of different global health programs. We would find it concerning if other analyses found that SMC is significantly less cost-effective than we currently estimate.

How we’ve accounted for this

  • We know about two SMC cost-effectiveness studies we could use to check our analysis: Diawara et al. 2021 (a study of SMC in one district in Mali)276 and Nonvignon et al. 2016 (investigating one region in Ghana).277 There may be other similar studies we haven’t seen.
  • To date (December 2023), we have only skimmed both studies. Based on an initial look, Diawara et al. finds a higher cost-per-death-averted estimate than GiveWell (~$14,500)278 , and Nonvignon et al. finds an estimate either in the same range as GiveWell’s analysis (~$3,300) or higher (~$9,900), depending on whether only the SMC provider’s costs or “societal” costs (including donated resources and caregiver time and expenses) are taken into account.279
  • We find it moderately concerning that these papers seem to find overall higher cost-per-death-averted estimates than GiveWell’s model, but we haven’t checked our cost-effectiveness analysis against either tool in detail, and there may be a number of definitional and methodological reasons why our estimates differ. We may use these to check our analysis in the future.

6.8 Does SMC have unintended negative consequences?

Why is this important? When trying to estimate the total impact of SMC, we need to offset the benefits with any negative impacts.

How we’ve accounted for this

  • We explicitly account for two negative impacts from distributing SMC elsewhere in our analysis:
    • The possibility that it delays the development of immunity to malaria and makes children more susceptible in the long term (-4% adjustment).
    • Serious side effects (-1% adjustment).
  • One negative impact that we might not be fully accounting for is the development of future resistance to the drugs used in SMC. We currently account for drug resistance with a -4% adjustment (in the Sahel), but this is only intended to capture the past development of drug resistance since the original studies of SMC were conducted (as these could affect the effectiveness of SMC in the present). (More) We do not account for the risk that widespread delivery of SMC will contribute to growing resistance in the future. We are planning to continue to monitor data on resistance in the future. One factor that mitigates our concern about increasing resistance is that the malaria community (including WHO) continues to support SMC.
  • Overall, we think the negative impacts of SMC are relatively small, in comparison to the benefits, and in the same range as other programs we see as very effective. We’re reasonably confident about this conclusion, although we may not have considered all the possible downsides and we have not investigated some of these issues recently.

6.9 How accurate was our analysis of SMC in hindsight?

Why is this important? Our cost-effectiveness analyses are “forward-looking.” They aim to project the future impact of funding a program at the time we make a grant decision. We would be concerned if backwards checks of our analysis found significant problems we hadn’t considered at the time we made our decisions, or were overly optimistic in general.

How we’ve accounted for this

  • In general, GiveWell has paid less attention to backwards checks to understand how accurate our predictions were, and more attention to making our forward projections as accurate as possible. This is a weakness in our approach and something we aim to improve in the future.
  • As part of our research on SMC, we have conducted some backward-looking analysis to inform our cost-effectiveness analysis (details in footnote).280
  • We have not yet conducted a backwards-looking analysis of our SMC cost-effectiveness analysis as a whole, to understand how accurate our predictions were. We are hoping to prioritize this in the future. We are also hoping to conduct a backwards-looking analysis of what happened in cases where GiveWell declined to fund an SMC program (i.e., did other funders fill the resulting gap). We are planning to use this to triangulate our adjustment for the impact of our funding on other actors’ spending (details in footnote).281

6.10 Will SMC remain impactful in the future?

Why is this important? Our cost-effectiveness analysis aims to model the impact of our funding at the time we make a grant, but this may reflect program delivery several years in the future. We could be making mistakes if we don’t anticipate likely changes in the malaria landscape.

How we’ve accounted for this

  • We have generally paid attention to making our analysis accurate at the time we make a grant. We want to improve this aspect of our work in the future and think in more detail about possible future changes, although this is inherently uncertain.
  • One particular uncertainty in our analysis is the impact of malaria vaccines. As of December 2023, WHO has recommended two malaria vaccines.282 We expect a significant share of future malaria funding to go toward the vaccine rollout.
  • We have seen one paper, published in August 2023, that found SMC co-delivered with the RTS,S vaccine was more effective at averting malaria than either intervention alone (protective efficacy of ~58% compared to SMC and ~59% compared to RTS,S).283 It’s possible that this implies delivering both interventions together would be more cost-effective than delivering either alone (since there could be cost savings from using a single delivery platform for multiple interventions). Conversely, it could be that the vaccine rollout makes SMC less cost-effective overall (because the overall malaria burden will be lower). We are planning to investigate this question in more detail in the future.
  • We’re also unsure about future developments in resistance to the drugs used in SMC. Our current analysis assumes that resistance is only a small negative adjustment to SMC programs delivered in the Sahel today, but it’s possible that this will change as resistance grows over time (more above). We are planning to do a more detailed update on resistance in the future using more recent data.

7. Previous SMC grants

Sources

Document Source
ACCESS-SMC Partnership 2020 Source
ACCESS-SMC, Annex III - ACCESS-SMC Evaluation (preliminary reports) Source
ACCESS-SMC, Evaluation of Seasonal Malaria Chemoprevention Source
ACCESS-SMC, Fact sheet, 2016 Source
ACCESS-SMC, Progress update, Pharmacovigilance: Monitoring SMC drug safety, 2017 Source
ACCESS-SMC, The cost of SMC in the Sahel region of Africa, 2017 Source
Bationo et al. 2023 Source
Bhatt et al. 2015 Source
Bleakley 2010 Source
Bojang et al. 2011 Source
Brown and Lilford 2006 Source
Cairns et al. 2012 Source
Cairns et al. 2021 Source
Cissé et al. 2006 Source
Cissé et al. 2016 Source
Cochrane, "About us," 2023 Source (archive)
Conteh et al. 2010 Source
Cutler et al. 2010 Source
Diawara et al. 2021 Source
Dicko et al. 2008 Source
Dicko et al. 2011 Source
Dicko et al. 2023 Source
Gilmartin et al. 2021 Source
GiveWell blog, "Sequence thinking vs. cluster thinking," 2014 Source
GiveWell, "Helen Keller International — Vitamin A Supplementation (January 2023)" Source
GiveWell, "Malaria Consortium — SMC Renewal in Nigeria, Burkina Faso, Chad, and Togo (January 2023)" Source
GiveWell, "Why we don’t use subnational malaria mortality estimates in our cost-effectiveness models," 2018 Source
GiveWell, Analysis of Malaria Consortium's cost per SMC cycle administered, October 2022 Source
GiveWell, Analysis of subnational mortality rates for SMC Source
GiveWell, Analysis of the counterfactual value of other actors' spending, 2023 Source
GiveWell, CEA for GiveDirectly's unconditional cash transfers Source
GiveWell, Cost of illness adverted adjustment write-up, 2023 Source
GiveWell, Cost of illness averted model for malaria treatment Source
GiveWell, Counterfactual value of government funds Source
GiveWell, Discount rate, 2020 Source
GiveWell, Estimating multiplier for benefits experienced by other household members Source
GiveWell, GiveDirectly – November 2020 version Source
GiveWell, GiveWell's 2020 moral weights Source
GiveWell, Global Fund counterfactual spend Source
GiveWell, Global Fund malaria funding, 2024-2026 Source
GiveWell, Likelihood of crowding out GFATM/PMI, December 2022 Source
GiveWell, Likelihood of crowding out GFATM/PMI, October 2020 Source
GiveWell, Malaria income effect size, 2023 Source
GiveWell, Moral weights and discount rate Source
GiveWell, Morbidity from malaria Source
GiveWell, Pryce et al. mortality calculations supplemental spreadsheet, 2023 Source
GiveWell, Rainfall in Nigeria by state and month, 2022 Source
GiveWell, Serious adverse events caused by seasonal malaria chemoprevention Source
GiveWell, SMC cost-effectiveness analysis 2.0, 2023 Source
GiveWell, SMC coverage in RCTs, 2023 Source
GiveWell, SMC fourth to fifth round calculations, January 2022 Source
GiveWell, SMC indirect effects in Malaria Consortium programs relative to Cisse et al. 2016 Source
GiveWell, Supplemental intervention-level adjustments for CEA, 2023 Source
GiveWell, Update of Meremikwu et al. 2012 meta-analysis (forest plot) Source
GiveWell, US allocations to Gavi and Maternal and Child Health, 2022 Source
GiveWell's non-verbatim summary of a conversation with Professor Sir Brian Greenwood, January 4, 2017 Source
Global Fund, "Global Fund Board Hails Record-Breaking Seventh Replenishment Final Outcome of US$15.7 Billion," 2022 Source (archive)
Global Fund, "Seventh replenishment: Fight for what counts" Source (archive)
Global Fund, "Unfunded quality demand" Source (archive)
Idro et al. 2010 Source
Imperial College London, "Malaria modelling: Tools and data," 2023 Source (archive)
Institute for Health Metrics and Evaluation, GBD 2019 methods appendix Source
Institute for Health Metrics and Evaluation, GBD 2019 summaries, "Malaria—Level 3 Cause" Source (archive)
Institute for Health Metrics and Evaluation, GBD results, prevalence of malaria, Burkina Faso, Nigeria, Senegal, Togo, 2019 Source
Institute for Health Metrics and Evaluation, GBD results, prevalence of vitamin A deficiency, Niger and Nigeria, 2019 Source
IRD Global, home page Source (archive)
Issiaka et al. 2020 Source
Jamison et al. 2006 Source
Kirakoya-Samadoulougou et al. 2022 Source
Konaté et al. 2011 Source
Kweku et al. 2008 Source
Lengeler 2004 Source
Malaria Atlas Project, Explorer Source (archive)
Malaria Consortium, 2018 SMC coverage report Source
Malaria Consortium, 2019 impact report Source
Malaria Consortium, 2019 SMC coverage report Source
Malaria Consortium, 2020 SMC philanthropy report Source
Malaria Consortium, 2021 SMC coverage report Source
Malaria Consortium, 2021 SMC philanthropy report Source
Malaria Consortium, 2022 SMC philanthropy report Source
Malaria Consortium, Net-target project report: Nigeria, 2020 (redacted) Source
Malaria Consortium, SMC programme start-up guide: Nigeria, 2015 Source
Matangila et al. 2015 Source
Meremikwu et al. 2012 Source
NDiaye et al. 2016 Source
Nonvignon et al. 2016 Source
Patouillard et al. 2011 Source
Pitt et al. 2017 Source
Pryce, Richardson, and Lengeler 2018 Source
Rethink Priorities, Malaria Deaths: A Comparison of WHO and IHME Estimates, 2023 Unpublished
Ross et al. 2011 Source
Sangaré et al. 2022 Source
Sesay et al. 2011 Source
SMC Alliance, Celebrating 10 years of SMC, 2023 Source
Tagbor et al. 2016 Source
Tchouatieu, “GiveWell SMC report: Feedback from an independent review,” 2023 Unpublished
Uyoga et al. 2019 Source
WebMD, "Malaria: Causes, symptoms, and treatment," 2023 Source (archive)
White 2018 Source
WHO, "Malaria," 2023 Source (archive)
WHO, "Updated WHO recommendations for malaria chemoprevention among children and pregnant women," 2022 Source (archive)
WHO, "WHO recommends R21/Matrix-M vaccine for malaria prevention," 2023 Source (archive)
WHO, Policy Recommendation: Seasonal Malaria Chemoprevention, 2012 Source
WHO, SMC field guide, 2013 Source
WHO, SMC field guide, 2nd ed., 2023 Source (archive)
WHO, World malaria report, 2022 Source (archive)
Zongo et al. 2015 Source
  • 1

    This range includes Burkina Faso, Togo, and various states in Nigeria. See this row in our cost-effectiveness analysis. Note, we exclude Chad, Mozambique, Uganda, and DRC from this range for the reasons discussed in this section.

  • 2

    "Malaria is a life-threatening disease spread to humans by some types of mosquitoes. It is mostly found in tropical countries. It is preventable and curable…The infection is caused by a parasite and does not spread from person to person…There are 5 Plasmodium parasite species that cause malaria in humans and 2 of these species – P. falciparum and P. vivax – pose the greatest threat. P. falciparum is the deadliest malaria parasite and the most prevalent on the African continent. P. vivax is the dominant malaria parasite in most countries outside of sub-Saharan Africa." World Health Organization (WHO), "Malaria," 2023

  • 3

    WHO, "Malaria," 2023.

  • 4

    WHO, "Malaria," 2023.

  • 5

    "Infants, children under 5 years, pregnant women, travellers and people with HIV or AIDS are at higher risk." WHO, "Malaria," 2023

  • 6

    “Globally, malaria deaths reduced steadily over the period 2000–2019, from 897 000 in 2000 to 577 000 in 2015 and to 568 000 in 2019. In 2020, malaria deaths increased by 10% compared with 2019, to an estimated 625 000. Estimated deaths declined slightly in 2021 to 619 000.” WHO, World malaria report, 2022, xxi.

  • 7

    WHO, "Malaria," 2023.

  • 8

    WHO, "Malaria," 2023.

  • 9

    “SMC is the intermittent administration of a curative dose of antimalarial medicine during the malaria season to asymptomatic children, regardless of whether the child is infected with the malaria parasite – that is, asymptomatic children are not tested for malaria before SMC administration. The objective of SMC is to establish antimalarial drug concentrations in the blood that clear existing infections and prevent new ones during the period of greatest malaria risk. SMC is recommended in areas of highly seasonal P. falciparum malaria transmission… The priority target areas for SMC implementation are those where:

    • P. falciparum malaria transmission is highly seasonal and the majority (>60%) of clinical malaria cases occur within 4 consecutive months – where data on malaria from the health management information system are unreliable, rainfall data could be used as a proxy for seasonality in incidence (at least 60% of annual rainfall in 4 consecutive months); and
    • the clinical attack rate of malaria (without SMC) is at least 0.1 episodes per child during the transmission season in the target group.”

    World Health Organization, Seasonal malaria chemoprevention with sulfadoxine–pyrimethamine plus amodiaquine in children: a field guide, 2nd edition, 2023, pg. 2.

  • 10

    Children with confirmed malaria are not given SMC. In locations where rapid diagnostic tests (RDTs) and antimalarial treatments are available, children with danger signs for malaria (including a fever) are tested, and given SMC if the test is negative. In locations where tests and antimalarials are not available, all children with malaria danger signs (including a fever) are not given SMC and are referred to a health facility for appropriate care. Malaria Consortium, comments on a draft of GiveWell's Malaria Consortium report, November 9, 2023.

    “SMC should not be given to:

    • a child with an acute febrile illness or a severe illness – these children need to be referred to the nearest health facility for appropriate care (or tested and, if positive for malaria, treated on the spot with an antimalarial in countries where rapid diagnostic tests and ACT are available in the community as part of the SMC campaign;
    • a child taking co-trimoxazole (e.g. HIV-positive child receiving co-trimoxazole prophylaxis);
    • a child who has received a dose of either SP or AQ during the previous 4 weeks; or
    • a child who is allergic to either SP or AQ."

    World Health Organization, Seasonal malaria chemoprevention with sulfadoxine–pyrimethamine plus amodiaquine in children: a field guide, 2nd edition, 2023, pg. 3

  • 11

    “WHO recommends that medicines used as first- or second-line malaria treatment in a country not be used for chemoprevention in that country. The combination of SP+AQ is currently recommended for SMC for the following reasons.

    • In the clinical trials that provided the evidence base for WHO recommendations, SP+AQ conferred greater protection than other medicine combinations.
    • There are no indications that the chemoprevention efficacy of SP+AQ is diminishing in Africa.
    • The SP+AQ regimen is well tolerated and relatively inexpensive.
    • The SP+AQ regimen confers protection for 28 days.”

    WHO, SMC field guide, 2nd ed., 2023, pg. 3.

  • 12

    See World Health Organization, World malaria report 2023, p. 64, table 7.1.

  • 13

    See WHO, Policy Recommendation: Seasonal Malaria Chemoprevention, 2012.

  • 14

    “Seasonal Malaria Chemoprevention (SMC) is recommended in areas of highly seasonal malaria transmission across the Sahel sub-region. A complete treatment course of amodiaquine plus sulfadoxine-pyrimethamine (AQ+SP) should be given to children aged between 3 and 59 months at monthly intervals, beginning at the start of the transmission season, to a maximum of four doses during the malaria transmission season (provided both drugs retain sufficient antimalarial efficacy). . . . Target areas for implementation are areas where:

    • Malaria transmission and the majority of clinical malaria cases occur during a short period of about four months [Note 3: Areas where on average more than 60% of clinical malaria cases occur within a maximum of 4 months; these areas are characterized by more than 60% of the average annual rainfall falling within 3 months.].
    • the clinical attack rate of malaria is greater than 0.1 attack per transmission season in the target age group, and
    • AQ+SP remains efficacious (>90% efficacy).” WHO, Policy Recommendation: Seasonal Malaria Chemoprevention, 2012, pg. 2.

  • 15

    "In 2021, the malaria programs in Burkina Faso, Nigeria, and Uganda introduced five monthly SMC cycles in areas where the transmission season is slightly longer." Malaria Consortium, 2021 SMC philanthropy report, pg. 17.

  • 16
    • Malaria Consortium began piloting SMC in Nampula, Mozambique, where it targeted approximately 70,000 children in two districts in 2020-21, and Karamoja, Uganda, where it targeted approximately 90,000 children in two districts in 2020-21.
    • In 2021, Malaria Consortium supported the second half of the 2020-21 SMC round: "Phase 1 of the study involved delivering four monthly SMC cycles to around 70,000 children in two districts of Nampula province (Figure 7) during the 2020/21 transmission season… Activities completed in 2020, including procurement of SPAQ, training of SMC implementers, and distribution of the first two monthly SMC cycles were discussed in Malaria Consortium’s 2020 SMC philanthropy report. Cycles 3 and 4 of the 2020/21 round were implemented as scheduled in January and February 2021. Both districts finished SMC distribution on 11th February." Malaria Consortium, 2021 SMC philanthropy report, pg. 24.
    • "In 2020, the National Malaria Control Division (NMCD) approached Malaria Consortium with a request to support an SMC implementation study in Karamoja (Figure 13) to investigate the feasibility, acceptability, and impact of SMC. The study employs a similar two-phase design as the study Malaria Consortium is conducting in Mozambique and is described in more detail in the research section below. The first project phase involved SMC delivery to around 90,000 children in two districts. Taking into account local malaria transmission patterns, five SMC cycles were implemented between May and September 2021. This was the first time SMC was implemented in the country." Malaria Consortium, 2021 SMC philanthropy report, pg. 37.

  • 17

    “The original recommendation restricted SMC use to the Sahel subregion of Africa; SMC could not be recommended, at the time, in areas outside the Sahel with highly seasonal malaria transmission, such as in southern Africa, due to high levels of resistance to the medicines (SP and AQ) in those areas. The updated recommendation recognizes that countries in other parts of Africa with highly seasonal variation in malaria burden could also benefit from SMC, and that the availability of new medicines could make it a viable intervention in these areas.
    The original recommendation stated that a maximum of 4 monthly doses of SMC should be given during the malaria transmission season. The updated guidance states that SMC should be given during peak malaria transmission season, without defining the specific number of monthly cycles… While the original recommendation restricted SMC use to children less than 6 years of age, the new recommendation recommends this intervention for children at high risk of severe malaria, which may extend to older children in some locations.” WHO, "Updated WHO recommendations for malaria chemoprevention among children and pregnant women," 2022.

  • 18

    “Children in age groups at high risk of severe malaria are eligible. Malaria programmes should use local data to determine which age groups are at high risk of severe malaria. In most countries with intense seasonal malaria transmission, these are children below 5 years of age (1).” WHO, SMC field guide, 2nd ed., 2023, pg. 2.

  • 19

    “The original recommendation restricted SMC use to the Sahel subregion of Africa; SMC could not be recommended, at the time, in areas outside the Sahel with highly seasonal malaria transmission, such as in southern Africa, due to high levels of resistance to the medicines (SP and AQ) in those areas. The updated recommendation recognizes that countries in other parts of Africa with highly seasonal variation in malaria burden could also benefit from SMC, and that the availability of new medicines could make it a viable intervention in these areas.
    The original recommendation stated that a maximum of 4 monthly doses of SMC should be given during the malaria transmission season. The updated guidance states that SMC should be given during peak malaria transmission season, without defining the specific number of monthly cycles… While the original recommendation restricted SMC use to children less than 6 years of age, the new recommendation recommends this intervention for children at high risk of severe malaria, which may extend to older children in some locations.” WHO, "Updated WHO recommendations for malaria chemoprevention among children and pregnant women," 2022.

  • 20

    These include pilots of SMC in Mozambique, Uganda, and South Sudan. See Malaria Consortium, 2022 SMC philanthropy report 2022, pgs. 41-46 (Mozambique), 53-55 (South Sudan), and 58-63 (Uganda) for more details. GiveWell has also co-funded (with the Bill and Melinda Gates Foundation) RCTs of SMC in Mozambique and Uganda, which we plan to use as evidence to inform our analysis of SMC in regions outside the Sahel. The Mozambique RCT is discussed on this grant page.

  • 21

    “SMC is primarily delivered door-to-door by trained community distributors. A full course of SP plus AQ (SPAQ) is given over three consecutive days. On the day of the community distributor’s visit to a household, one tablet of SP and one tablet of AQ are dispersed in water and administered under the supervision of a community distributor. This is called directly observed treatment (DOT). The remaining two doses of AQ are given to the caregiver to disperse and administer once daily over the next two days.” Malaria Consortium, SMC Philanthropy report 2020, pg. 5.

  • 22

    Malaria Consortium notes that it only provides funding for SMC campaigns in locations supported by philanthropic funding, including GiveWell. In locations where other funding sources are used, the funding is provided by donors, not Malaria Consortium. Malaria Consortium, comments on a draft of GiveWell’s Malaria Consortium page, November 9, 2023.

  • 23

    “SMC campaigns are implemented under the leadership of national malaria programs and through countries’ existing health system structures. Consequently, Malaria Consortium’s role in supporting SMC varies from country to country.” Malaria Consortium, SMC philanthropy report 2020, pg. 6.
    See Malaria Consortium, 2020 SMC philanthropy report, pgs. 7-8, for a detailed breakdown of Malaria Consortium’s role in the program. For details on the proportion of funding Malaria Consortium contributes to each SMC program, see our analysis of the cost per SMC cycle delivered.

  • 24

    Burkina Faso, Chad, Mozambique, Togo, Uganda, four provinces in DRC, and nine states in Nigeria.

  • 25

    The locations in our analysis not discussed in detail in this report are Chad, Mozambique, Uganda, and DRC.
    Chad: As of 2023, we estimate that the Malaria Consortium SMC program in Chad is below our funding bar. In January 2023, we made a grant for the SMC program in Chad, intended as exit funding. We largely do not focus on Chad in this report because we expect to discontinue funding the SMC program in Chad in the future. See this grant page for further details.
    Mozambique, Uganda and DRC: Historically, SMC has only been delivered at large scale in the Sahel region of west Africa, and Malaria Consortium has only recently scaled up its SMC program in Mozambique and Uganda (it has not yet begun delivering SMC in DRC). See this grant page for details on this scale-up in Mozambique. At the time of writing (December 2023), we have not fully completed or published our analysis of SMC outside the Sahel (which may differ substantially in effectiveness because of more widespread drug resistance in East Africa). Our analysis in this report therefore focuses on analyzing the impact of SMC in the Sahel.

  • 26

    We used the following process:

    • Three GiveWell staff members familiar with our research on SMC gave confidence intervals for the parameters in the Simple CEA sheet of our cost-effectiveness analysis. The intervals were for Malaria Consortium’s program in Burkina Faso (the example program we discuss in this report). These intervals address the following question: “For each parameter, provide an upper and lower bound for the range of values that you think has a 50% probability of containing the true value (i.e. a 25% probability the true value is lower than the range, and a 25% probability the true value is higher).”
      • Where we felt that it was difficult to form an intuitive judgment about a parameter (e.g., because it comprised several separate parameters), we gave intervals for each individual parameter.
    • A fourth GiveWell staff member reviewed the intervals given by the first three staff members and decided upon a final interval for each parameter, using their subjective judgment.
    • We applied the intervals used for Burkina Faso to other countries as follows:
      • We first calculated both the relative differences and absolute differences between our best guesses and our 25th and 75th percentile values for each parameter. For example, if our best guess value for a parameter was 50% with 25th and 75th percentile values of 25% and 75%, then the relative differences would be -50% and +50%, and the absolute differences would be -25 percentage points and +25 percentage points.
      • We assigned confidence intervals to each country that matched the widths of our confidence intervals for Burkina Faso, putting some weight on the relative difference between the confidence interval bounds and the best guess and some weight on the absolute difference. The higher our best guess value for a parameter in a country was compared to our best guess for Burkina Faso, the more weight we put on the absolute difference between the confidence interval bounds and our best guess. The purpose of this method was to avoid assigning extremely wide confidence intervals to parameters in countries where our best guess was much higher than our best guess for Burkina Faso.
    • Finally, we used monte carlo simulations to transform the confidence intervals for individual parameters into an aggregated confidence interval for each parameter. These are the intervals that appear in the summary of this page, and the Burkina Faso columns in the Sensitivity Analysis sheet of our cost-effectiveness analysis.

  • 27

    $1 million is an arbitrary amount that we use to quantify the benefits of the program in the rest of our analysis. See this row in our cost-effectiveness analysis.

  • 28

    This range includes Burkina Faso, Togo, and various states in Nigeria. See this row in our cost-effectiveness analysis. We exclude Chad, Mozambique, Uganda, and DRC from this range for the reasons discussed in this section.

  • 29

    See our calculations here.

  • 30

    Although we have data available on Malaria Conosortium’s programs between 2015 and 2021, we use data from 2018 to 2021 only in our estimates. This is because in general we would expect a program to incur some start-up costs, and for the costs of a more mature program to be a better guide to future costs. Note that for Malaria Consortium’s SMC program specifically there does not appear to have been a trend towards lower costs of SMC over time, with the exception of high costs in 2015. See this row of our analysis for a year-by-year comparison.

  • 31

    See the "Target population" rows in this sheet of our analysis.

  • 32

    “Several limitations may be noted, however. First, target populations used for calculation of administrative coverage were estimated on the basis of official population figures, which were often based on outdated national census data and adjusted for projected population growth. Estimates of population sizes may not adequately reflect population movements due to migration or internal displacement." Malaria Consortium, 2021 SMC coverage report, p. 56.

  • 33

    "In previous years, only one coverage survey was conducted at the end of the annual SMC round to measure coverage achieved by Malaria Consortium’s SMC program. This meant that administrative data had to be relied on to gauge coverage while the campaign was ongoing. As gathering and compiling administrative data is a time-consuming process, this limited the program’s capacity to take evidence-based corrective action and make adaptations to the intervention during the campaign. For this reason, Malaria Consortium decided to implement coverage surveys following each of the four SMC cycles in each of the three implementation countries." Malaria Consortium, 2018 SMC coverage report, pg. 12.
    "End-of-cycle (EoC) surveys employing the lot quality assurance sampling methodology following cycles 1 to 3 to enable implementing teams to identify areas of low coverage and rapidly take corrective actions to improve SMC delivery in subsequent cycles." Malaria Consortium, 2019 SMC coverage report, pg. 4.
    "Comprehensive end-of-round (EoR) surveys following the completion of the SMC round (that is, after cycle 4) to assess SMC performance across all four monthly cycles." Malaria Consortium, 2019 SMC coverage report, pg. 4.
    Note: there were no nationally-representative post-round surveys in Burkina Faso in 2021, so we estimate the number of cycles delivered in that year’s campaign using coverage rates from 2018 to 2020 and the 2021 campaign target population. We would expect this method to be slightly less accurate than using 2021 coverage surveys. See this sheet for our calculations.

  • 34

    Our estimates of the number of cycles delivered based on post-cycle surveys and the number of cycles delivered based on post-round surveys differ by 2% in 2017, 13% for 2018, 24% for 2019, 9% for 2020, and 6% for 2021. In all years except 2017, the post-cycle surveys produce a higher estimate for the number of cycles delivered. See this sheet of our analysis ("Difference" rows) for our calculations.
    We would expect the post-round surveys to be more accurate because:

    • We have previously heard from Malaria Consortium that some researchers for post-cycle surveys were recruited from among state-level health authority staff. We would generally expect that fully independent surveys will be more accurate.
    • Post-cycle surveys have generally found extremely high coverage (close to 100% in some cases).
    • Malaria Consortium has told us it puts more weight on post-round results, as post-round surveys are explicitly designed to achieve representative samples, whereas post-cycle surveys are designed to assess whether each program area unit met a target coverage level.

    See this section of our report on Malaria Consortium for further details.

  • 35

    See this section of our analysis.

  • 36

    Our estimates of the number of cycles delivered based on post-cycle surveys and the number of cycles delivered based on post-round surveys differ by 2% in 2017, 13% for 2018, 24% for 2019, 9% for 2020, and 6% for 2021. In all years except 2017, the post-cycle surveys produce a higher estimate for the number of cycles delivered. See this sheet of our analysis ("Difference" rows) for our calculations.

  • 37

    See this section of the report for further details.

  • 38

    Specifically, we weight the % of children who received 1 / 2 / 3 / 4 / 5 cycles of SMC by the number of cycles received, and multiply this by the target population. See this section for an example.

  • 39

    We refer to this figure as “person-months of coverage,” i.e., the number of children protected by SMC for one month.

  • 40

    See this sheet in our analysis, “Person-months of coverage” rows.

  • 41

    See Malaria Consortium, SMC programme start-up guide: Nigeria, 2015, Figure 1, pg. 8. "DOT" stands for "Directly Observed Treatment."
    "Day 1 SP and AQ should be administered by the drug distributor as DOT.
    If the child vomits or spits out the drugs within 30 minutes, a second dose should be given." Table 1, Pg. 8, Malaria Consortium, 2018 SMC coverage report.

  • 42

    See this row of our analysis for the adjustment by country.
    We estimate this adjustment by:

    • Using adherence estimates (the number of children who received a full three-day course of SMC as a proportion of those who received the first day, according to their caregivers) from surveys conducted by Malaria Consortium in 2020 and 2021. These result in adherence estimates of 94 - 100%, varying by survey. See this sheet (for the 2020 surveys) and this sheet (for the 2021 surveys).
    • We apply a 90% adjustment (i.e., -10%) to these figures to account for social desirability bias (the tendency of survey participants to over-report "good" behaviors). We would expect this to inflate caregivers’ self-reported adherence to some extent, although the specific estimate we use is a rough best guess.
    • Finally, we estimate the overall impact of non-adherence on the number of children protected by SMC by applying a 50% discount to our estimates of non-adherence. This reflects our best guess that receiving the first dose only of SMC provides approximately 50% of the protection against malaria compared to receiving all three doses. This is a very rough guess, and further research could lead us to update this estimate up or down. However, because our non-adherence adjustment is small overall, we have not prioritized further work on this.

    This produces an overall adherence adjustment of 93-94% (i.e., -7% to -6%, varying by location). See this sheet for our calculations. This adjustment means that our overall estimate of the number of SMC cycles delivered should be interpreted as the adjusted number of cycles delivered, accounting for some cycles providing a lower level of protection because children did not swallow all three days’ doses.

  • 43

    We break down this spending by country and year in the following sheets:

  • 44

    See this row in our analysis.

  • 45

    See this row in our analysis. We break down this spending by country and year in the following sheets:

    Our calculations on other funders’ contributions are based on information shared by Malaria Consortium. For Nigeria, we sum all the contributions reported by Malaria Consortium to estimate the total. For UNICEF’s contribution to the SMC Burkina Faso in 2021 (where it provided SMC implementation costs in two districts), we do not know the precise costs. Instead, we estimate these costs based on (i) the proportion of the target population in those districts and (ii) Malaria Consortium’s implementation costs (excluding research, external relations, and management costs). See this row for our calculation.

  • 46

    “SMC in the three eligible regions is supported by Malaria Consortium’s philanthropic funding through co-funding arrangements with the Global Fund and UNICEF (Figure 14).”
    See also Table 14, p. 56, in Malaria Consortium, 2022 SMC philanthropy report.

  • 47

    See this cell in our cost-effectiveness analysis. The calculations for the $1.50 figure are available here.

  • 48

    See this row in our analysis.

  • 49

    The study found that these contributions accounted for ~12% of the costs of delivering SMC. This figure does not include research and overhead costs (which are included in our estimates of Malaria Consortium’s spending on more recent campaigns), so we adjust this estimate downward to account for these additional costs. This results in our total estimate of 10-11% of total costs (varying by country).
    See this row in our analysis (which calculates government contributions as a proportion of all costs for both 2015 and 2016) and the following sheets for more granular data:

    Note that this figure includes recurrent costs only (i.e., excluding start-up costs, which we would expect to apply only when a program is started for the first time).
    “Total program costs are divided into start-up and recurrent costs. The start-up costs of an SMC program represent those incurred on a one-time basis such as initial launch meetings, the production of videos, the development and recording of radio spots, and other activities that would not need to be repeated each year. 16 The recurrent costs refer to those incurred every year (e.g. for drugs, equipment, recurrent meetings and training). Costs incurred when a new program is started (i.e. on a one-time basis) and are not repeated every year.”
    ACCESS-SMC, The cost of SMC in the Sahel region of Africa, 2017, pg. 14.
    See the following sheets for our specific calculations by country and by year:

  • 50

    Including these costs, we estimate that each cycle delivered costs $1.48 in Burkina Faso. Excluding these costs, we estimate a cost of $1.32 summarized in the table above.

  • 51

    Our reasoning for this assumption is that, if Malaria Consortium spends more on SMC campaigns in a given country, the amount of government time dedicated to these campaigns is likely to increase in response. Our best guess is that this is not true for other NGOs’ spending on Malaria Consortium-supported campaigns. We assume that these resources would have been spent on SMC in any case regardless of the size of the Malaria Consortium’s SMC program.

  • 52

    “Several limitations may be noted, however. First, target populations used for calculation of administrative coverage were estimated on the basis of official population figures, which were often based on outdated national census data and adjusted for projected population growth. Estimates of population sizes may not adequately reflect population movements due to migration or internal displacement." Malaria Consortium, 2021 SMC coverage report, p. 56.

  • 53

    “SMC in the three eligible regions is supported by Malaria Consortium’s philanthropic funding through co-funding arrangements with the Global Fund and UNICEF (Figure 14).” See also Malaria Consortium, 2022 SMC philanthropy report, Table 14, p. 56.

  • 54

    We received information on these costs in August 2023, but have not yet incorporated it into our cost analysis.

  • 55

    See, e.g., Malaria Consortium, 2021 SMC coverage report, pg. 51.
    “Table 28 shows the proportions of ineligible children 60–119 months who received SPAQ, based on data from EoR surveys: these were 28.2 percent in Burkina Faso, 17.5 percent in Chad, a weighted average of 31.9 percent (33.3–36.1) across the seven states in Nigeria surveyed, and 9.7 percent in Togo. Administration of SPAQ to overage ineligible children varied markedly between Nigerian states, according to cycle 5 EoR data (Table 29). In these four countries, the proportion of ineligible children who received SMC in the last cycles of the 2021 SMC round was lower than in 2020, when this was 34.9 percent (31.0–39.0) in Burkina Faso, 44.4 percent (41.7–47.0) in Chad, 35.0 percent (33.3–36.1) in Nigeria, and 32.7 percent (30.4– 35.1) in Togo. It should be noted, however, that surveys in these four countries were not designed to provide a representative sample of children 60–119 months; estimates of the proportion of this group receiving day 1 SPAQ are likely to represent an overestimate as only children in this group residing in households with eligible children 3–59 months were included in the sample.”

  • 56

    A rough average estimate in these nine studies was $7.89, accounting for inflation since the studies were conducted. This compares to a total cost estimate in our analysis of roughly $5.60 to around $9, varying by location (see this row). See this section of our cost per cycle spreadsheet.
    We’re very uncertain about this comparison, since we haven’t checked details of the underlying programs or the costing methods used, and we can imagine different studies’ methodologies having a significant impact on the estimates reached.
    Our analysis included the following studies:

  • 57

    We apply a separate -2% adjustment for “quality of monitoring and evaluation” (more), but this may not account for the extent of the bias.

  • 58

    “SMC should be implemented during the peak malaria transmission period, when the incidence of malaria is highest. SMC courses should be given at 28-day intervals, beginning at the start of the transmission season and continuing for 3–5 cycles, depending on the local context.
    SMC administration should be chosen to cover the period when children are at greatest risk for malaria infection. The seasonal distribution of malaria (number of confirmed malaria cases per month) should be described by district in each country to define the ideal timing to start and end SMC. Rainfall data can also be used to estimate peak transmission periods. Depending on the seasonal patterns of malaria transmission, the timing and number of SMC cycles (3–5) may vary between countries and in different parts of the same country (see section 3.2).”
    WHO, SMC field guide, 2nd ed., 2023, p. 2.

  • 59

    “In the 11 health districts supported by Malaria Consortium where five SMC cycles were implemented, the annual round commenced on 10 June. The remaining 18 health districts implemented four SMC cycles and started the annual round on 10 July.” Malaria Consortium, 2022 SMC Philanthropy report, p. 38.

  • 60

    (30% x 5) + (70% x 4)) = 4.3. See this spreadsheet for our calculations.

  • 61

    This range includes Burkina Faso, Togo, and various states in Nigeria. See this row in our cost-effectiveness analysis. We exclude Chad, Mozambique, Uganda, and DRC from this range for the reasons discussed in this section.

  • 62

    See this section of our cost-effectiveness analysis.

  • 63

    This range includes Burkina Faso, Togo, and various states in Nigeria. See this row in our cost-effectiveness analysis. We exclude Chad, Mozambique, Uganda, and DRC from this range for the reasons discussed in this section.

  • 64

    See this row in our cost-effectiveness analysis. This range includes Burkina Faso, Togo, and various states in Nigeria. We exclude Chad, Mozambique, Uganda, and DRC from this range for the reasons discussed in this section.

  • 65

    See this row in our cost-effectiveness analysis.

  • 66

    See this row in our cost-effectiveness analysis.

  • 67

    Cochrane, formerly known as the Cochrane Collaboration, is a not-for-profit research organization which synthesizes health research in systematic reviews.
    “Cochrane is an international network with headquarters in the UK, a registered not-for-profit organization, and a member of the UK National Council for Voluntary Organizations. …There are now over 7,500 Cochrane Systematic Reviews which we publish in the Cochrane Library. We also play a key role in developing new methods in evidence synthesis.” Cochrane, "About us," 2023..

  • 68

    “Intermittent treatment, also known as ’intermittent preventive treatment’ or ’intermittent presumptive treatment’ (IPT), is an alternative strategy and is defined as “the administration of a full therapeutic course of an antimalarial or antimalarial combination to a selected, target population at specified times without determining whether or not the subject is infected.”(Greenwood 2010).”
    Our understanding is that IPTc and SMC are identical and SMC has been adopted as the intervention name by the field since Meremikwu et al. 2012 was published. “SMC, formerly known as ‘intermittent preventive treatment of malaria in children’, is defined as “intermittent administration of full treatment courses of an antimalarial medicine during the malaria season to prevent malarial illness with the objective of maintaining therapeutic antimalarial drug concentrations in the blood throughout the period of greatest malarial risk”. The SMC strategy consists of administering a maximum of four treatment courses of SP + AQ at monthly intervals to children aged 3–59 months in areas of highly seasonal malaria transmission.” WHO, SMC field guide, 2013, p. 7.

  • 69

    “IPTc prevents approximately three quarters of all clinical malaria episodes (rate ratio 0.26; 95% CI 0.17 to 0.38; 9321 participants, six trials, high quality evidence).” Meremikwu et al. 2012, abstract.

  • 70

    “Seven trials (12,589 participants), including one cluster-randomized trial, met the inclusion criteria. All were conducted in West Africa, and six of seven trials were restricted to children aged less than 5 years.” Meremikwu et al. 2012, abstract.

  • 71

    CI = Confidence interval.

  • 72

    “Overall, IPTc prevented around three-quarters of clinical malaria episodes during the intervention period (rate ratio 0.26, 95% CI 0.17 to 0.38; 9321 participants, six trials; Analysis 1.1).” Meremikwu et al. 2012, pg. 12.

  • 73

    “IPTc also prevented around three-quarters of severe malaria episodes (rate ratio 0.27, 95% CI 0.10 to 0.76; 5964 participants, two trials; Analysis 1.2).” Meremikwu et al. 2012, pg. 12.

  • 74

    “Two trials distributed and promoted the use of ITNs to both the intervention and control groups (Dicko 2011 and Konate 2011). Despite ITN use being reported as >90% in both treatment arms, IPTc had high protective efficacy against both clinical malaria (rate ratio 0.22, 95% CI 0.13 to 0.38; 5964 participants, two trials; Analysis 2.1)”. Meremikwu et al. 2012, pg. 12.

  • 75

    “The number of deaths observed in these trials was very low. Although fewer deaths were seen in the children who received IPTc the difference was not statistically significant (9533 participants, six trials; Analysis 1.4).” Meremikwu et al. 2012, pg. 12.
    “IPTc probably produces a small reduction in all-cause mortality consistent with the effect on severe malaria, but the trials were underpowered to reach statistical significance (risk ratio 0.66, 95% CI 0.31 to 1.39, moderate quality evidence).” Meremikwu et al. 2012, abstract.

  • 76

    “IPTc probably produces a small reduction in all-cause mortality consistent with the effect on severe malaria, but the trials were underpowered to reach statistical significance (risk ratio 0.66, 95% CI 0.31 to 1.39, moderate quality evidence).” Meremikwu et al. 2012, abstract.

  • 77

    “IPTc prevents approximately three quarters of all clinical malaria episodes (rate ratio 0.26; 95% CI 0.17 to 0.38; 9321 participants, six trials, high quality evidence), and a similar proportion of severe malaria episodes (rate ratio 0.27, 95% CI 0.10 to 0.76; 5964 participants, two trials, high quality evidence). These effects remain present even where insecticide treated net (ITN) usage is high (two trials, 5964 participants, high quality evidence).” Meremikwu et al. 2012, abstract.

  • 78

    “IPTc probably produces a small reduction in all-cause mortality consistent with the effect on severe malaria, but the trials were underpowered to reach statistical significance (risk ratio 0.66, 95% CI 0.31 to 1.39, moderate quality evidence).” Meremikwu et al. 2012, abstract.

  • 79

    “Relative to the AL only group, the adjusted hazard ratio (aHR) was 0.62 (95% CI: 0.41, 0.93), P = 0.020 in the SMC group, a protective efficacy of 38.5% (95% CI 7.28%, 59.2%)” Tagbor et al. 2016, pg. 229.
    “A randomised controlled trial was conducted in Kinshasa, DRC. Enrolled schoolchildren were assigned to a passive control arm (n = 212), sulfadoxine/pyrimethamine (SP) (n = 202) or SP plus piperaquine (SP/PQ) (n = 202)... After 12 months, the Hb level increased by 0.20 g/dL (95% CI -0.61 to 0.47; P = 0.168) and 0.39 g/dL (0.12-0.66; P < 0.01) in the SP and SP/PQ arms, respectively. SP treatment reduced anaemia, malaria parasitaemia and clinical malaria by 10% (0-20%; P = 0.06), 19% (2-33%; P = 0.042) and 25% (-32 to 57%; P = 0.37), respectively. The corresponding values for SP/PQ were 28% (19-37%; P < 0.001), 40% (26-52%; P < 0.001) and 58% (17-79%; P < 0.01). No deaths or severe adverse events (SAEs) were observed. SP/PQ offered substantial protection against anaemia, malaria parasitaemia and clinical malaria and showed no SAEs. SP/PQ, a combination of two long-acting non-artemisinin-based antimalarials, may be a valuable option for IPTsc in Africa.” Matangila et al. 2015, Abstract.

  • 80

    “IPTsc with SP or SP/PQ was given at baseline (November 2012), at Month 4 (March 2013) and at Month 7 (June 2013).” Matangila et al. 2015, pg. 340.

  • 81

    “262 children 6 months-10 years in Kambila, Mali were randomized to receive either IPT with SP twice at eight weeks interval or no IPT during the transmission season of 2002 and were followed up for 12 months.” Dicko et al. 2008, pg. 1.

  • 82

    This analysis also excludes Tagbor et. al. 2011, the only RCT in the Cochrane review which did not estimate the impact of SMC on malaria cases. “Tagbor 2011…
    Outcomes
    1. Parasitaemia
    2. Severe anaemia
    3. Adverse events”
    Meremikwu et al. 2012, p. 31.

  • 83

    See: GiveWell update of Meremikwu et al. 2012 meta-analysis (Forest plot)

  • 84

    “The results of clinical trials indicate that a high level of protection against uncomplicated clinical malaria is likely to be maintained for only 4 weeks after administration of each treatment course of SP+ AQ; thereafter, protection appears to decay rapidly.” WHO, SMC field guide, 2013, p. 4.

  • 85

    The trials in our adapted meta-analysis delivered an unweighted average of 3.8 cycles. We estimate that these trials had an unweighted average follow-up period of approximately ~4.1 months (although this is only a rough estimate because not all trials report precise follow-up periods).

    • Cissé et al. 2006 followed up for 13 weeks, for three months of SMC (~3.25 months follow-up).
      • “During 13 weeks of follow-up, the intervention led to an 86% (95% CI 80–90) reduction in the occurrence of clinical episodes of malaria” pg. 663.
      • “one dose of artesunate and one dose of sulfadoxine-pyrimethamine, given on three occasions during the malaria transmission season, was undertaken in 1203 Senegalese children aged 2–59 months.” pg. 659.
    • Dicko et al. 2011 delivered three cycles of SMC and followed up 6-7 weeks after the last round of SMC (~3.5 months follow-up).
      • “After screening, eligible children aged 3–59 mo were given a long-lasting insecticide-treated net (LLIN) and randomised to receive three rounds of active drugs or placebos.” pg. 1.
      • “Passive surveillance for clinical malaria started at the time of the administration of the first dose of IPTc in August 2008 and continued until the end of the malaria transmission season in November/December 2008, 6–7 wk after the last round of IPTc.” pg. 4.
    • Konaté et al. 2011 delivered three cycles of SMC and followed up six weeks after the last round of SMC (~3.5 months follow-up).
      • “Treatment courses were given in August, September, and October during the peak malaria transmission season, with 1 mo intervals between treatments.” pg. 4.
      • “A cross-sectional survey of all children enrolled in the study was carried out 6 wk after the last course of IPTc had been given (November 2008). Fever within the last 24 h was documented, and axillary temperature, weight, and height were measured. Thick and thin blood films and filter paper blood spots were prepared and Hb concentration was measured.” pg. 4.
    • Kweku et al. 2008 appears to have followed up for six months for six treatments delivered monthly.
      • “Children received study drugs every 28 days on six occasions.”
      • ”During the six months of the intervention period, the incidence of anaemia, the primary trial endpoint, was significantly lower in all IPTc groups compared to the placebo group.”
    • Sesay et al. 2011: It is unclear from the text what the precise period of surveillance was. Our best guess is surveillance covered a period of 3.5 months, for three cycles delivered at monthly intervals.
      • “SP/AQ (person months at risk, 2248)” (Table 4); “SP/AQ (n = 639)” (Table 1). 2248/639=3.5 months.
      • “monthly IPTc with a single dose of sulphadoxine/pyrimethamine (SP) plus three doses of amodiaquine (AQ) or SP and AQ placebos given by village health workers (VHWs) on three occasions during the months of September, October and November”
    • Tagbor et al. 2016: Children received five cycles of SMC. It is unclear from the text what the precise period of surveillance was. It reports results “during the SMC period” but we are uncertain whether this period extended beyond four weeks after the last treatment. We assume the follow-up period was five months.
      • “SMC or placebo was delivered on five occasions during the rainy season.” Abstract, pg. 224.
      • “The protective efficacy of SMC was lower than expected during the SMC period, at around 38%, with no significant protection over the whole study period” pg. 233.
    • Average number of cycles = (3 + 3 + 3 + 6 + 3 + 5) / 6 = ~3.8
    • Average estimated follow-up period in months = (3.25 + 3.5 + 3.5 + 6 + 3.5 + 5) / 6 = ~4.1.

  • 86

    “The figure of 75% from the 2011 systematic review underlies your impact predictions. This figure refers to the efficacy of 3 cycles over almost 4-month evaluation period. SMC gives a high degree of protection for 28 days, protection then drops off rapidly, with little effect remaining by about 6 weeks. The estimate over one month from the same trials, was 86%. Similar estimates were obtained in operational settings using case control studies (85% over 1 month and 60% over the next 2 weeks). In order to predict the impact of SMC over 4 or 5 cycles, it may be helpful to be more explicit, the current method may under-estimate the impact in that there is some benefit beyond 4 weeks, this is relevant for children who receive 4 cycles, they get perhaps 2 weeks of slightly lower protection at the end of the 4-month period, there are also children who receive say month 1,3 and 4, they get some protection during month 2, etc. This could be more explicitly calculated using existing data.” Paul Milligan, review of GiveWell SMC report, November 2023 (unpublished).

  • 87
    • "We inspected the forest plots to detect overlapping CIs, applied the Chi2 test and a P value of 0.10 was used as the cut-off value to determine statistical significance. We also estimated the I2 statistic with values of 30 to 59%, 60 to 89%, and 90 to 100% used to denote moderate, substantial and considerable levels of heterogeneity respectively." Meremikwu et al. 2012, pg. 8.
    • "Heterogeneity: Tau2 = 0.20, Chi2 = 84.28, df = 5 (P < 0.00001); I2 = 94%” Meremikwu et al. 2012, Analysis 1, pg. 40.

  • 88

    “The size of this effect varied from a 45% reduction in Mali (Dicko et al. 2008) to an 86% reduction in Senegal (Cissé et al. 2006). This variation could be explained by differences in efficacy between the antimalarial regimen used, by variation in local transmission or resistance patterns, or other factors related to the conduct of the trials. However there were insufficient trials to make meaningful conclusions from subgroup analyses exploring the effects of these factors (Analysis 2.1; Analysis 3.1).” Meremikwu et al. 2012, pg. 12.

  • 89

    Dicko et al. 2008 had (i) a longer follow-up (ii) fewer cycles of SMC, and (iii) longer time interval between cycles than the other trials.

    • Dicko et al. 2008 report reduction in incidence of malaria in the period 52 weeks after the start of the intervention. The other RCTs in the Cochrane review report incidence between 13 weeks and 6 months after the start of the intervention. “Cisse et al [12] in Senegal found a reduction of 86% in incidence of clinical malaria children less than five years of age over 13 weeks period. This higher efficacy can be explained by the shorter time interval between treatments (four weeks instead of eight weeks), the number of intermittent treatments (three instead of two) and the shorter duration of the follow up (13 instead 52 weeks) [12].” Dicko et. al. 2008, pg. 7.

    See this section of the report for details on the number of cycles and time interval for other trials included in the review. Follow-up periods:

    • Cissé et al. 2006: “During 13 weeks of follow-up, the intervention led to an 86% (95% CI 80–90) reduction in the occurrence of clinical episodes of malaria”, pg. 663.
    • Dicko et al. 2011: “Incidence rate expressed as number of episodes/child/year. Note that this is based on the 3-mo surveillance period and does not correspond to an annual rate.” pg. 9.
    • Konaté et al. 2011: “From mid-September to the end of November 2008, 150 enrolled children were randomly selected and surveyed each week to monitor the prevalence of malaria infection.” pg. 4.
    • Kweku et al. 2008: “During the six months of the intervention period, the incidence of anaemia, the primary trial endpoint, was significantly lower in all IPTc groups compared to the placebo group”, Pg. 5.
    • Sesay et al. 2011: “SP/AQ (person months at risk, 2248)” (Table 4); “SP/AQ (n = 639)” (Table 1). 2248/639=3.5 months.

    Tagbor et al. 2016 had (i) low coverage (ii) low malaria incidence prior to program delivery (iii) unreliable adherence data, and (iv) the course did not start until the high-transmission season was already in progress.

    • “The lower efficacy may partly be due to relatively low coverage of all five SMC cycles (36.4% in the SMC group), but even among children who received five cycles of SMC, protective efficacy was around 47% during the SMC period, lower than the efficacy of 70-80% seen in other studies of monthly SP-AQ [4, 6, 7], or monthly artesunate-amodiaquine”, pg. 233.
    • “Overall efficacy might also be lower because incidence is relatively low in the study area as a whole: 7 of the 13 study communities have an annual incidence of malaria below 100 cases per 1000 child-years at risk, the lower limit suggested for SMC to be cost-effective [8]. In three small communities with a high prevalence of malaria at baseline and high incidence during the transmission season (Korase, Sarpeh and Timeabu) the protective efficacy of SMC (any number of cycles) was more consistent with earlier studies: 69.7% (95% CI: 22.7%, 88.1%). Potentially supporting this is the fact that there was no benefit of SMC on anaemia, in contrast to findings from SMC studies in high transmission areas [6, 7] but agreeing with areas of lower transmission [1].” pg. 233.
    • “This could be a chance finding, because malaria incidence was low and confidence intervals for the efficacy are wide. However, if correct, it could suggest failure to complete the 3-day course of SMC (SP plus AQ on day one, AQ only on days two and three): although SP alone retains reasonably high efficacy in most of West Africa, the efficacy of SP plus AQ is higher [5]. Adherence data were collected but implausibly high values of adherence were reported, as seen elsewhere [23]”, pg. 233.
    • “Low efficacy may also be a consequence of the timing of administration of the first cycle of SMC. At the baseline survey in July, prevalence was around 26%, suggesting that the transmission season was well underway by this point. Although SP-AQ is likely to have very good curative efficacy in West Africa [24, 25], the benefit of SMC will likely be greater where children are protected prior to exposure each year. If SMC was started too late in 2012 relative to the start of the transmission season, this suggests that five cycles of SMC will be insufficient in this setting, as there was also some malaria incidence in January 2013. Seven or possibly eight cycles could cover the period between May and January, but this raises new questions of cost-effectiveness, and the safety and acceptability of a substantial additional number of SMC cycles.” pg. 233.

  • 90

    See our separate review of Malaria Consortium for details on Malaria Consortium’s SMC program.

  • 91
    • Cissé et al. 2006: "Intermittent treatment: IPTi with sulfadoxine-pyrimethamine plus artesunate given once monthly" Meremikwu et al. 2012, pg. 23.
    • Dicko et al. 2011: "Sulphadoxine Pyrimethamine (SP) SP + Amodiaquiune AQ or Placebo tablets were given during the peak malaria transmission season, with one month intervals between treatments." Meremikwu et al. 2012, pg. 25.
    • Konaté et al. 2011: "Sulphadoxine Pyrimethamine (SP) SP + Amodiaquiune AQ or Placebo tablets were given in August, September, and October during the peak malaria transmission season, with one month intervals between treatments", Meremikwu et al. 2012, pg. 27.
    • Kweku et al. 2008: "Monthly artesunate plus amodiaquine reduced the incidence of malaria by 69% (95% CI: 63%, 74%) and anaemia by 45% (95% CI: 25%,60%)", abstract.
    • Sesay et al. 2011: "During the 2008 malaria transmission season, 1,277 children under five years of age resident in villages within the rural Farafenni demographic surveillance system (DSS) in North Bank Region, The Gambia were randomized to receive monthly IPTc with a single dose of sulphadoxine/pyrimethamine (SP) plus three doses of amodiaquine (AQ) or SP and AQ placebos given by village health workers (VHWs) on three occasions during the months of September, October and November, in a double-blind trial", abstract.
    • Tagbor et al. 2016: "SMC involves monthly administration of the long-acting antimalarial drugs sulfadoxine-pyrimethamine and amodiaquine (SP-AQ) to all children under five years of age", pg. 224.

  • 92
    • Cissé et al. 2006: "Intermittent treatment: IPTi with sulfadoxine-pyrimethamine plus artesunate given once monthly" Meremikwu et al. 2012, pg. 23.
    • Dicko et al. 2011: "Sulphadoxine Pyrimethamine (SP) SP + Amodiaquiune AQ or Placebo tablets were given during the peak malaria transmission season, with one month intervals between treatments." Meremikwu et al. 2012, pg. 25.
    • Konaté et al. 2011: "Sulphadoxine Pyrimethamine (SP) SP + Amodiaquiune AQ or Placebo tablets were given in August, September, and October during the peak malaria transmission season, with one month intervals between treatments", Meremikwu et al. 2012, pg. 27.
    • Kweku et al. 2008: "Intermittent treatment: IPTc with artesunate plus amodiaquine (AS+AQ) monthly or every two months, or sulphadoxine-pyrimethamine (SP) every two months and placebo over a period of six months" Meremikwu et al. 2012, pg. 28.
    • Sesay et al. 2011: "During the 2008 malaria transmission season, 1,277 children under five years of age resident in villages within the rural Farafenni demographic surveillance system (DSS) in North Bank Region, The Gambia were randomized to receive monthly IPTc with a single dose of sulphadoxine/pyrimethamine (SP) plus three doses of amodiaquine (AQ) or SP and AQ placebos given by village health workers (VHWs) on three occasions during the months of September, October and November, in a double-blind trial", abstract.
    • Tagbor et al. 2016: "SMC involves monthly administration of the long-acting antimalarial drugs sulfadoxine-pyrimethamine and amodiaquine (SP-AQ) to all children under five years of age", pg. 224.

  • 93

    "Seven trials (12,589 participants), including one cluster-randomized trial, met the inclusion criteria. All were conducted in West Africa, and six of seven trials were restricted to children aged less than 5 years." Meremikwu et al. 2012, pg. 2.

  • 94

    "Seven trials (12,589 participants), including one cluster-randomized trial, met the inclusion criteria. All were conducted in West Africa, and six of seven trials were restricted to children aged less than 5 years." Meremikwu et al. 2012, pg. 2.
    Tagbor et al. 2016, which is not included in the Cochrane review: "SMC involves monthly administration of the long-acting antimalarial drugs sulfadoxine-pyrimethamine and amodiaquine (SP-AQ) to all children under five years of age", pg. 224.
    The Cochrane trial which treated children above age five is Dicko et. al. 2008, which we exclude from our meta-analysis.
    “Dicko 2008…
    Participants 262 children aged from months to 10 years”. Meremikwu et al. 2012, pg. 24.

  • 95
    • Cissé et al. 2006: "1136 children aged 2–59 months received either one dose of artesunate plus one dose of sulfadoxine-pyrimethamine or two placebos on three occasions during the malaria transmission season." pg. 659.
    • Dicko et al. 2011: "After screening, eligible children aged 3–59 mo were given a long-lasting insecticide-treated net (LLIN) and randomised to receive three rounds of active drugs or placebos" pg. 1.
    • Konaté et al. 2011: "Sulphadoxine Pyrimethamine (SP) SP + Amodiaquiune AQ or Placebo tablets were given in August, September, and October during the peak malaria transmission season, with one month intervals between treatments" Meremikwu et al. 2012, pg. 27.
    • Kweku et al. 2008: "Intermittent treatment: IPTc with artesunate plus amodiaquine (AS+AQ) monthly or every two months, or sulphadoxine-pyrimethamine (SP) every two months and placebo over a period of six months" Meremikwu et al. 2012, pg. 28.
    • Sesay et al. 2011: "During the 2008 malaria transmission season, 1,277 children under five years of age resident in villages within the rural Farafenni demographic surveillance system (DSS) in North Bank Region, The Gambia were randomized to receive monthly IPTc with a single dose of sulphadoxine/pyrimethamine (SP) plus three doses of amodiaquine (AQ) or SP and AQ placebos given by village health workers (VHWs) on three occasions during the months of September, October and November, in a double-blind trial", abstract.
    • Tagbor et al. 2016: "SMC or placebo was delivered on five occasions during the rainy season", pg. 224.

  • 96
    • Cissé et al. 2006: "Use of bednets was limited in the study population. In the control group, 128 children (23%) slept under a bednet, compared with 116 (21%) in the active treatment group", pg. 662.
    • Dicko et al. 2011: "The coverage of ITNs at baseline was 33.4% (312/935) in Siby, 84.7% (563/665) in Djoliba, and 89.8% (2,207/2,458) in Ouelessebougou." pg. 3.
    • Konaté et al. 2011: "the proportions of children who used an ITN before the intervention were also similar (less than 0.5%)." pg. 5.
    • Kweku et al. 2008: 20.1%-21.7%. Figure 1.
    • Sesay et al. 2011: "93% of the study subjects slept under an ITN", pg. 5.
    • Tagbor et al. 2016: "Firstly, due to a recent distribution campaign, around 80% of children were regularly sleeping under an insecticide-treated net." pg. 231.

    Note: we have not reviewed how coverage was measured in each study, and there may be differences in definitions between these reported figures.

  • 97

    See this row in our cost-effectiveness analysis. Note that we believe there are still differences between study contexts and contemporary SMC that are likely to have a bearing on our cost-effectiveness analysis (e.g., differences in baseline mortality), but these are accounted for elsewhere in our analysis.

  • 98

    See this row in our cost-effectiveness analysis.

  • 99

    Other reasons why we might expect published studies to overstate efficacy, on average:

    • Researcher bias: researchers are incentivized to find positive results in their research. Researchers may also have some degree of freedom in how data is analyzed and presented, which increases the likelihood of positive results being found and published.
    • Implementation of the intervention: On average, we would expect an intervention to be delivered more effectively in a controlled study environment than in a large-scale program. This effect is separately (but may not be fully) accounted for through other parameters in our analysis, e.g., our analysis of cost per SMC cycle, which accounts for the percentage of children reached in contemporary programs).

  • 100

    See this row of our cost-effectiveness analysis.

  • 101

    Note: This adjustment assumes that the findings reported in the Cochrane meta-analysis are “intention-to-treat” estimates (including all children originally allocated to receive SMC) rather than “per-protocol” estimates (including only children who actually received SMC). The Cochrane analysis reports using intention-to-treat estimates where possible. Our review of the underlying studies found that four studies report intention-to-treat estimates and two report modified intention-to-treat estimates (whereby only children who receive the first course of the drug or placebo were included in the analysis). We do not see this as a serious concern to trial validity because:

    • In one of the two trials that used modified intention-to-treat (Cissé et. al. 2006), only 13 children did not receive the first dose out of 1088 randomized (see the diagram on this sheet).
    • In the other trial that used modified intention-to-treat (Kweku et. al. 2008), 151 children did not receive the first dose out of 2,602 enrolled. However, our understanding is that children were randomized and given the first dose at the same time in this study, so that all children who were randomized received at least one dose (see the diagram on this sheet). We think that this poses minimal threat to trial validity.

    For the type of analysis used, see this column in our accompanying spreadsheet. Sesay et. al. 2011 (type of analysis not discussed in spreadsheet) reports: “For the primary outcome of malaria incidence during the study period, analysis was by intention to treat.”
    “We aimed to carry out the analysis according to the intention to-treat principle. However, when there was loss to follow up, a complete-case analysis was employed, such that, patients for whom no outcome was reported were excluded from the analysis. This analysis assumes that the patients for whom an outcome is available are representative of the original randomized patients.” Meremikwu et al. 2012, pg. 9.

  • 102

    Six of the seven trials in the Cochrane meta-analysis were individually randomized, and one was cluster randomized: “Seven trials (12,589 participants), including one cluster-randomized trial, met the inclusion criteria. All were conducted in West Africa, and six of seven trials were restricted to children aged less than 5 years.”
    The cluster randomized trial is Tagbor et. al. 2011, which we exclude from our adjusted meta-analysis because it is not included in the Cochrane meta-analysis of incidence of clinical malaria. Our understanding is that this is because it reported incidence of fever episodes without parasitological confirmation. “Clinical malaria: Six trials reported on incidence of clinical malaria, while one reported incidence of fever episodes without parasitological confirmation (Tagbor 2011).” Meremikwu et al. 2012, pg. 9.
    We also exclude Dicko et. al. 2008 from our analysis, and add Tagbor et. al. 2016 (for the reasons discussed above). Tagbor et. al. 2016 was individually randomized: “method Individually randomised, placebo-controlled trial in the Ashanti Region of Ghana. A total of 2400 children aged 3–59 months received either: (i) a short-acting ACT for case management of malaria (artemether-lumefantrine, AL) plus placebo SMC, or (ii) a long-acting ACT (dihydroartemisinin-piperaquine, DP) for case management plus placebo SMC or (iii) AL for case management plus active SMC with sulphadoxine-pyrimethamine and amodiaquine.” Tagbor et. al. 2016, abstract.

  • 103

    “IPTc probably produces a small reduction in all-cause mortality consistent with the effect on severe malaria, but the trials were underpowered to reach statistical significance (risk ratio 0.66, 95% CI 0.31 to 1.39, moderate quality evidence).” Meremikwu et al. 2012, abstract.

  • 104

    See this row in our cost-effectiveness analysis.

  • 105

    “IPTc also prevented around three-quarters of severe malaria episodes (rate ratio 0.27, 95% CI 0.10 to 0.76; 5964 participants, two trials; Analysis 1.2).” Meremikwu et al. 2012, pg. 12.

  • 106

    “IPTc probably produces a small reduction in all-cause mortality consistent with the effect on severe malaria, but the trials were underpowered to reach statistical significance (risk ratio 0.66, 95% CI 0.31 to 1.39, moderate quality evidence).” Meremikwu et al. 2012, abstract.

  • 107

    See this row in our cost-effectiveness analysis. This range includes Burkina Faso, Togo, and various states in Nigeria. We exclude Chad, Mozambique, Uganda, and DRC from this range for the reasons discussed in this section.

  • 108

    See this row in our cost-effectiveness analysis. This range includes Burkina Faso, Togo, and various states in Nigeria. We exclude Chad, Mozambique, Uganda, and DRC from this range for the reasons discussed in this section.

  • 109

    Note: this figure does not appear directly in our cost-effectiveness analysis, because we present the annual mortality rate and the proportion of malaria mortality occurring in the SMC season in different rows.

  • 110

    Children 3 to 59 months old are the target group for SMC (see discussion above for further details). The GBD estimates do not use 3-59 month olds as a grouping, and so we need to restructure the data to estimate mortality rates for this group.
    The main simplifying assumption we make in this conversion is that we assume the number of deaths from malaria in 1-2 month olds is the same as in 3-11 month olds. Our understanding is that this assumption may be overly conservative, and relatively fewer deaths from malaria may occur in 1-2 month olds than later in the post neonatal period. We have assumed an equal mortality rate because it's simple and doesn't make a big difference to our results. See this section of our analysis for further details.

  • 111

    See our estimates for Nigeria in this sheet. Note that we also use province-level data for Mozambique, Uganda, and DRC but we largely exclude these countries from our discussion in this report for the reasons discussed above.

  • 112

    The GBD model that our analysis relies on already incorporates the expected impact of another malaria intervention, insecticide-treated nets (ITNs), into its estimates of malaria mortality. Alexandra Walker, Engagement Officer, IHME, email to GiveWell, July 21, 2021 (unpublished). See this section of our separate report on insecticide-treated nets for further information.
    In most countries, our understanding is that ITN campaigns have been occurring on relatively consistent three-year intervals and therefore we would guess that these estimates are relatively up-to-date. However, in Nigeria, these campaigns take place at the state level on a rolling basis. We expect these campaigns to have become more frequent in the years since 2019 than in the years before (details on reasoning below), and would therefore expect the Nigeria estimates to be out of date. This implies that malaria mortality in Nigeria is somewhat lower than the GBD estimates, and therefore that SMC is somewhat less cost-effective. We account for this with a -5% adjustment in our analysis for states in Nigeria except FCT. See this row.
    Reasoning for thinking ITN campaigns have become more frequent since 2019 in Nigerian states where Malaria Consortium delivers SMC:

    • GiveWell began funding ITN campaigns in Nigeria for the first time in three of the nine states currently (as of September 2023) supported by Malaria Consortium with GiveWell funding (Bauchi, Benue, and Plateau states), beginning in 2021. We would expect the impact of this funding to cause people in all 11 PMI-supported states to receive an ITN earlier than they otherwise would. See this grant page and this grant page for more details.
    • In 2020, Nigeria's National Malaria Elimination Programme negotiated with the World Bank, Islamic Development Bank, and African Development Bank to secure loan funding for malaria control in 13 states which were not previously supported for malaria programs by the two largest external malaria funders, the Global Fund and PMI. Three of these 13 states (Borno, FCT, and Kogi) are currently (as of September 2023) supported by Malaria Consortium with GiveWell funding. See Malaria Consortium, Net-target project report: Nigeria, 2020 (redacted), pg. 7.

    We apply a larger (-35%) adjustment for the Federal Capital Territory (FCT) because our understanding is that the most recent insecticide-treated bed net campaign in FCT was in 2011, but a campaign is planned. We expect this will mean lower malaria mortality rates than implied by GBD. See this cell note for our reasoning.

  • 113

    See this row in our cost-effectiveness analysis.
    Note: in some cost-effectiveness analyses for other interventions, we adjust GBD estimates of baseline mortality to account for the impact of the intervention already being "baked in" to the estimates. For example, for insecticide-treated nets our understanding is that the GBD estimates aim to capture the overall mortality risk in each location, and incorporate the expected benefit of previous net programs. Because our analysis aims to estimate the impact of increasing access to nets among children who would not otherwise have them, we need to adjust the mortality estimates to account for this.
    We do not include an adjustment like this for SMC because our understanding from discussions with IHME researchers is that the GBD estimates do not incorporate the expected benefits of SMC programs, which have only recently been scaled up in recent years.
    Alexandra Walker, Engagement Officer, IHME, email to GiveWell, July 21, 2021 (unpublished).
    For details of our adjustment for insecticide-treated nets, see this section of our separate report.

  • 114

    This understanding is based on multiple conversations with IHME researchers. Detailed modeling assumptions for the GBD estimates are available in Institute for Health Metrics and Evaluation, GBD 2019 methods appendix.

  • 115

    "Despite its importance, current knowledge on the nature and drivers of changing endemicity in sub-Saharan Africa is remarkably weak. National health records in 32 highly endemic countries (together accounting for about 90% of the global malaria burden) are considered inadequate to assess trends in malaria cases. This stems from low care-seeking rates (many malaria cases are not seen at formal health facilities), incomplete record keeping and curation (many recorded cases are never captured in surveillance databases), and historically poor access to parasitological diagnosis (malaria cases were often diagnosed presumptively with poor specificity).” Bhatt et al. 2015.

  • 116

    We have not yet received permission from Rethink Priorities to publish this report. Rethink Priorities, Malaria Deaths: A Comparison of WHO and IHME Estimates, 2023 (unpublished).

  • 117

    “That data, in fact, shows that WHO mortality estimates are on aggregate nearly 17% less than those published by the IHME, for the assessed year 2019.”
    “Later in this report, we will discuss the methodological changes that resulted in the clear split between pre-2021 and post-2021 WHO data. At a glance, the interested viewer can tell that the World Health Organization’s recent changes adjusted aggregate deaths upward significantly, by about 18% over the assessed period 2000-2019.” Rethink Priorities, Malaria Deaths: A Comparison of WHO and IHME Estimates, 2023 (unpublished), pg. 4.

  • 118

    “The bulk of malaria deaths come from the Sub-Saharan Africa (SSA) region (85%-91%, depending on the source). In this region, the difference between the 2021 WHO data and the most recent IHME model is around 7% for the assessed years 2017-2019, compared to a global gap of around 11%. Over the period 2000-2019, the overall difference for sub-Saharan Africa is around 12%, whereas the global gap between the two organizations' estimates is closer to 17%.” Rethink Priorities, Malaria Deaths: A Comparison of WHO and IHME Estimates, 2023 (unpublished), pg. 6.

  • 119

    “As deaths for children under 5 are a significant component of GiveWell’s model, we looked further into whether adults are indeed the root of the remaining gap between the IHME and WHO datasets. Our primary finding is that the World Health Organization consistently reports about 20-30% higher deaths of children under five than the IHME, whereas its figures for adult deaths are about 50% lower. “ Rethink Priorities, Malaria Deaths: A Comparison of WHO and IHME Estimates, 2023 (unpublished), pg. 7.

  • 120

    “There is significant controversy over verbal autopsy data and adult deaths. One of the experts we interviewed expressed a worry that IHME data overstates adult mortality and understates child mortality as a pervasive modeling flaw. Indeed, we have found that there is an ongoing debate in the literature about the age structure of malaria mortality. In the 1990s and 2000s, it was widely accepted that in endemic areas, malaria was mostly fatal to children under 5.
    When the IHME’s 2012 estimates were released, researchers noticed that the large estimated number of deaths in people older than 5 years in Africa was out of step with other estimates, and criticized the IHME’s reliance on verbal autopsy data, stating “We do not believe that verbal autopsy can distinguish severe malaria from other severe febrile illnesses and so is fundamentally flawed as an estimator of malaria mortality” (White et al, 2012, p. 560).” Rethink Priorities, Malaria Deaths: A Comparison of WHO and IHME Estimates, 2023 (unpublished), pg. 16.

  • 121

    See this row in our cost-effectiveness analysis.

  • 122

    “Indirect consequences of P. falciparum infection include anemia (unless anemia is linked to acute high-density parasitemia as a direct cause), low birthweight, growth retardation, or undernutrition. In addition, malaria infection can increase the severity of other comorbid infectious diseases through immune suppression or enhanced invasive capacities across physical barriers to infection (for example, blood and tissue). Previous approaches to the global burden of disease have assumed that each death must be attributed to a single cause and can be fitted into the fixed disease-mix matrix of all causes (Murray and Lopez 1997).” Jamison et al. 2006, p. 204.

  • 123

    "During randomized controlled intervention trials aimed at reducing the incidence of infection (but not 100 percent protective), the all-cause mortality of children is often reduced more than would be attributed by VA diagnosis of malaria. For example, in Kilifi the proportion of deaths of children under five years attributed to malaria by VA was 34 percent (R. W. Snow, unpublished data). During a randomized controlled trial of insecticide-treated bednets in the same area, the incidence of malaria infection was reduced by 50 percent (Snow et al. 1996), which was sufficient to reduce all-cause mortality by 33 percent (Nevill et al. 1996). More dramatically, in The Gambia, insecticide-treated bednets reduced all-cause mortality by over 60 percent, and yet the VA-diagnosed contribution of malaria to all-cause mortality among control populations was only 16 percent (Alonso et al. 1993). This has led some to speculate that malaria infection is a contributor to broad causes of mortality beyond the direct fatal consequences of infection (Molineaux 1997)." Jamison et al. 2006, p. 204.
    "Data on all-cause mortality of children under five from DSS studies undertaken across a broad range of malaria transmission settings in Sub-Saharan Africa were analyzed against the prevalence of P. falciparum infection at each site. Weighted least-squares regression was used to model the contiguous relationships between all-cause mortality and parasite prevalence rates, allowing for the square of parasite prevalence (for possible saturation of parasite prevalence), timing, location, and the sampling precision of each study (Snow, Korenromp, and Gouws 2004). The unadjusted median all-cause child mortality rate for low prevalence areas of childhood infection (less than 25 percent) was 10.9 per year per 1,000 children under five (IQR 7.8–17.6). This rose dramatically to 39.1 per year per 1,000 children (IQR 32.8–52.2) among populations exposed to childhood parasite prevalence risks greater than or equal to 25 percent. In the regression model, mortality increased significantly with parasite prevalence, but this effect leveled off at higher prevalence rates. The model suggested that, in rural DSS sites throughout Sub-Saharan Africa, all-cause mortality increases by more than twofold (25–30 deaths per 1,000 children under five years old) over the prevalences of malaria infection covered by the DSS sites, and parasite prevalence explained 64 percent of the variation between sites in all-cause under-five mortality. By contrast, the direct estimation of malaria-specific mortality presented earlier for children living under stable endemic conditions was only 28.2 percent." Jamison et al. 2006, p. 206.

  • 124

    We do not have documentation from these conversations and so our notes are unpublished.

  • 125

    Our analysis is based on Pryce, Richardson, and Lengeler 2018, a Cochrane meta-analysis of insecticide-treated nets. The meta-analysis is informative about indirect deaths averted from reduced malaria because it estimated the impact of nets both on all-cause mortality and on malaria incidence. Combined with other sources of data, this allows us to infer the relationship between direct malaria mortality and indirect malaria mortality.
    Pryce et. al. 2018 estimates that:

    • Nets reduce malaria episodes by 45% compared to no nets.
    • Nets reduce all-cause child mortality by 17%, compared to no nets.

    “Insecticide‐treated nets reduce child mortality from all causes by 17% compared to no nets (rate ratio 0.83, 95% CI 0.77 to 0.89; 5 trials, 200,833 participants, high‐certainty evidence). This corresponds to a saving of 5.6 lives (95% CI 3.6 to 7.6) each year for every 1000 children protected with ITNs. Insecticide‐treated nets also reduce the incidence of uncomplicated episodes of Plasmodium falciparum malaria by almost a half (rate ratio 0.55, 95% CI 0.48 to 0.64; 5 trials, 35,551 participants, high‐certainty evidence) and probably reduce the incidence of uncomplicated episodes of Plasmodium vivax malaria (risk ratio (RR) 0.61, 95% CI 0.48 to 0.77; 2 trials, 10,967 participants, moderate‐certainty evidence).” Pryce, Richardson, and Lengeler 2018, abstract.
    In the studies that measured reduced malaria episodes, we estimate (based on GBD estimates from the locations and the times when the studies were conducted) that approximately 15% of deaths in these studies were direct malaria deaths. Cross referencing these figures implies that to achieve an all-cause mortality reduction of 17%:

    • ITN distributions reduced malaria incidence by 45%, and the ratio of direct to indirect malaria deaths is 1.5.
    • ITN distributes reduced malaria incidence by 75%, and the ratio of direct to indirect malaria deaths is 0.5.
    • Or somewhere in between these, e.g., ITN distributions reduced malaria incidence by 60%, and the ratio of direct to indirect malaria deaths is 0.9.

    Our calculations are available here.
    Our best guess is that the ratio of 1.5 indirect deaths for every direct death is an overestimate, because:

    • The studies that provide more than 99% of the weight in the Pryce, Richardson, and Lengeler meta-analysis took place in the 1980s, 1990s and 2000s. It is plausible that the ratio of direct to indirect deaths has fallen since that time in malaria endemic countries as overall health has improved and under five mortality has decreased.
    • We use national-level estimates from the Global Burden of Disease project from the countries where the studies in Pryce, Richardson, and Lengeler 2018 were conducted as inputs to our analysis (see here). Intuitively, we might expect the Pryce, Richardson, and Lengeler 2018 RCTs to be conducted in areas where malaria mortality was higher than the national average. If that assumption is correct, our analysis would be likely to overestimate the share of indirect deaths and underestimate the share of direct malaria deaths.

    See this section of our report on insecticide-treated nets for further information.

  • 126

    “As an example, after all adjustments we estimate that in-line chlorination in Kenya reduces all-cause mortality by 11% in children under five and 2% in people five and over. In contrast, an estimate obtained by multiplying the reduction in diarrhea risk caused by in-line chlorination by the GBD estimate of under-five deaths caused by enteric infections in Kenya is 2.9%, suggesting that our estimate is about 3.7 times as large as expected based on indirect estimation methods. This implies that for each enteric infection death averted in children under five, 2.7 deaths are averted from other causes.” See this section of our report on water quality interventions.

  • 127

    See this row in our cost-effectiveness analysis.

  • 128

    We could attempt to estimate the indirect deaths ratio using the following parameters:

    • The impact of SMC on malaria from the Cochrane meta-analysis.
    • The impact of SMC on all-cause mortality from the Cochrane meta-analysis (although this was underpowered, as discussed above).
    • A best guess about what share of all-cause child mortality was attributable to malaria at the time the Cochrane studies were conducted (e.g., from the Global Burden of Disease Project).

    This would be similar to the analysis we have conducted for insecticide-treated nets here.

  • 129

    “Most estimates of the burden of malaria are based on its direct impacts; however, its true burden is likely to be greater because of its wider effects on overall health. Here we estimate the indirect impact of malaria on children’s health in a case-control study, using the sickle cell trait (HbAS), a condition associated with a high degree of specific malaria resistance, as a proxy indicator for an effective intervention. We estimate the odds ratios for HbAS among cases (all children admitted to Kilifi County Hospital during 2000–2004) versus community controls. As expected, HbAS protects strongly against malaria admissions (aOR 0.26; 95%CI 0.22–0.31), but it also protects against other syndromes, including neonatal conditions (aOR 0.79; 0.67–0.93), bacteraemia (aOR 0.69; 0.54–0.88) and severe malnutrition (aOR 0.67; 0.55–0.83). The wider health impacts of malaria should be considered when estimating the potential added benefits of effective malaria interventions.” Uyoga et. al. 2019, abstract.

  • 130

    See this row in our cost-effectiveness analysis. Note: we use a lower value (52% for Mozambique and DRC and 62% in Uganda), because our understanding is that malaria seasonality tracks rainfall less well in East Africa. However, we largely exclude these countries from our discussion in this report for the reasons discussed above.

  • 131

    Consistent with this assumption, a World Health Organization report claims that SMC offers a high level of protection for about four weeks after the last treatment, after which protection appears to decline. We have not independently vetted this claim.
    “The results of clinical trials indicate that a high level of protection against uncomplicated clinical malaria is likely to be maintained for only 4 weeks after administration of each treatment course of SP+ AQ; thereafter, protection appears to decay rapidly.” WHO, SMC field guide, 2013, pg. 4.

  • 132

    “SMC is recommended for use in areas with highly seasonal malaria transmission, and it is likely to be most cost-effective where the burden of malaria is highest in children. The suitability of an area for SMC is determined by the seasonal pattern of rainfall, malaria transmission and the burden of malaria. SMC is recommended for deployment in areas: . . . where more than 60% of the annual incidence of malaria occurs within 4 months.” WHO, SMC field guide, 2013, pg. 8.

  • 133

    “The priority target areas for SMC implementation are those where: • P. falciparum malaria transmission is highly seasonal and the majority (>60%) of clinical malaria cases occur within 4 consecutive months.” WHO, SMC field guide, 2nd ed., 2023, pg. 2.

  • 134

    "In sites considered suitable for SMC, the median fraction of incidence occurring in the 4 consecutive months of peak transmission was 77% and the mean 75.7% (Supplementary Table S1,2). We therefore assumed that, on average, 75% of the annual burden occurred in the SMC period”. Cairns et al. 2012, pg. 9.

  • 135

    Since SMC programs have expanded geographically since 2012 (with 60% of malaria incidence in the high transmission as the lower bound for most of this expansion period), we would expect average seasonality in the locations covered by this grant to be somewhat lower than those found in the 2012 paper.

  • 136

    “SMC is conducted during the high transmission period. The start and end dates depend on the pattern of malaria transmission, which generally correlates with rainfall.” WHO, SMC field guide, 2013, pg. 19.

  • 137

    We have only investigated the relationship between rainfall and the number of SMC cycles delivered in Nigeria, not in Burkina Faso (the other country in the Sahel where Malaria Consortium delivers five cycles of SMC in some locations). See this row in our cost-effectiveness analysis for the number of cycles by location.
    In Nigeria, the following GiveWell-funded states in Nigeria receive five cycles of SMC (as of December 2023):

    • Kogi
    • Nasarawa
    • Plateau
    • Oyo
    • Federal Capital Territory (FCT)

    We estimate (based on using a single city as a proxy) that 69% to 81% of annual rainfall falls within the four month rainy season in these states. This compares to 86% to 90% in GiveWell-funded states in Nigeria that receive four cycles of SMC (Bauchi, Borno, Kebbi, and Sokoto). See this spreadsheet for our calculations.

  • 138

    This range includes Burkina Faso, Togo, and various states in Nigeria. See this row in our cost-effectiveness analysis. Note, we exclude Chad, Mozambique, Uganda, and DRC from this range for the reasons discussed in this section.

  • 139

    See this row in our cost-effectiveness analysis. This range includes Burkina Faso, Togo, and various states in Nigeria. We exclude Chad, Mozambique, Uganda, and DRC from this range for the reasons discussed in this section.

  • 140

    See this row in the “Sensitivity analysis” section cost-effectiveness analysis.
    For ease of modeling, we quantify our uncertainty ranges for benefits other than mortalities averted among people under age 5 as percentage adjustments applied to other modeled benefits of the program. For Burkina Faso:

    • Our 25th to 75th percentile range for averted deaths among older children and adults is a +3% adjustment to a +11% adjustment.
    • This roughly translates to between 2% and 8% of the total modeled benefits of the program, as shown in the table above.
    • This conversion is imperfect because varying the proportion of the program's total impact that is contributed by one type of benefit will also affect the proportion of impact contributed by other types of benefits, but we think the values presented in the table are a reasonably good approximation of our uncertainty level for this parameter.

  • 141

    “SMC was introduced into three districts over three years in central Senegal using a stepped-wedge cluster-randomised design” Cisse et al. 2016, abstract.
    “Fifty-four health posts were randomised. Nine started implementation of SMC in 2008, 18 in 2009, and a further 18 in 2010, with 9 remaining as controls. In the first year of implementation, SMC was delivered to children aged 3–59 months; the age range was then extended for the latter two years of the study to include children up to 10 years of age.” Cisse et al. 2016, abstract.
    See Cisse et al. 2016, Fig 1 Trial profile.
    Stepped-wedge studies involve sequentially rolling out a program to participants (either people or locations) over a number of time periods. Outcomes are compared between the people or locations which receive the program in a particular time period compared to those which don’t.
    “Stepped wedge randomised trial designs involve sequential roll-out of an intervention to participants (individuals or clusters) over a number of time periods. By the end of the study, all participants will have received the intervention, although the order in which participants receive the intervention is determined at random.” Brown and Lilford 2006, abstract.

  • 142

    “SMC implementation started in September 2008 in Zone 1, treating about 14,000 children aged 3–59 months each month (mid-September, mid-October, and mid-November). About 90,000 children under ten years of age were treated in 2009, and about 160,000 were treated in 2010.” Cisse et al. 2016, pg. 9.

  • 143
    • “Delivery was organised by health post using community health workers who visited each household once per month during the transmission seasons of 2008, 2009, and 2010 (in mid-September, mid-October, and mid-November) to administer SMC to children aged 3–59 months (in 2008) and to children up to 10 years of age in 2009 and 2010. The dose of SP and the first dose of AQ was administered by the health worker or by the mother/carer under supervision by the health worker; the remaining daily doses of AQ were left with the mother to administer at home over the next two days.” Cisse et al. 2016, pg. 7.
    • “In 2008, co-blister packs of Dualkin (Pfizer) (SP [500/25 mg tablets] and AQ [200 mg tablets]) were used. These tablets had to be broken and crushed to make half doses for infants. In 2009, 200-mg AQ tablets (Chongqing Qinyang Pharmaceutical) and 500/25-mg SP (Shijizhuang Ouyi Pharmaceutical) tablets were used, and, in 2010, sweetened dispersible tablets of SP (500/25 mg) and of AQ (153 mg) (Kinapharma, Ghana) were given.” Cisse et al. 2016, pg. 7.

  • 144

    “In the first year of implementation, SMC was delivered to children aged 3–59 months; the age range was then extended for the latter two years of the study to include children up to 10 years of age.” Cisse et al. 2016, abstract.

  • 145

    “The incidence rate ratio comparing SMC and control areas, adjusted for effects of calendar time, for RDT-confirmed malaria during the transmission season was 0.43 In children under five years of age and 0.39 for children five to nine years of age (Table 2), giving reductions of 57% (95% CI 48%–63%) in the younger group and 61% (95% CI 55%–67%) in the older group, an overall effectiveness of 60% (95% CI 54%–64%, p < 0.001).” Cisse et al. 2016, pg. 11.

  • 146

    “Among age groups too old to receive SMC, incidence was reduced by 26% (95% CI 18%–33%, p < 0.001) (RDT-confirmed), or 29% (95% CI 21%–35%, p < 0.001) (confirmed and unconfirmed cases) in areas where SMC was delivered to children compared to control areas (Fig 3).” Cisse et al. 2016, pg. 11.

  • 147

    “During the transmission seasons, the mortality rate among children aged 3–59 months in SMC areas was 4.6 per 1,000 child-years at risk (197 deaths) and 4.5/1,000 in control areas (159 deaths). Among children aged 5–9 years, the mortality rates were 1.30/1,000 in SMC areas (45 deaths) and 1.2/1,000 in control areas (43 deaths). Mortality in infants under three months of age was 19/1,000. The mortality rate ratio SMC: control given by a random effect model that included cluster, time, and age effects was 0.89 (95% CI 0.65–1.2, p = 0.442) for the 3–59 months age group and 0.97 (95% CI 0.61–1.6, p = 0.916) for the 5- to 9-year-olds. The pooled estimate for both age groups combined was 0.90 (95% CI 0.68–1.2 p = 0.496) (S3 Fig, S2 Table).” Cisse et al. 2016, pg. 10.

  • 148

    “Sample size calculations to detect an effect on all-causes mortality were done for a range of scenarios, with the assumption that the mortality rate (all causes) would be 20 per 1,000 in children 3–59 months of age (based on estimates from the Niakhar DSS) and assuming 30% of deaths were due to malaria (determined from verbal autopsies in Niakhar and the proportion of deaths in hospital attributed to malaria). We also assumed a coefficient of variation between health posts of 0.25 to 0.3 and SMC effectiveness against malaria to be 70%. Calculations using the method of Hemming and Girling [13] for the stepped-wedge design, assuming k = 0.3 and a median population of children aged 3–59 months of 1,400 per cluster, gave a power of about 17%, 52%, and 99% for reductions of 5%, 10%, and 20%, respectively, if the mortality rate was 10/1,000 and 11%, 32%, and 86% power for the same reductions if the rate was 20/1,000.” Cisse et al. 2016, pg. 8.

  • 149

    “We did not observe an effect on all-cause mortality. There had been a gradual reduction in mortality rates in the years before the study related to improved vaccination coverage and access to primary health care [16], but the sudden drop in mortality in our study was unexpected.” Cisse et al. 2016, pg. 15.

  • 150

    See this section in our cost-effectiveness analysis.

  • 151

    See this row in our cost-effectiveness analysis.

  • 152

    “In the first year of implementation, SMC was delivered to children aged 3–59 months; the age range was then extended for the latter two years of the study to include children up to 10 years of age.” Cisse et al. 2016, abstract.

  • 153
    • We estimate that, as a proportion of the overall population, Malaria Consortium’s program treats 63% of the population treated in Cissé et. al. 2016 (see this cell in our analysis). This is based on children aged 3-59 months being 53% of the treated population in Cissé et. al. 2016, and data from a 2016 SMC survey which found that 18.7% of the population treated were actually 6 or 7 years old (e.g., because of implementer error; 53% x (100% + 18.7%) = 63%). See this sheet for our calculations.
    • We assume that doubling the proportion of the population treated results in a ~3x increase in indirect effects (see this row in our analysis). This assumption is based on an unpublished modeling study shared with us by Professor Matthew Cairns, a researcher at London School of Hygiene and Tropical Medicine. The paper used the Imperial College malaria model (available here) to simulate 4 monthly cycles of SMC targeted at different age ranges with 80% coverage, fitting efficacy over time with Zongo et al. 2015. It suggests treating 5-10 year olds would result in ~3x the indirect effects of treating 3 month to 4 year olds in the countries in which Malaria Consortium works. Professor Matthew Cairns, email to GiveWell, September 19, 2018 (unpublished).
    • We combine these factors to scale our estimate of indirect benefits. Based on the estimate that ~doubling the proportion of the population covered increases the indirect effects by ~3x, we assume that the indirect effects scale according to raising the proportion of the population covered to the power of log, base 2 (3). Combined with the estimate that, as a proportion of the overall population, Malaria Consortium’s program treats 63% of the population treated in Cissé et. al. 2016, this results in an overall adjustment for this factor of 47% (63%^(log(3,2)) = 47%). See this cell for our calculation.

    See this sheet for our calculations.

  • 154

    We have seen early unpublished results of a modeling study shared with us by Professor Matthew Cairns, a researcher at London School of Hygiene and Tropical Medicine, which suggests that indirect effects would be ~35% lower in the typical area covered by Malaria Consortium, than in Cisse et al. 2016 due to higher transmission rates in areas covered by Malaria Consortium. As above, the paper used the Imperial College malaria model (available here) to simulate 4 monthly cycles of SMC targeted at different age ranges with 80% coverage, fitting efficacy over time with Zongo et al. 2015. Professor Matthew Cairns, email to GiveWell, September 19, 2018 (unpublished).
    We have not vetted this model in detail, but checked estimates from the Institute for Health Metrics and Evaluation, which suggest malaria prevalence is substantially lower in Senegal than most countries in which Malaria Consortium works (estimates available here).

  • 155

    See this row in our analysis.

  • 156

    Professor Cairns told us the study used the Imperial College malaria model, which we have not vetted in detail, to simulate 4 monthly cycles of SMC targeted at different age ranges with 80% coverage, fitting protective efficacy over time with Zongo et al. 2015. Zongo et. al. 2015 evaluated an SMC program using dihydroartemisinin-piperaquine (DHAPQ) rather than SPAQ (used in Malaria Consortium’s program).

    • “The WHO recommends that children living in areas of highly seasonal malaria transmission in the Sahel subregion should receive seasonal malaria chemoprevention (SMC) with sulfadoxine-pyrimethamine plus amodiaquine (SPAQ). We evaluated the use of dihydroartemisinin-piperaquine (DHAPQ) as an alternative drug that could be used if SPAQ starts to lose efficacy.” Zongo et al. 2015, Abstract.

  • 157
    • Total modeled benefits excludes our supplemental adjustments discussed in sections 4.5 and 4.6.
    • See this row in our cost-effectiveness analysis. This range includes Burkina Faso, Togo, and various states in Nigeria. We exclude Chad, Mozambique, Uganda, and DRC from this range for the reasons discussed in this section.

  • 158

    For ease of modeling, we quantify our uncertainty ranges for benefits other than mortalities averted among people under age 5 as percentage adjustments applied to other modeled benefits of the program. For Burkina Faso:

    • Our 25th to 75th percentile range for income effects is a +10% adjustment to a +28% adjustment (to the total modeled benefits of the program). See this row in the “Sensitivity analysis” section of our cost-effectiveness analysis.
    • This roughly translates to between 9% and 22% of the total modeled benefits of the program, as shown in the table above.
    • This conversion is imperfect because varying the proportion of the program's total impact that is contributed by one type of benefit will also affect the proportion of impact contributed by other types of benefits, but we think the values presented in the table are a reasonably good approximation of our uncertainty level for this parameter.

  • 159

    Bleakley 2010.

  • 160

    Cutler et al. 2010.

  • 161

    “We use the national malaria eradication program in India in the 1950s as a quasi experiment and exploit geographic variation in malaria prevalence prior to the eradication campaign. We compare gains for cohorts born before and after the program in areas with varying pre-eradication prevalence.” Cutler et al. 2010, p. 73.

  • 162

    “This study uses the malaria-eradication campaigns in the United States (circa 1920), and in Brazil, Colombia and Mexico (circa 1955) to measure how much childhood exposure to malaria depresses labor productivity.” Bleakley 2010, abstract.

  • 163

    “We use the national malaria eradication program in India in the 1950s as a quasi experiment and exploit geographic variation in malaria prevalence prior to the eradication campaign. We compare gains for cohorts born before and after the program in areas with varying pre-eradication prevalence. These differences-in-differences estimates show no gains in literacy or primary school completion between areas that experienced large reductions in malaria and those that experienced small reductions in malaria. We do, however, observe modest relative increases in economic status (proxied by household per capita household expenditure) for prime age men. This effect is robust to using localized sources of geographic variation and to instrumenting for pre-eradication prevalence using geographic and climate factors, although in our most demanding specification (identified using within-region variation and including district linear trends) the point estimates remain similar but become imprecise. We do not observe significant increases in expenditure as a result of the program for women, and these gender differences cannot be explained by differences in the household composition of treated men and treated women.” Cutler et al. 2010, p. 73.
    “Relative to non-malarious areas, cohorts born after eradication had higher income as adults than the preceding generation. These cross-cohort changes coincided with childhood exposure to the campaigns rather than to pre-existing trends. Estimates suggest a substantial, though not predominant, role for malaria in explaining cross-region differences in income.” Bleakley 2010, p. 1.

  • 164

    We think that major economics journals would be less likely to publish studies finding that there was no impact of malaria eradication on economic outcomes. Relying only on these two studies could mean we are systematically overestimating the income effects in our analysis.

  • 165

    See this row in our analysis spreadsheet.

  • 166

    This adjustment is a subjective best guess which we have quantified in the following way:

    • Adjustment for study quality: Our best guess is that the pooled estimate from Bleakley and Cutler et. al. is overestimated because of the methodological concerns identified above. We apply a 40% adjustment to our initial estimate to account for this.
    • Adjustment for gender disparities: Both studies only find evidence of economic gains for men (Bleakley’s analytic sample does not include women, and Cutler et. al. find an effect of approximately 0 for women). We think it is plausible that averting malaria cases today would result in economic gains for women that Cutler et al. did not find (e.g., because of women’s higher labor force participation today than when the campaigns were conducted), but there may still be reasons why women would not see economic gains as large as men. To account for this, we apply a further adjustment of 75% to our estimate. This is based on a guess that the effect for women might be around half as large as for men, and women are approximately 50% share of the population.
    • (40% x 75%) - 100% = -70%. See this row in our analysis spreadsheet.

  • 167

    Note: we use an estimate of 0.65% increase in adult income per case averted for SMC programs, and 0.58% for another malaria program that we have funded, mass distribution of insecticide-treated nets (ITNs). This is because we have seen evidence that ITNs reduce malaria incidence (new cases) more than prevalence (the proportion of the population with detectable parasites, whether or not they have symptoms). More in our separate report on ITNs here.

  • 168

    See this section of our cost-effectiveness analysis for our full calculations. Our analysis consists of the following steps:

    The number of children who will benefit from SMC

    • Children aged 3 - 59 months: we use our estimates of the number of children treated per $1m spent by Malaria Consortium (e.g., ~176,000 in Burkina Faso).
    • Children aged 5 - 14 years: Our analysis assumes that children aged 0 - 5 will see larger benefits from SMC (because they are treated directly), but children aged 5 - 14 will also see some spillover benefits because of reduced malaria in the wider population. To quantify the number of children who will benefit from spillovers, we use the ratio of 5 - 14 year olds to <5 year olds in each country in our analysis, based on population estimates from the GBD Project. For example, we assume there are 2 children aged 5 - 14 for every 0 - 5 year old in Burkina Faso. This implies ~353,000 5-14 year olds will be exposed to SMC spillovers per $1m spent by Malaria Consortium in Burkina Faso (~176,000 x 2).

    Reduction in malaria cases from SMC

    • Malaria incidence: We use estimates of baseline malaria incidence in each location in our analysis from the GBD. These estimates are only available disaggregated into 1 - 4 year old and 5 - 14 year old age groups, and so we use incidence for 1 - 4 year olds as a proxy for children who receive SMC (in fact, children aged 3 - 59 months are eligible to receive SMC). We do not adjust these estimates, with the exception of a downward adjustment (varying by state) in Nigeria. This is because we think the GBD estimates might not account for a recent increase in insecticide-treated net campaigns in some states in Nigeria, and could therefore overestimate malaria burden. This is the same adjustment that we apply to malaria mortality in Nigeria (discussed above).
    • Impact of SMC on malaria cases: To estimate the number of cases that will be averted by SMC, we multiply malaria incidence for each age group by:

  • 169

    This is based on an assumption that people join the labor force on average at the age of 17. We also assume:

    • Children between the ages of 0 and 14 are the only group who will see income benefits from the long-run effects of reduced childhood exposure to malaria.
    • SMC averts more malaria cases in children under 5 (~70%) than in children ages 5-14. There are ~15 years between program implementation and starting work for children under-5, and ~7 years for children ages 5-14. Taking a weighted average implies an average gap of ~12.5 years before income benefits begin. See this row in our cost-effectiveness analysis.

  • 170

    Although this is a very rough guess, our cost-effectiveness analysis is not highly sensitive to the value we use. This is because we use a discount rate to discount the value of future economic benefits (as discussed in the prior footnote), meaning benefits in the distant future contribute less to our bottom line.

  • 171

    The spreadsheet references deworming because we created it to estimate the household multiplier for income gains from deworming. We have used the same assumptions for SMC.

  • 172

    This is because we think they would be challenging to model effectively or their likely effect is small enough that we do not expect explicit modeling to be worth the time investment required.

  • 173

    See this row of our cost-effectiveness analysis.

  • 174

    Because this estimate varies by location, it does not appear on our supplemental spreadsheet. See this row in our cost-effectiveness analysis for this estimate by location.

  • 175
    • Step one: Estimate the likely size of each effect (a rough subjective best guess).
    • Step two: Evaluate each effect using three criteria, each assigned a score of up to 3:
      • Can it be objectively justified (i.e., is there direct evidence for the effect)? (Where 0 is little evidence and 3 is strong direct evidence.)
      • How easy would it be to model? (Where 0 is impossible and 3 is simple.)
      • Consistency: is this effect included in our cost-effectiveness analysis for other programs? (Where 1 is no, 2 is partially, and 3 is yes.)
    • Step three: Convert the total three criteria score out of 9 into a percentage (e.g., a total score of 5 converts to a percentage of 50%).
    • Step four: Weight our estimate of each effect by this percentage to produce our overall adjustment (e.g., an effect size guess of 15% multiplied by a three criteria score of 60% = an adjustment of 9% (15% * 60% = 9%)).

    We use the method described for all supplemental intervention-level adjustments in our cost-effectiveness analysis for ITNs, with the exception of:

    • Our estimate of costs saved from averted malaria. We account for these savings using an explicit model, discussed in more detail here.
    • The adjustment for marginal funding going to lower-priority areas, which we estimate case-by-case for each country based on our understanding of how the National Malaria Program is likely to prioritize funding.

  • 176

    See this spreadsheet for more details on our reasoning for each adjustment.

  • 177

    “Cerebral malaria is the most severe neurological complication of infection with Plasmodium falciparum. With over 575,000 cases annually, children in sub-Saharan Africa are the most affected. Surviving patients have an increased risk of neurological and cognitive deficits, behavioral difficulties and epilepsy making cerebral malaria a leading cause of childhood neuro-disability in the region. The pathogenesis of neuro-cognitive sequelae is poorly understood: coma develops through multiple mechanisms and there may be several mechanisms of brain injury.” Idro et al. 2010.

  • 178

    Rough calculations underlying our estimate are available here. These estimate the years lost to disability (YLDs) as a proportion of the overall malaria disease burden, measured in DALYs (disability-adjusted life years).

  • 179

    “The best estimates of the causal contribution of malaria to anaemia in a particular setting come from randomized trials of malaria control interventions [19, 20]. A review of 29 community-based studies of insecticide-treated nets (ITNs), anti-malarial chemoprophylaxis, and insecticide residual spraying found that among children < 5 years exposed to between 1 and 2 years of malaria control, mean relative risk for a haemoglobin concentration < 11 g/dL was 0.73 (95% CI 0.64–0.81), and for a haemoglobin < 8 g/dL was 0.40 (95% CI 0.25–0.55) compared with the control groups not exposed to these malaria interventions.” White 2018, 371.
    The Pryce, Richardson, and Lengeler 2018 meta-analysis we use to estimate the impact of insecticide-treated nets (ITNs) on mortality also finds an effect of ITN distribution on hemoglobin levels (which determine anemia).
    “Five trials reported the mean haemoglobin in ITN and no-nets arms. Pooled analysis of the trials showed that ITNs were associated with a mean difference of a 1.29 increase in percentage PCV (95% CI 0.42 to 2.16; 5 trials, 11,489 participants, Analysis 1.9).” Pryce, Richardson, and Lengeler 2018, p. 16.

  • 180

    Note that reduced anemia is one of the proposed mechanisms for malaria income effects, which we model separately. This adjustment is intended to capture the shorter-term benefits of averted anemia separately from its possible role in income effects.

  • 181

    See the specific intervention tabs of this spreadsheet for a comparison.

  • 182

    See this row in our cost-effectiveness analysis.

  • 183

    As of December 2023, these are the Against Malaria Foundation, Malaria Consortium, Helen Keller International, and New Incentives. See our top charities page here.

  • 184

    Our model estimates a 14% upward adjustment for insecticide-treated nets (here), a 20% upward adjustment for seasonal malaria chemoprevention (here) and a 21% upward adjustment for vitamin A supplementation (here).

  • 185

    We use state-level data on malaria burden for Nigeria (and Mozambique, although we do not focus on Mozambique in this report).

  • 186

    “From 1995 onwards, coinciding with the results from the main African ITN trials, a vigorous debate arose about the possibility that the short-term mortality improvements observed in trials of 1–2 years' duration could be offset by increased mortality at older ages — a "delayed mortality" effect (4, 5). The underlying hypothesis was that immunity to malaria would develop more slowly under reduced transmission, leading to a longer period of susceptibility. No direct evidence was available at the time either to support or to refute this hypothesis.” Lengeler 2004.

  • 187

    “Three studies continued to monitor children for a full transmission season after IPTc was stopped. There was no observed rebound increase in malaria in children who had received the intervention compared to controls (2299 participants, three trials; Analysis 1.1).” The three studies are Cisse 2006, Dicko 2008 and Kweku 2008. Meremikwu et al. 2012, pgs. 12, 40.

  • 188

    “SP and amodiaquine (AQ) are used in combination (as SP/AQ) for SMC. It is very likely that if SP/AQ is used for SMC on a wide scale, resistance will eventually develop (as has been the case with nearly all antimalarial drugs used in the past).
    Resistance to SP/AQ would require mutations granting resistance to both SP and AQ, since these operate via independent mechanisms. If a parasite is resistant to either SP or AQ but still fully sensitive to the other, this is unlikely to reduce SP/AQ's overall efficacy.” GiveWell's non-verbatim summary of a conversation with Professor Sir Brian Greenwood, January 4, 2017.

  • 189

    “Recently, the median time for resistance to a given drug to develop has been roughly 10 to 15 years; for instance, resistance to sulfadoxine-pyrimethamine (SP) resistance took roughly 10 years to emerge in Southeast Asia (where SP was being used as a mainline treatment for malaria).” GiveWell's non-verbatim summary of a conversation with Professor Sir Brian Greenwood, January 4, 2017.
    “Professor Greenwood’s best guess is that SP/AQ will still be working, at least partially, in 5-10 years’ time, but this is quite uncertain (it is possible, though unlikely, that an unanticipated type of resistance mutation could appear at any time).” GiveWell's non-verbatim summary of a conversation with Professor Sir Brian Greenwood, January 4, 2017.

  • 190

    “SP has been used alone for intermittent preventive treatment in pregnancy (IPTp) for roughly 20 years. This is likely to have contributed to resistance to SP developing in Tanzania and East Africa, and SP is no longer very effective in those places for IPTp. However, it is still effective in much of the rest of Africa. A mutation associated with a particularly high level of SP resistance that appeared in Southeast Asia (the DHFR 164 mutation) also appeared in Uganda but has not spread in Africa.” GiveWell's non-verbatim summary of a conversation with Professor Sir Brian Greenwood, January 4, 2017.

  • 191

    “AQ has relatively little history of resistance, compared to SP. Some resistance has been reported, but overall efficacy has remained high despite extensive use in West Africa for 40 years.” GiveWell's non-verbatim summary of a conversation with Professor Sir Brian Greenwood, January 4, 2017.

  • 192

    “While Professor Greenwood thinks it is likely that SP/AQ will maintain its effectiveness for up to 10 years, if resistance to SP/AQ does develop, dihydroartemisinin-piperaquine (DHAPQ) could be used to replace its role in prevention. However, there is some risk that broad use of an artemisinin-based drug like DHAPQ for prevention could encourage artemisinin resistance, e.g., by encouraging an artemisinin-resistant parasite that already exists in Cambodia to emerge in Africa. While Professor Greenwood would be concerned about using artemisinin in prevention, he would not rule it out as an option, and other experts might disagree about the level of risk.” GiveWell's non-verbatim summary of a conversation with Professor Sir Brian Greenwood, January 4, 2017.

  • 193

    This analysis was based on (at the time) unpublished results that were subsequently published in the ACCESS-SMC Partnership 2020 study (the main findings from our analysis that we cite here are the same in the ACCESS-SMC Partnership 2020 study, although we haven’t investigated to check if every claim

    from the results we reviewed are identical). We cite ACCESS-SMC Partnership 2020 in this report.

  • 194

    “In children younger than 5 years…The prevalence of the quintuple mutation associated with resistance to sulfadoxine–pyrimethamine (triple mutation in pfdhfr with pfdhps-437Gly and pfdhps-540Glu) was 0.4% (0.2–0.8) in 2016 and 0.7% (0.3–1.5) in 2018 (prevalence ratio 1.8 [0.7–5.0). In the 10–30 age group, the corresponding estimates for the combination of pfcrt-CVIET, pfmdr1-86Tyr and pfmdr1-184Tyr were 0.7% (0.4–1.2) in 2016 and 0.4% (0.1–0.8) in 2018 (prevalence ratio 0.5 [0.2–1.2]), and for the quintuple mutation (triple mutation in pfdhfr with pfdhps-437Gly and pfdhps-540Glu), 0.2% (0.1–0.5) in 2016 and 1.0% (0.6–1.6) in 2018 (prevalence ratio 4.8 [1.7–13.7]).” ACCESS-SMC Partnership 2020, pg. 1836.

  • 195

    “In children younger than 5 years, the combination of the pfcrt-CVIET (amino acid positions 72–76), pfmdr1- 86Tyr, and pfmdr1-184Tyr variants, associated with resistance to amodiaquine, was found with a prevalence of 1.3% (95% CI 0.9–2.0) in 2016 and 0.5% (0.2–1.4) in 2018 (prevalence ratio 0.4 [0.1–1.1])...In the 10–30 age group, the corresponding estimates for the combination of pfcrt-CVIET, pfmdr1-86Tyr and pfmdr1-184Tyr were 0.7% (0.4–1.2) in 2016 and 0.4% (0.1–0.8) in 2018 (prevalence ratio 0.5 [0.2–1.2]).” ACCESS-SMC Partnership 2020, pg. 1836.”

  • 196

    We use an adjustment of -10% for this parameter outside the Sahel. This is because we assume that the risk of SMC campaigns accelerating drug resistance is higher in areas with high baseline resistance. See this row in our cost-effectiveness analysis.

  • 197

    Burkina Faso: As of 2022, GiveWell funding supports around half of the total SMC program in Burkina Faso (2m children targeted out of a total target population of 4.4m). See Malaria Consortium, 2022 SMC philanthropy report, pg. 36. In addition, there appears to be moderate variation in malaria prevalence across locations in Burkina Faso (see the map here). We interpret this to imply a significant scope to prioritize locations in the absence of GiveWell funding. We account for this with a relatively large adjustment (-15%).
    Togo: While SMC in Togo was fully funded in 2020 and 2021, the national malaria program in Togo has a history of choosing to implement fewer SMC cycles across all eligible areas rather than prioritizing higher-risk areas when faced with underfunding. See Malaria Consortium, 2021 SMC philanthropy report, pg. 35. In addition, there appears to be relatively little variation in malaria rates across the regions in Togo where SMC is delivered (Centrale, Kara, and Savanes see the map here). Together, we think there is less scope to prioritize locations in the absence of GiveWell funding and the Togolese government may deliver a lower number of SMC cycles rather than prioritize by location. We therefore use a relatively lower adjustment (-5%) compared to Burkina Faso.
    See this row of our cost-effectiveness analysis for more details on our reasoning.

  • 198

    See Malaria Consortium, 2022 SMC philanthropy report, pg. 47.

  • 199
    • The two main studies we reviewed are:
      • NDiaye et al. 2016, a paper on the safety of SMC in 776,191 children under 10 in Senegal.
      • A 2017 progress report for the ACCESS-SMC project, which focuses on efforts to improve drug safety surveillance systems in targeted countries, and reports adverse event rates captured via the improved surveillance systems.
    • In NDiaye et. al. 2016, the researchers did not attribute any deaths to SMC. One severe (symptoms preventing usual activities) and no serious (potentially life-threatening) adverse events were attributed to SMC. In the ACCESS-SMC report, among 95 million SMC doses administered, 40 serious adverse events (0.000042 percent), and no “drug-related deaths”, were reported.
      • “investigation of cause of death using verbal autopsies did not suggest that any deaths were related to SMC drugs.” NDiaye et al. 2016, pg. 12.
      • “The review panel considered that only the extra-pyramidal syndrome was likely to be related to study drugs... No serious adverse events attributable to SMC were reported after giving 776,191 documented treatments.” NDiaye et al. 2016, pg. 8.
    • ACCESS-SMC, Progress Update, Pharmacovigilance: Monitoring SMC drug safety, 2017, Table 1.

  • 200

    See the “Back-of-the-envelope calculation” section of this document.

  • 201

    The substudy within NDiaye 2016 recorded and classified causes of all deaths in a catchment area for 12 health posts via verbal autopsies. We would expect this method to be more reliable at detecting serious adverse events than other methods because it does not rely on caregivers or clinicians to link an adverse event to SMC. “In twelve health posts, purposively selected to be representative of the study area, all deaths under 10 years of age in the catchment population were investigated by verbal autopsy.” Ndiaye et al. 2016.

  • 202

    Using the most pessimistic assumptions based on Ndiaye et al. 2016, we estimated:

    • A rate of 2.9 deaths per 100,000 children treated with SMC.
      • In 2010, when the most strengthening of the verbal autopsy approach had been implemented in Ndiaye et al. 2016, there were 11 deaths in SMC areas. None of the deaths were attributed to serious adverse events, though they did include 3 from “other causes”., The numerator for the risk of death is 3 deaths from the 10 health posts receiving SMC that had verbal autopsy in 2010. For the denominator, 468,067 children received a first dose of SMC in 2010 across 45 health posts receiving SMC in 2010. Assuming an equal population of children per 45 health post, the rate would by (3/((468067/45)*10))*100000=2.9. See Ndiaye et al. 2016, tables 2 and 3.
    • An estimate (based on GiveWell’s cost-effectiveness analysis for Malaria Consortium’s SMC program in April 2023) that 497 deaths would be averted by SMC for every 233,480 children reached, or (497/233480)*100000 = 213 per 100,000 children treated.
    • This implies that, under the most pessimistic assumptions, per 100,000 children treated, SMC would result in 2.9 total deaths from serious adverse events, or 1.4% of the mortality reduction of 213 deaths.
    • This rate is pessimistic given it assumes that all 3 deaths classified as “other causes” in the Ndiaye et al. 2016 substudy were actually due to SMC, whereas study investigators did not classify any of the deaths as due to serious adverse events. If none of the deaths were due to serious adverse events, that would imply a rate of 0%, implying a range of effects from 0% to 1.4% of the mortality benefit.
    • Given the very small number of deaths included in the numerator and the likely challenges of conducting a high-quality verbal autopsy study, we do not think these estimates are very precise, but rather suggest a rough order of magnitude for potential adverse effects, if they exist.

    GiveWell, SMC Adverse Events Analysis, June 2023 (unpublished)

  • 203

    See this section of the report for further details on each stage of our analysis.

  • 204

    See this section of our separate report on Malaria Consortium for more details on these surveys and their methodology.

  • 205

    “Two trials distributed and promoted the use of ITNs to both the intervention and control groups (Dicko 2011 and Konate 2011).” Meremikwu et al. 2012, pg. 13.
    “Two trials distributed and promoted the use of ITNs to both the intervention and control groups (Dicko 2011 and Konate 2011). Despite ITN use being reported as >90% in both treatment arms, IPTc had high protective efficacy against both clinical malaria (rate ratio 0.22, 95% CI 0.13 to 0.38; 5964 participants, two trials; Analysis 2.1) and severe malaria (rate ratio 0.27, 95% CI 0.10 to 0.76; 5964 participants, two trials; Analysis 1.2).” Meremikwu et al. 2012, pg. 13.

  • 206
    • "We are only permitted to procure products from suppliers that meet WHO stringent guidelines for pre-qualification of malaria drugs to ensure high-quality and efficacy of SP and AQ." Malaria Consortium, comments on a draft of GiveWell’s report on Malaria Consortium’s SMC program, October 2017 (unpublished).
    • "...there is not yet any other supplier who manufactures both products and packages them in a 1SP 3AQ treatment slide combination, irrespective of WHO guidelines. Unpackaged products could be used but it is significantly more inefficient with a much higher risk of error." Malaria Consortium, comments on a draft of GiveWell’s report on Malaria Consortium’s SMC program, October 2019 (unpublished).

  • 207

    Malaria Consortium, comments on a draft of report on Malaria Consortium’s SMC program, October 2020 (unpublished).

  • 208

    See this row in our cost-effectiveness analysis. Note, we also consider other grantee-level factors in our analysis that we do not discuss in this report:

    • Ineffective goods (whether the program might use goods that are out of date, poorly made, or otherwise ineffective)
    • Goods purchased and left in storage until they expire
    • Non-funding bottlenecks (i.e., a grantee experiences problems that mean it cannot spend its allocated funds)

    We do not discuss these factors in this report because we assign a 0% downward adjustment in each case. See this section of our cost-effectiveness analysis for more details on our reasoning. We do discuss our 0% adjustment for double treatment, because estimating the proportion of children who would receive a program if it were not for GiveWell funding is a core assumption informing our funding decisions.

  • 209

    We also consider:

    • Ineffective goods (whether the program might use goods that are out of date, poorly made, or otherwise ineffective).
    • Goods purchased and left in storage until they expire.

    We exclude these factors from our discussion here since we assign a 0% adjustment in both cases, and we have considered double treatment in more detail. See our reasoning here for more details.

  • 210

    "By 2014, eight Sahelian countries were implementing SMC, reaching about 2.5 million children. . . . Between 2015 and 2017, the Achieving Catalytic Expansion of SMC in the Sahel (ACCESS-SMC) project, funded by Unitaid and led by Malaria Consortium, accelerated the scale-up of SMC across the region. At its peak, ACCESS-SMC reached close to seven million children in Burkina Faso, Chad, The Gambia, Guinea, Mali, Niger and Nigeria. . . . In 2021, SMC was implemented in 15 countries, targeting around 45 million children, up from 33.5 million children in 13 countries the year before. The increase in targeted children was primarily due to the expansion of SMC to new states in Nigeria, while the increased number of countries implementing SMC was a result of SMC pilot projects in Mozambique and Uganda — the first countries outside of the Sahel to test the use of SMC as a malaria prevention strategy. According to unpublished data compiled by the SMC Alliance, a workstream under the RBM Partnership to End Malaria’s Country/Regional Support Partner Committee, the number of children targeted globally was just under 48 million in 2022." Malaria Consortium, 2022 SMC philanthropy report, pg. 12.

  • 211
    • “Weekly administration of SP or AQ for chemoprophylaxis was discontinued because of the potential for serious adverse drug reactions. As there are limited data on cumulative toxicity of AQ, it is not recommended to shorten the interval between cycles to less than 28 days. National malaria programmes should not deploy antimalarials containing SP or AQ for first- or second-line treatment in areas implementing SMC. It is important to engage with the private sector to prohibit the sale of medicines containing SP or AQ in areas 5 implementing SMC. This will prevent potential repeated exposure to SP or AQ at short intervals, which may cause serious toxicity.” WHO, SMC field guide, 2nd ed., 2023, p. 4.
    • Our understanding that this decision was taken partly in order to minimize the growth of resistance is based on feedback from Dr. André Tchouatieu,Director, Access & Product Management at Medicines for Malaria Venture. Dr. André Tchouatieu, “GiveWell SMC report: Feedback from an independent review”, October 2023 (unpublished).

  • 212

    See this section in our cost-effectiveness analysis.

  • 213

    These estimates feed into our cost-effectiveness analysis as part of our estimates of the cost per cycle of SMC delivered. Note that we separately account for the risk that children vomit the first day of SMC back and therefore benefit less from SMC with a -4% adjustment, discussed above.

  • 214

    See this section of our separate report on Malaria Consortium for further discussion.

  • 215

    See this cell in our cost-effectiveness analysis for further details.

  • 216

    See this cell in our cost-effectiveness analysis for further details.

  • 217

    This estimate suggests that activities funded by the Burkina Faso government are approximately 1.5 times as cost-effective as direct cash transfers. We estimate that each US dollar donated to GiveDirectly’s direct cash transfer program generates 0.00335 units of value. 0.005 / 0.00335 = ~1.5.

  • 218

    This estimate suggests that activities funded by the Global Fund and/or PMI are approximately 4.5 times as cost-effective as direct cash transfers. We estimate that each US dollar donated to GiveDirectly’s direct cash transfer program generates 0.00335 units of value. 0.015 / 0.00335 = ~4.5.

  • 219

    See this row in our cost-effectiveness analysis. When rounded, this appears as 0%.

  • 220

    See this section for more details on how we reach this estimate.

  • 221

    We estimate that each US dollar spent by a domestic government in a country receiving SMC generates 0.005 units of value, compared to 0.105 for Malaria Consortium’s spending on SMC in Burkina Faso. 0.005 / 0.105 = ~5%.
    For details of how we estimate these values, see this section.

  • 222
    • We estimate that each US dollar spent by the Burkina Faso government generates 0.005 units of value if used for other activities.
    • In total, we think that each $1m spent by Malaria Consortium causes the Burkina Faso government to incur ~$118,000 in in-kind costs.
    • This implies that shifting these resources away from other activities results in ~600 units of value being lost. (~118,000 x 0.005) = ~600.
    • See this row in our cost-effectiveness analysis.

  • 223

    See this row in our cost-effectiveness analysis.

  • 224

    299 / ($1,000,000 x 0.105) = ~0.3%. See this row in our cost-effectiveness analysis. Due to rounding, this value appears as 0%.

  • 225

    See this row in our cost-effectiveness analysis.

  • 226

    Our understanding is that the Global Fund and PMI are the largest external malaria funders in sub-Saharan Africa and therefore the most likely to replace GiveWell’s funding for Malaria Consortium.
    We revisit these probabilities for each grant we make. For Burkina Faso, we most recently estimated these probabilities as part of a 2023 grant. See our grant page here.

  • 227

    See these rows in our cost-effectiveness analysis.

  • 228

    In this scenario, we do not think that Malaria Consortium’s funding is causing more SMC to be delivered, because other funders would have replaced its spending in any case.

  • 229

    We roughly estimate that the Global Fund and PMI’s spending on other activities generates 0.015 units of value per dollar, compared to 0.105 for Malaria Consortium’s spending in Guinea. 0.015 / 0.105 = ~14%, or ~1/7. See this section of the report for how we estimate the 0.015 figure.

  • 230

    We estimate that each dollar spent by the Global Fund and/or PMI generates 0.015 units of value if used for other activities and that Malaria Consortium generates 0.105 units of value per dollar spent on SMC in Burkina Faso. Per $1m, the Burkina Faso government replacing Malaria Consortium’s spending would generate ~105,000 units of value on SMC ($1m x 0.105) and lose ~15,000 units of value ($1m x 0.015) that would have been spent on other programs. ~105,000 - ~15,000 = ~90,000. See this row in our cost-effectiveness analysis.

  • 231

    -45,000 / 105,000 = ~-43%. See this row in our cost-effectiveness analysis.

  • 232

    See these rows in our cost-effectiveness analysis.

  • 233

    See these rows in our CEA for scenario 2 and these rows for scenario 4. Note: we also gave probabilities for two other scenarios:

    • Scenario 1: The Burkina Faso government would replace Malaria Consortium’s costs.
    • Scenario 3: The Burkina Faso government would spend the same amount and the program would be smaller.

    We exclude these scenarios from the summary here because we assigned a 0% probability to each scenario, so they do not affect our bottom line. See this section and this section in our cost-effectiveness analysis for scenarios 1 and 3, respectively.

  • 234

    “In 2019, Burkina Faso achieved 100 percent geographical SMC coverage, with all of the country’s 70 health districts reached. Full geographical coverage has been maintained in subsequent years.” Malaria Consortium, 2022 SMC philanthropy report, p. 36.

  • 235

    "In this process, the Global Fund, PMI and Malaria Consortium also agreed to split health districts supported by the World Bank until 2019 and covered by the Global Fund in 2020. As a result of the changes, Malaria Consortium’s support will increase to 27 health districts." Malaria Consortium, 2020 SMC philanthropy report, p. 12.

  • 236

    See Malaria Consortium, 2022 SMC philanthropy report, p. 36, table 4.

  • 237

    The Global Fund underwent its latest three-year funding “replenishment” in late 2022 (for interventions to be delivered in 2024-2026). The total value of the replenishment was $15.7 billion, compared to a target of at least $18 billion. This represents a fairly stable level of funding compared to the previous replenishment (a 3.3% nominal increase in the overall funding allocated to countries). The funding available for malaria was also stable.
    "The Global Fund’s Seventh Replenishment is the world’s opportunity to rise to the challenge and take bold action to protect everyone, everywhere from the deadliest infectious diseases. Our target is to raise at least US$18 billion. This is the minimum required to get the world back on track toward ending HIV, TB and malaria, to build resilient and sustainable systems for health and strengthen pandemic preparedness, making the world more equitable and safer from future threats." Global Fund, "Seventh replenishment: Fight for what counts"
    “The Board of the Global Fund to Fight AIDS, Tuberculosis and Malaria welcomed the Seventh Replenishment outcome of US$15.7 billion [ download in English ] during a 3-day meeting this week in Geneva.” Global Fund, "Global Fund board hails record-breaking seventh replenishment final outcome of US$15.7 billion," 2022.
    $4.2 billion was allocated to countries for malaria interventions, compared to $4.0 billion in the previous replenishment. See GiveWell’s summary here.

  • 238

    We factored this into our analysis of funging probabilities here. This was one factor in reducing our best guess of the chance that the Global Fund / PMI would replace Malaria Consortium’s funding from 60% (here) to 50%.

  • 239

    See this row in our cost-effectiveness analysis.

  • 240

    See this row in our cost-effectiveness analysis. The value is derived from our calculations in our analysis of the counterfactual value of other actors' spending spreadsheet.

  • 241

    0.005 / 0.105 = ~5%, or ~1/20.

  • 242

    See this cell in our supplementary analysis.

  • 243

    Health

    • We use data from the Uganda National Health Expenditure Accounts in 2013-14 (available here) as a proxy for how domestic governments in low-income countries allocate their health spending. In that year, approximately 30% of the health spending was allocated to HIV/AIDS, 20% to malaria, and the rest to other programs (see this column in our accompanying spreadsheet).
    • We roughly guess the $ value required to save a life for eight of the largest categories of spending (e.g., we guess $30,000 of spending on HIV/AIDS programs is required to save one life (here), compared to $10,000 for malaria, here). As a simplifying assumption, we divide each category of spending into spending intended to save adult lives (e.g., HIV/AIDS), or spending intended to save under-five lives (e.g., malaria) (see this column in our accompanying spreadsheet). This allows us to calculate the weighted average cost required to save one child’s life ($12,254) and one adult life ($30,000).
    • Finally, we translate these figures into standardized units of value using GiveWell’s moral weights for the value of averting a child death (116 units) and adult death (73 units). We also use our estimates of the proportion of spending allocated to programs intended to save adult lives (55%) and under-five lives (45%). This results in an overall weighted average of 0.0056 units of value per $ spent on health programs by domestic governments. See this row for our calculations.

    Education

    • As with health, we use data from Uganda as a proxy for how domestic governments in low-income countries allocate their education spending. Specifically, we rely on World Bank 2014 data that 7% of Uganda’s GDP in 2014 went towards primary education, 3% went towards secondary education and 2% went towards tertiary education (here). We then rescale these figures, resulting in estimates that 59% of education spending in Uganda goes towards primary education, 25% towards secondary education and 16% towards tertiary education (here).
    • As a simplifying assumption, we assume that the primary benefit of education spending is that it increases long-run income and consumption. We benchmark our estimates to our estimates of the impact of GiveDirectly’s unconditional cash transfer program (which we think also increases income and consumption). Specifically, we roughly guess that primary education spending is 100% as valuable as GiveDirectly’s program, secondary education is 70% as valuable and tertiary education is 50% as valuable (here).
    • We take a weighted average of these values, using the % of government spending on each category of education as weights. This results in an overall estimate of 0.0028 units of value per $ spent on education programs by domestic governments. See this row for our calculations.

    Social security
    We use a similar approach to estimating the value of social security spending as education spending. Specifically:

    • We use data from Uganda as a proxy for how domestic governments in low-income countries allocate their social security spending. We use World Bank 2014 estimates that social security spending accounted for 0.62% of Uganda’s GDP (here). Within this, 0.29% of GDP is in-kind spending, 0.1% is social pension spending, 0.1% is conditional cash transfers, and 0.05% is other cash transfers (here).
    • We roughly guess the value generated by each of these four largest categories of social security spending, excluding the smaller categories. As with education spending, we use the simplifying assumption that the primary benefit of social security spending is that it increases income and consumption. Benchmarking to GiveDirectly’s cash transfer program, we guess that cash transfers generate 100% of the value of GiveDirectly’s program and in-kind and social pension spending each generate 70% of the value (here).
    • We take a weighted average of these values, using the % of government spending on each category of social security spending as the weights (re-scaled to ignore smaller categories). This results in an overall estimate of 0.0026 units of value per $ spent on social security programs by domestic governments. See this row for our calculations.

  • 244

    See this row in our supplementary analysis.

  • 245

    See this row in our cost-effectiveness analysis. Note: although we focus on the Global Fund in our analysis, we also factor in the possibility that PMI might replace Malaria Consortium’s funding for SMC if GiveWell did not provide it. This means our analysis relies on the assumption that the counterfactual value of PMI’s spending is the same as the counterfactual value of the Global Fund’s spending.

  • 246

    0.015 units of value for Global Fund spending on activities other than SMC / 0.105 units of value generated by Malaria Consortium spending on SMC = ~14%, or 1/7.

  • 247

    HIV (calculated on this sheet)

    • As a simplifying assumption, we assume that all Global Fund spending on HIV will be used for antiretroviral therapy (ART). This is because our understanding is that this is the dominant focus of the Global Fund’s HIV programming.
    • We estimate that spending on ART is 1.4x as cost-effective as direct cash transfers. This is based on the following estimates (drawn from published literature):
      • ART costs $540 per person to deliver per year
      • Treating someone with HIV with ART extends their remaining life expectancy by 18 years, on average (from 9 to 27 years).
    • We also use GiveWell’s standard discount rate of 4% to discount the costs of ART over time also, the standard GiveWell moral weight for averting one year of life lived with disease/disability (YLD) (2.3), and a -20% adjustment to account for lost effectiveness due to possibility of poor program implementation.

    This produces an overall estimate that each $ spent on ART is 1.1x as cost-effective as direct cash transfers. See these columns for our calculations and this column for references from the secondary literature.
    Tuberculosis (calculated on this sheet)

    • As a simplifying assumption, we assume that all Global Fund spending on TB will be used for TB screen-and-treat programs.
    • We estimate the cost-effectiveness of screen-and-treat based on GiveWell’s (unpublished) analysis of IRD Global’s TB screen-and-treat program. Our analysis includes the following key estimates:
      • It costs $8 to treat one person with screen-and-treat.
      • The health burden of tuberculosis is 84% as high in areas supported by the Global Fund as it is in areas supported by IRD Global’s program.
      • Screen-and-treat has a 0.12% probability of averting a death.
      • The additional benefits of averted morbidity from TB are 4% as high as the benefits from averted mortality.
    • We use a moral weight of 60 for the value of averting one person’s death with screen-and-treat. This is lower than GiveWell’s standard value of 70 for averting the death of an adult, because the age profile of people who die from TB is older and GiveWell’s moral weights assign higher values to averting deaths at younger ages.
    • As with HIV, we use a -20% adjustment to account for lost effectiveness due to possibility of poor program implementation.

    This produces an overall estimate that each $ spent on screen-and-treat is 3.2x as cost-effective as direct cash transfers. See these columns for our calculations and this column for references from the secondary literature.

  • 248

    The programs and their cost-effectiveness estimates are (see here):

    • Malaria vector control: LLINs (campaign): 9.7x as cost-effective as direct cash transfers
    • Malaria vector control: LLINs (continuous): 5.6x as cost-effective
    • Malaria vector control: Indoor residual spraying (IRS): 3.2x as cost-effective
    • Malaria vector control: Other: 0.5x as cost-effective
    • Malaria case management (CM): 2.5x as cost-effective
    • Malaria specific prevention interventions (prophylaxis): 13.6x as cost-effective
    • Malaria (other): 0.5x as cost-effective

    We estimate the value of each program in a separate sheet in this spreadsheet.

  • 249

    We use confidential data from the Global Fund’s Unfunded Quality Demand (UQD) register to estimate how reallocated funding in the Global Fund’s portfolio is most likely to be used. The UQD register contains funding for programs that national governments would ideally like to deliver, but cannot because they do not have sufficient funding from their Global Fund grant. These programs may subsequently be funded through savings from other programs, reallocation of funding in the Global Fund’s wider portfolio, or private sector contributions.
    We use data on the proportion of different malaria programs were funded from the UQD register between 2017 and 2019, and what proportion of these programs were funded by either savings from savings on other malaria programming ("grant savings") or reallocated funding within the Global Fund’s wider portfolio ("portfolio optimization").

  • 250

    GiveWell's non-verbatim summary of conversations with Malaria Consortium staff, November 7 and November 9, 2016 (unpublished).
    "The ACCESS-SMC partnership:

    • Malaria Consortium is leading the ACCESS-SMC project, tracking its impact, managing the procurement of SMC drugs and supporting malaria control programmes to implement SMC in Burkina Faso, Chad and Nigeria.
    • Catholic Relief Services is the lead-subrecipient and contributing to tracking the reach and impact of the project and supporting malaria control programmes to implement SMC in Guinea, Mali, Niger and The Gambia." ACCESS-SMC, Fact sheet, 2016, pg. 2.

  • 251
    • See ACCESS-SMC, Evaluation of Seasonal Malaria Chemoprevention. Our understanding is that this study was later published as part of the ACCESS-SMC 2020 paper, and in Cairns et. al. 2021, all of which report case-control studies in the same countries in the same years, and all of which report almost identical results. We haven’t tried to understand any differences between these studies, and we have only reviewed the findings from the evaluation shared by Malaria Consortium in detail.
    • “Case-control studies in five countries in 2015 and 2016 were used to measure the protective efficacy of each SMC monthly treatment. 820 cases and 1,637 controls were recruited in 2015, and 1,433 cases and 2,867 controls in 2016. SMC was associated with an 89% reduction in malaria incidence for 4 weeks after treatment, and 62% from 5 to 6 weeks after treatment, compared with children who had not received SMC or whose last dose was more than 6 weeks before.” ACCESS-SMC, Annex III - ACCESS-SMC Evaluation (preliminary reports), pg. 2.
    • “Effectiveness of SMC treatments (in terms of the percentage reduction in clinical malaria incidence in the 28 days and 29–42 days after administration of the first daily dose of SMC each month) was estimated in five countries during the 2015 (The Gambia and Mali) and 2016 (Burkina Faso, Chad, The Gambia, Mali, and Nigeria) transmission seasons with case-control studies.” ACCESS-SMC Partnership 2020, pg. 1832.
    • “SMC treatment was associated with a protective effectiveness of 88.2% (95% CI 78.7–93.4) over 28 days in case-control studies (2185 cases of confirmed malaria and 4370 controls). ACCESS-SMC 2020, abstract.
    • “Case–control studies were carried out in Mali and The Gambia in 2015, and in Burkina Faso, Chad, Mali, Nigeria, and The Gambia in 2016… In all 7 individual case–control studies, a high degree of personal protection from SMC against clinical malaria was observed, ranging from 73% in Mali in 2016 to 98% in Mali in 2015. The overall OR for SMC within 28 days was 0.12 (95% CI: 0.06, 0.21; p < 0.001), indicating a protective effectiveness of 88% (95% CI: 79%, 94%). Effectiveness against clinical malaria for SMC 29–42 days ago was 61% (95% CI: 47%, 72%).” Cairns et al., 2021.

  • 252
    • "Data for impact indicators outlined in Table 1 were abstracted from the national HMIS over the years of 2013 to 2018 for Burkina Faso and Chad, and 2017 to 2018 for Nigeria and were retrospectively analysed." Malaria Consortium, 2019 impact report, pg. 4.
    • "To generate estimates of the level of impact of SMC, a Poisson regression model was fitted to the number of cases confirmed by RDT or microscopy reported to health facilities during the SMC distribution months (August, September, October, November) between 2013 and 2018 (or 2017-2018 for Nigeria)." Malaria Consortium, 2019 impact report, pg. 6.

  • 253

    “The reduction in the number of malaria outpatient cases, inpatient cases, and malaria deaths in hospital in children younger than 5 years associated with introduction of SMC, according to cases reported in HMIS databases and individual patient data, was estimated with a difference-in-differences approach (appendix pp 19–21). Individuals aged 5 years and older were the control age group, with data on the same age groups in areas that did not introduce SMC as additional controls. Poisson regression models were fitted to the data on numbers of cases before and during the intervention period in Stata. This approach corrected for changes in testing rates and use of insecticide-treated bednets, which increased in some countries during the study period but changed similarly in both age groups, and for the effect of removal of patient charges in Burkina Faso from 2016 (appendix pp 11–13).” ACCESS-SMC 2020, pg. 1833.

  • 254

    “The reduction in the number of malaria outpatient cases, inpatient cases, and malaria deaths in hospital in children younger than 5 years associated with introduction of SMC, according to cases reported in HMIS databases and individual patient data, was estimated with a difference-in-differences approach (appendix pp 19–21). Individuals aged 5 years and older were the control age group, with data on the same age groups in areas that did not introduce SMC as additional controls…
    The Gambia and Burkina Faso had established District Health Information System 2 (DHIS2) databases before SMC scale-up and these national databases were used for analyses of the effect of SMC on the number of reported outpatient malaria cases, the number of reported severe (hospitalised) cases, and the number of deaths in district hospitals attributed to malaria. In the other five countries, data on confirmed outpatient cases were collected from outpatient clinics. In each country, facilities were selected (~30 per country) that had used parasitological confirmation of malaria cases for at least one year before introduction of SMC; had retained clinic registers; and were in areas where SMC was to be delivered via ACCESS-SMC starting in 2015 or 2016, or would not have implemented SMC by 2016.” ACCESS-SMC Partnership 2020, pg. 1833.

  • 255

    “Over 2015–16, the estimated reduction in confirmed malaria cases at outpatient clinics during the high transmission period in the seven countries ranged from 25.5% (95% CI 6.1 to 40.9) in Nigeria to 55.2% (42.0 to 65.3) in The Gambia.” ACCESS-SMC 2020, abstract, and pg. 1836, table 2.

  • 256

    “In Burkina Faso and The Gambia, implementation of SMC was associated with reductions in the number of malaria deaths in hospital during the high transmission period, of 42.4% (95% CI 5.9 to 64.7) in Burkina Faso and 56.6% (28.9 to 73.5) in The Gambia.” ACCESS-SMC Partnership 2020, abstract.

  • 257

    “The surveys also showed that a high percentage of children aged 6–7 years received SMC (53.0% [95% CI 48.7–57.3] of children surveyed in 2015 [n=1695] and 62.4% [55.7–69.1] in 2016 [n=2062] were treated at least once).” ACCESS-SMC Partnership 2020, pg. 1835.

  • 258

    This understanding is based on: David Schellenberg, Professor of Malaria and International Health at the London School of Hygiene and Tropical Medicine, email to Givewell, November 6, 2023 (unpublished).

  • 259

    “ In 2015, among eligible children, mean coverage per month was 76.4% (95% CI 74.0–78.8), and 54.5% children (95% CI 50.4–58.7) received all four treatments. Similar coverage was achieved in 2016 (74.8% [72.2–77.3] treated per month and 53.0% [48.5–57.4] treated four times).” ACCESS-SMC Partnership 2020, abstract.

  • 260

    See this section of the report.

  • 261

    "Efficacy of SMC treatments will be measured using the case control approach. Malaria cases, and controls who do not have malaria, will be recruited concurrently, and the dates of the doses of SMC they received noted. The efficacy of SMC can then be calculated as a function of the time since treatment using case-control analysis. It is essential that dates of SMC doses are accurately documented, and that malaria cases are parasitologically confirmed. Controls will be selected from the community, in the neighbourhood where the case lived at the time they had malaria. Trained fieldworkers will collect information about cases and controls, and make home visits to record bednet use and other household factors that may act as confounders. Microscopy will be used to confirm cases and to measure parasite density. Controls will be confirmed to be negative for P.falciparum, by RDT." ACCESS-SMC, Evaluation of Seasonal Malaria Chemoprevention, pg 10.

  • 262

    "820 cases and 1,637 controls were recruited in 2015, and 1,433 cases and 2,867 controls in 2016. SMC was associated with an 89% reduction in malaria incidence for 4 weeks after treatment, and 62% from 5 to 6 weeks after treatment, compared with children who had not received SMC or whose last dose was more than 6 weeks before", ACCESS-SMC, Annex III - ACCESS-SMC Evaluation (preliminary reports), pg. 2.

  • 263

    “Methods
    Quasi-experimental design comparing changes in outcomes during the high transmission period (August–November) between SMC and non-SMC health districts before (2013–2014) and after intervention (two rounds in 2015 and 2016). Health indicators (number of uncomplicated malaria cases (UM) and severe malaria cases (SM)) from 19 health districts (8 in intervention and 11 in comparison group) were extracted from the District Health Information System (DHIS2)-based platform including health facilities data. Effect on incidence was assessed by fitting difference-in difference mixed-effects negative binomial regression model at a log scale.
    Results
    The two rounds of SMC were associated with a reduction of UM incidence (ratio of incidence rate ratio (IRR) 69% (95% CI 55–86%); p = 0.001) and SM incidence (ratio of IRR = 73% (55–95%), p = 0.018) among under five children.” Kirakoya-Samadoulougou et al. 2022, abstract.

  • 264

    “All causes mortality rate per 1000 person-years was 8.29 in the control areas compared to 3.63 in the intervention areas; age and gender adjusted mortality rate ratio 0.44 (95% CI 0.22–0.91), p = 0.027. The incidence rate of all causes hospital admissions was 19.60 per 1000 person-years in the intervention group compared to 33.45 per 1000 person-years in the control group, giving an incidence rate ratio (IRR) adjusted for age and gender of 0.61 (95% CI 0.44–0.84), p = 0.003”. Issiaka et al. 2020, abstract.

  • 265

    "A review of the quality of HMIS data for each country was conducted, which showed varied results between and within countries...Checks for data completeness were conducted for each country, as defined by the proportion of all monthly reports for each indicator available. Briefly, Burkina Faso showed high completeness of data with less than 1% of missing data points for the relevant indicators. Data coming from Chad and Nigeria showed low completeness ranging from 36- 42%. To mitigate potential bias from low completeness, any health facility or district with more than 1 month missing data during the SMC distribution period was excluded...Further checks for data accuracy revealed data entry errors as well as major health data reporting issues. First, common in all countries were obvious data entry errors most likely due to typing errors, entering data into the incorrect field, recording data in the wrong month, etc. The level of data accuracy varied across and within countries. Additionally, the analysis also revealed major data reporting issues that bring into question the overall accuracy of data that is captured in the national HMIS. For example, in Burkina Faso, approximately 30% of the data points reported more malaria cases in children under 5 than there were children seen that month." Malaria Consortium, 2019 impact report, pg. 7.
    "However, as previously discussed, the quality of data, the limitations of HMIS data reporting, and the inability to adjust for major factors affecting malaria transmission contribute to noise in the analysis and result in inaccurate estimates of effect...With regards to quality of impact data, in this analysis, we sought to analyse impact indicators available through the national HMIS from Burkina Faso, Chad, and Nigeria. Overall, the analysis indicated that accurate health data reporting is still a point of improvement for health systems and limitations to the data must be considered when interpreting results from national HMIS. Inaccurate data with no means of verification contribute to noise in the analysis and result in either dampening or amplification of estimated effects." Malaria Consortium, 2019 impact report, pg. 9.

  • 266

    Sangaré et al. 2022 (figure 3) and Bationo et al. 2023 (figure 8).

  • 267

    This understanding is based on an unpublished conversation with Professor Thomas Churcher, Professor of Infectious Disease Dynamics at Imperial College London, September 1, 2023.

  • 268

    “The total number of malaria cases reported by health facilities in Burkina Faso between 2013 and 2020 was 75,559,397, with 7,146,026 in 2013; 8,278,408 in 2014; 8,286,453 in 2015; 9,785,822 in 2016; 11,915,816 in 2017; 11,970,321 in 2018; 687,2720 in 2019 and 11,303,831 in 2020. From 2013 to 2020, the Sahelian, Sahelo-Sudanese and Sudanese zones accounted for 22%, 59% and 19% of malaria cases, respectively.
    The average annual incidence was 480 cases of malaria per 1000 inhabitants. It increased from 407 cases per 1000 inhabitants in 2013 to 518 per 1000 inhabitants in 2020 with a peak of 598 cases per 1000 inhabitants in 2017 (Table 1).” Sangaré et. al. 2022.
    “The national malaria control program recorded 55,417,532 malaria cases from January 2013 to December 2018 for a population that increased from 17,322,796 in 2013 to 20,244,079 in 2018 [53]. The median malaria incidence was 739.06 cases per 100,000 population/week over the entire period (range 294.46;2427.85). The highest incidences for the years 2013 to 2015 were observed between late July and early November. During the years 2016 to 2018, high incidences were observed between mid-June and early November. In addition to that from epidemic year 2013–2014 to 2016–2017, we observed two peaks in the incidence rate distribution: the first and high one around 500 cases per 100,000 inhabitants/week and the second around 1500 cases per 100,000 inhabitants/week (Fig 1).” Bationo et. al. 2023.

  • 269

    “In Cameroon in early 2022, ahead of the year’s SMC campaign, a decision was taken to pilot a new approach to distribution known as routine distribution. Social mobilisation, household enumeration and delivery of SMC was performed by the existing cadre of community health workers (CHWs) in 11 health districts out of the 47 eligible for SMC. These CHW were responsible for covering all of the children in their catchment areas in a period of less than 10 days. The expectation is that CHW familiarity with the families in their communities should facilitate these required steps. In contrast, in the other 36 SMC districts, SMC was conducted as a mass campaign, requiring recruitment, training and payment of a large number of distributors. Based on the reported results of the 2022 campaign, more than 98% of enumerated children were reached with SMC. No difference was noted in terms of performance between campaign and routine health districts; however, the cost of routine distribution was much lower ($1.24/child treated in mass campaign mode vs. $1.02 in routine mode). A total of 2 147 CHWs were involved in mobilization and distribution in the routine mode. This number would have been 4 102 in mass campaign mode. Similarly, Togo is currently conducting research on how SMC could be integrated into the routine delivery of community health services.” SMC Alliance, Celebrating 10 years of SMC, 2023, pg. 14.
    Note that in the Cameroon example described above, our understanding is that delivery of SMC would still take place door-to-door in a time-limited window. This resembles a campaign in some respects. We’re unsure what delivery model the Togo pilot involves.

  • 270

    See WHO, Policy Recommendation: Seasonal Malaria Chemoprevention, 2012.

  • 271

    These changes included delivering five cycles rather than four in some locations in Nigeria and Burkina Faso beginning in 2021, and pilots of SMC in Mozambique and Uganda in 2020–2021. See this section for more.

  • 272

    See World Health Organization, World malaria report 2023, p. 64, table 7.1.

  • 273

    Dr. André Tchouatieu, “GiveWell SMC report: Feedback from an independent review”, October 2023 (unpublished).

  • 274

    “Overall SMC is and will remain for many years one of the most cost-effective interventions to prevent malaria. Investment towards such intervention is valuable in all senses, despite some of the criteria used to assess the worth of the investment remaining unclear. However, cost-effectiveness could be improved with the routinisation of the intervention in the health system services delivery.
    The clear success of SMC is however now posing issues of equity. In SMC areas, children up to 5 years of age are highly protected against the disease, anticipating a shift of the burden of the disease to older age groups and this needs to be addressed. In non-SMC areas, children in the same age group as those living in the SMC areas are not benefiting from any sort of protection apart from the LLINs which are also distributed in SMC areas and this can be considered inequitable. And worst is that this situation can happen within the same country, Nigeria being an example.” Dr. André Tchouatieu, “GiveWell SMC report: Feedback from an independent review”, October 2023 (unpublished).

  • 275

    Dr André Tchouatieu, email to GiveWell, December 6th, 2023.

  • 276

    “Methods
    Cost data were collected from pilot implementation of SMC in Kita district, which targeted 77,497 children aged 3–59 months. Starting in August 2014, SMC was delivered by fixed point distribution in villages with the first dose observed each month. Treatment consisted of sulfadoxine-pyrimethamine and amodiaquine once a month for four consecutive months, or rounds. Economic and financial costs were collected from the provider perspective using an ingredients approach. Effectiveness estimates were based upon a published mathematical transmission model calibrated to local epidemiology, rainfall patterns and scale-up of interventions. Incremental cost effectiveness ratios were calculated for the cost per malaria episode averted, cost per disability adjusted life years (DALYs) averted, and cost per death averted.” Diawara et al. 2021, abstract.

  • 277

    “Ghana adopted the WHO-recommended Seasonal Malaria Chemoprevention (SMC) strategy with a trial in the Upper West Region in 2015. The objective of this study was to estimate the cost-effectiveness of seasonal malaria chemoprevention.
    Methods
    Costs were analysed from provider and societal perspectives and are reported in 2015 US$. Data on resource use (direct and indirect costs) of the SMC intervention were collected from intervention records and a survey in all districts and at regional level. Additional numbers of malaria cases and deaths averted by the intervention were estimated based on prevalence data obtained from an SMC effectiveness study in the region. Incremental cost-effectiveness ratios (ICERs) were estimated for the districts and region. Sensitivity analyses were conducted to test the robustness of the ICERs.” Nonvignon et al. 2016, abstract.

  • 278

    “When costs were combined with modelled effects, the economic cost per malaria episode averted in children was US $4.26 (uncertainty bound 2.83–7.17), US $144 (135–153) per DALY averted and US $ 14,503 (13,604–15,402) per death averted.” Diawara et al. 2021, abstract.

  • 279

    “The economic cost per additional child death averted by the intervention was US$3298.36 from the provider perspective and US$9858.02 from the societal perspective. The financial cost per the SMC intervention delivered to a child under-five was US$9.66. The ICERs were sensitive to mortality rate used.” Nonvignon et al. 2016, abstract.
    “Costing was undertaken from both provider perspective (which included provider-related costs incurred only
    on delivery of the intervention) and societal perspective (which included cost incurred on delivery of the intervention, donations and the time and other expenses of caregivers). The analytic horizon was 4 months, which constituted the duration of the SMC implementation.” Nonvignon et al. 2016, pg. 5.

  • 280

    The main example of this analysis is an analysis of data on SMC coverage and costs incurred in previous Malaria Consortium-supported campaigns. We use this as part of our cost per SMC cycle analysis (more).

  • 281

    Intuitively, if other funders filled a high proportion or all of the resulting funding gap in a short space of time, it would imply that funding gaps are more likely to be filled in the absence of GiveWell funding than we’re currently assuming.

  • 282

    “The World Health Organization (WHO) has recommended a new vaccine, R21/Matrix-M, for the prevention of malaria in children. The recommendation follows advice from the WHO: Strategic Advisory Group of Experts on Immunization (SAGE) and the Malaria Policy Advisory Group (MPAG) and was endorsed by the WHO Director-General following its regular biannual meeting held on 25-29 September.
    WHO also issued recommendations on the advice of SAGE for new vaccines for dengue and meningitis, along with immunization schedule and product recommendations for COVID-19. WHO also issued key immunization programmatic recommendations on polio, IA2030 and recovering the immunization programme.
    The R21 vaccine is the second malaria vaccine recommended by WHO, following the RTS,S/AS01 vaccine, which received a WHO recommendation in 2021. Both vaccines are shown to be safe and effective in preventing malaria in children and, when implemented broadly, are expected to have high public health impact. Malaria, a mosquito-borne disease, places a particularly high burden on children in the African Region, where nearly half a million children die from the disease each year.” WHO, "WHO recommends R21/Matrix-M vaccine for malaria prevention," 2023.

  • 283

    “The combination of RTS,S/AS01E and SMC was superior to SMC (protective efficacy 57.7%, 95% CI 53.3 to 61.7) and to RTS,S/AS01E (protective efficacy 59.0%, 54.7 to 62.8) in preventing clinical malaria. RTS,S/AS01E was non-inferior to SMC (hazard ratio 1.03 [95% CI 0.95 to 1.12]). The protective efficacy of the combination versus SMC over the 5-year period of the study was very similar to that seen in the first 3 years with the protective efficacy of the combination versus SMC being 57.7% (53.3 to 61.7) and versus RTS/AS01E-alone being 59.0% (54.7 to 62.8).” Dicko et al. 2023, abstract.

Based on our level of uncertainty about the best guesses calculated in our cost-effectiveness analysis, GiveWell staff gave their subjective 25th - 75th percentile confidence interval for each parameter. This column is an aggregation of these intervals. The implied cost-effectiveness column shows, for each parameter, what the program's overall cost-effectiveness would be at the 25th and 75th percent level of confidence, holding all other parameters constant.
We use multiples of direct cash transfers as a benchmark for comparing the cost-effectiveness of different programs.
($1m / 5.69)
~($1,000,000 / (176,000 x (100% - 0%) x 0.67% x 70% x 79%))
(Multiples of the value of direct cash transfers)
(116/$1,500/0.00335)
(23x / 74% x (100% + 19%) x (100% - 8%) x (100% - (0.3% + 43%)))
(($27.5m + $0.25m) / 21m)
($1.32 x 4.3)
(1,000,000 / 5.69)
(75% x (100% - 5%) / 90% x 100%)
(0.38% x (1 + 0.75))
(0.67% x 70%)
(26% / 60%)
(79% x 43% x (100% - 69%))
(176,000 x 2)
(176,000 x 60% x 79% x 70%)
(353,000 x 25% x 11% x 70%)
(59,000 + 6,500)
(~65,000 x 0.23)
(~15,000 / (~81,000 + ~15,000))
(units of value per $)
($1,000,000 x 0.105)
(units of value per $)
(units of value per $)
-($118,000 x 0.005) x 50%
~$1,000,000 x (0.105 - 0.015) x 50%
(-300 / 105,000)
(-45,000 / 105,000)
(-0.3% + -43%)