Malaria Consortium — Renewal Grant for SMC in Nampula Province, Mozambique (January 2025)

Published: April 2025

Background

Malaria is a leading cause of under-5 mortality in Mozambique, and transmission is seasonal in the northern part of the country, with peak transmission between December and March.1 Seasonal malaria chemoprevention (SMC) involves giving children monthly courses of antimalarial medicines during peak malaria transmission periods. (more)

GiveWell previously co-funded Malaria Consortium to conduct SMC implementation and effectiveness studies in Nampula over the 2020-2021 and 2021-2022 seasons.2 We detailed the scope of these studies in Mozambique on a previous grant page. Prior to this, SMC had been studied and delivered primarily in the Sahel region of Africa.3 Following completion of these studies, GiveWell funded Malaria Consortium to deliver SMC in all 23 districts of Nampula over the 2022-2023 and 2023-2024 seasons, and again for the 2024-2025 season.4

Malaria Consortium’s SMC program is a GiveWell Top Charity, meaning we believe it is a highly cost-effective program and we have directed a significant amount of funding to it. See our intervention report on SMC for more on the intervention. See this page for all of our content on Malaria Consortium’s SMC program.

What we think this grant will do

We think this grant of up to $4 million will allow Malaria Consortium to deliver SMC at full scale across all 23 districts of Nampula province during the 2025-2026 and 2026-2027 seasons, by supplementing the roughly $11.5 million we estimate it has in rollover funds (more). The annual target population is around 1.6-1.7 million children under 5, or 3.3 million over the two-year span. After adjusting for coverage rates, self-report bias, and dose adherence, we estimate that the number of children receiving SMC will be approximately 2.3 million over the two years of the grant.5 We think this grant will cause children to receive SMC who would not otherwise have received it, which in turn will reduce under-5 malaria cases and deaths.

While we expect the two-year campaign to cost ~$16.7 million, we are recommending a much smaller grant of up to $4 million because, in early January 2025, the Mozambican government announced that the 2024-2025 SMC campaign in Nampula would be cancelled. At the time, there were concerns about post-election violence. There were also ongoing discussions between Malaria Consortium and the national malaria program about the resolution of payment issues relating to the 2023/24 round, which affected the decision to cancel the 2024/25 round.6 The cancelled campaign also resulted in costs which we have considered as part of the budget for this campaign, bringing the total cost of the campaigns to ~$17.5 million.7 After accounting for unspent campaign funds and additional leftover funding, Malaria Consortium will need an additional $3.5m-4m to carry out the 2025-2026 and 2026-2027 campaigns.

Malaria Consortium’s budget for the two years of implementation is $16,672,907,8 split roughly into:

• 36% intervention delivery costs
• 28% medicines, other commodities & freight
• 16% Malaria Consortium's program staff costs
• 11% management fee
• 6% digital, research, and external relations costs
• 2% Malaria Consortium operational costs
• 1% above-country costs (added by GiveWell)

Adding in $784,000 for staff costs incurred as a result of the cancelled 2024-2025 campaign results in an overall cost of $17,456,907.9

Why we made this grant

Cost effectiveness

We think this grant is likely to be around 9 times as cost-effective as unconditional cash transfers, our benchmark for comparing programs10 , which is just over our funding bar of 8x cash for Top Charities.11 The main reasons we expect this program to be cost-effective are:

• High malaria burden in Nampula – Our understanding is that malaria is a major cause of death among children under five years old in Nampula province:
  • Available data sources generally indicate that the annual under-5 malaria mortality rate in Mozambique is 0.2% to 0.3%.12 Sources of data on subnational malaria burden in Nampula show more disagreement (more), but based on the data we have seen, our best guess is that malaria burden in Nampula province is roughly 40% higher than the national average. We take this to imply an annual under-5 malaria mortality rate of ~0.35% in Nampula vs. our best guess of ~0.25% in Mozambique overall.
  • After accounting for a slightly lower malaria mortality rate in 3-59 month olds compared with all children under 5 (0.32% vs. 0.35%, due to high mortality among children aged 0-3 months ), and 0.75 indirect deaths per direct malaria death (see this section of our SMC intervention report for our reasoning behind this assumption), we estimate a ~0.56% malaria-attributable mortality rate for children 3-59 months in Nampula. This is a higher mortality rate than we estimate in most other areas where we fund SMC, surpassed only by Burkina Faso and a few states in northern Nigeria.13
  • However, our understanding is that Mozambique will be rolling out the R21 malaria vaccine over the coming months and years.14 While we don’t know on what timeline the vaccine will be rolled out and in which areas, we expect the vaccine rollout to reduce mortality from malaria in Mozambique. Our projected vaccine coverage rates (based on Mozambique’s other routine vaccination rates), model the malaria vaccine as reducing under-5 malaria mortality by 17% during this grant period.15 This leaves us with a final best guess mortality rate of 0.47% after accounting for the malaria vaccine rollout.
• Low cost of delivery – We think delivering SMC in Nampula is fairly inexpensive, though more expensive than in some countries where Malaria Consortium operates:
  • Our best guess is it costs roughly $2.11 to reach a child in Nampula with one cycle of SMC drugs. This figure is based on Malaria Consortium’s budget for the two years of implementation covered by this grant, with a series of adjustments to account for additional staff costs for the year with a skipped campaign,16 over-budgeting,17 non-MC costs,18 and imperfect coverage19 and adherence.20
  • While costs are higher than in the Sahelian programs where Malaria Consortium has a more established presence (we estimate that cost per cycle delivered was $1.64 in Burkina Faso and $1.09 in Nigeria last year), they’re lower than in any of the other non-Sahel programs. Costs in Nampula have been going down as the program has matured, and we think it’s plausible that costs will continue to fall a bit more, even after this grant period, before plateauing.21
• Effectiveness of SMC – We think SMC averts a substantial fraction of malaria mortality among children who receive it in Mozambique, though less so than in the Sahel. This is due to several factors:
  • Seasonality: While we think malaria transmission is less seasonally pronounced in Nampula than in the Sahel,22 our best guess is that just over half (52%) of the year’s malaria cases would otherwise happen in the four months during which Malaria Consortium will deliver SMC.
  • Effectiveness: Our best guess is that SMC averts about 60% of malaria mortality in children under 5 in non-Sahel settings like Nampula. That’s based on:
    • Effect sizes from a relatively strong body of experimental literature from the Sahel (where our estimate is that SMC averts of 79% malaria cases in treated children).23
    • A downward adjustment to account for higher levels of parasite resistance to SP24 in eastern and southern Africa compared with the Sahel, which is expected to result in lower effectiveness outside the Sahel.25 Our best guess is that SMC is around 25% less effective in places outside the Sahel, like Mozambique, where resistance to the drugs used in SMC is high.
    • An assumption that a reduction in malaria incidence translates 1:1 to a reduction in malaria mortality.26
  • Development of drug resistance: We think it’s possible that continued use of Sulfadoxine-Pyrimethamine with Amodiaquine (SPAQ) in high-resistance settings like Mozambique could contribute to further drug resistance in the future.27 We’ve attempted to account for this in the form of a 20% downward adjustment. While this represents our best guess at the moment, we’re particularly uncertain about this parameter and whether it adequately prices in potential downside from undermining future effectiveness (more).

See below a simplified version of our cost-effectiveness analysis for this grant.

What we are estimating	Best guess (rounded)	Confidence intervals (25th - 75th percentile)	Implied cost-effectiveness
Grant to Malaria Consortium	$17,456,907
Cost per child reached with SMC	$8	$6 - $9	7x - 11x
Number of children receiving SMC	2,310,104
Percent of children who would have received SMC without mass distribution	0%
Annual mortality rate from malaria and associated causes among children who do not receive SMC	0.47%	0.21% - 0.65%	4x - 12x
Proportion of malaria mortality occurring in SMC season	52%	41% - 64%	7x - 11x
Reduction in malaria mortality from receiving SMC	60%	42% - 71%	6x - 10x
Number of deaths averted among people under age five	3,310
Initial cost-effectiveness estimate
Cost per death averted (child mortality only)	~$5,273
Moral weight for each death averted	116
Value generated per dollar donated to GiveDirectly's unconditional cash transfer program28	0.0035
Initial cost-effectiveness estimate	~7x
Summary of primary benefits (% of modeled benefits)
Reduced child mortality	65%
Reduced mortality among older children and adults	10%
Income increases in later life	25%
Additional adjustments
Adjustment for additional program benefits and downsides	18%	4% - 27%	8x - 9x
Adjustment for grantee-level factors	-8%	-17% - -4%	7x - 9x
Adjustment for diverting other actors' spending into SMC ("leverage")	-1%
Adjustment for diverting other actors' spending away from SMC ("funging")	-20%	-47% - -8%	6x - 10x
Overall cost-effectiveness (multiples of cash transfers)	~9x

Other considerations:

Beyond cost-effectiveness, there are several additional factors that contributed to our decision to recommend this grant:

• Digitalization:
  • Starting in 2023, Malaria Consortium began implementing a digital tool called Salama for SMC delivery in Nampula.29 We think continued development and iteration of this system could improve program implementation through better real-time monitoring, more efficient supervision, and quicker payments to distributors. Successful, sustained implementation with Salama in Mozambique could also serve as a model for other countries. We think that learnings from Nampula could make for smoother digitalization of SMC campaigns elsewhere, though admittedly we don’t have a clear idea of how much that kind of knowledge dissemination happens in practice.
• Benefits to SMC elsewhere in Mozambique:
  • Our understanding is that Niassa province is beginning full-scale implementation of SMC as well, funded by the Global Fund and implemented through World Vision International’s sub-recipient, ADPP.30 While ADPP has experience in running door-to-door ITN campaigns, our understanding is that it doesn’t have SMC-specific experience. We think it’s plausible that having Malaria Consortium on the ground conducting SMC in a neighboring province could lead to smoother implementation for ADPP in Niassa, though we haven’t invested time in interrogating the mechanisms by which that would happen.
• Learning value:
  • We see modest opportunities for learning from this grant, in that it would give us a few more years of data to learn about how costs and coverage evolve over time. Nampula is a newer SMC program, and a large-scale one (10x the scale of the Karamoja, Uganda program, for example). To date, Nampula has displayed higher costs and lower coverage than the mature Sahelian programs. Over the course of this grant, we’d be able to see what happens to these two parameters over another two years of implementation. We note, however, that we’ve become more uncertain about our ability to learn given the cancellation of the ‘24-’25 round. We think it’s plausible that the missed campaign might interrupt whatever trends were in place and entail some backsliding in terms of program maturation.
• Malaria Consortium as a partner:
  • In general, our qualitative assessment of Malaria Consortium as an implementation partner is quite strong. We also judge their coverage monitoring to be of relatively high quality, though we think there is likely to be considerable room for improvement even from what we judge to be a high baseline.31 We expect to bring in a third party monitoring and evaluation (M&E) expert in 2025 to explore areas for improvement in more depth, and we expect that recommendations arising from that exercise can be implemented during this grant period.

Our main reservations

Our main reservations about this grant are:

• Drug resistance and its impact on effectiveness
  • Mozambique is a setting where malaria parasites show high levels of resistance to SP, and we remain uncertain about the impact of parasitic resistance on SMC’s effectiveness. We are also concerned that we could be under-appreciating downstream effects on future drug resistance. For now, we have attempted to account for these in our model through downward adjustments for reduced drug efficacy and the possibility of contributing to further resistance (more).
  • We expect to explore this question in more detail in the near future in order to inform our approach to SMC grantmaking outside the Sahel. We are currently evaluating several opportunities to support research on the relationship between SMC drug resistance, efficacy, and effectiveness, and our best guess is that we will make at least one grant along these lines in 2025.
• Risk of crowding out the Global Fund
  • We think there is a moderately high risk that we are crowding out funding for SMC in Nampula, in this case from the Global Fund; we incorporate that risk in our model in the form of a 20% funging penalty. The Global Fund currently supports SMC in Niassa province, but our understanding from conversations with stakeholders is that the government sees Nampula—as a high-burden province and Mozambique's largest—as the highest priority for SMC. We think that if we had decided not to fund Nampula, the Global Fund would likely have moved its Niassa funding to cover SMC in part of Nampula instead. We don’t think the Global Fund would have covered the full program in Nampula (which has around three times the target population of Niassa); our best guess is that they would have supported about 40% of it with the funding currently allocated to Niassa.32

Other reservations

• Low coverage rates
  • Coverage in Nampula has been relatively low during its two years of full-scale operation. While other Malaria Consortium programs typically achieve coverage rates above 90%, surveys in Nampula have found rates in the 70s.33 Malaria Consortium has told us that it considers anything under 80% "not acceptable."34 While Malaria Consortium attributes much of the low coverage to payment issues with distributors35 and expects these issues to improve, we're not yet sure how much improvement to expect. Lower coverage raises two concerns for us: those not reached may be among the most likely to benefit from the program (as we'd expect correlation between being hard to reach and having less access to other preventative measures), and it may be a negative update on overall implementation quality in Mozambique relative to other country programs.
• Implementation risks
  • Mozambique, like many of the places where Malaria Consortium implements SMC, has been experiencing recent security concerns.36 In other locations, the SMC campaigns have rarely been meaningfully delayed by insecurity, but in Mozambique, the government decided to cancel the ‘24-’25 campaign outright due to continuing post-election violence. There is a risk that future campaigns, such as those covered by this grant, could face similar challenges. Beyond this outbreak of violence, there’s been a longstanding insurgency in neighboring Cabo Delgado province that has periodically spilled over into Nampula, and it’s possible that election unrest could create fertile ground for more spillover.37
  • Even if future campaigns don’t get cancelled outright like this year’s, delays can mean one or more cycles being skipped and thus the loss of the opportunity for impact with some lost funding (e.g. for staff salaries), or all cycles going ahead but covering a relatively lower-transmission section of the year.
• Burden uncertainty
  • While there is broad agreement across data sources around malaria burden in Mozambique as a country, there are some discrepancies in subnational estimates across sources, and our final estimate for malaria burden in Nampula is sensitive to how much weight we put on different sources.
  • We estimate an annual malaria mortality rate for children under 5 of 0.351% in Nampula. This is based on four sources of data on national malaria burden (IGME,38 GBD,39 the 2022-23 DHS40 , and COMSA41 ) and three sources of data on subnational malaria burden (2022-23 DHS, COMSA and MAP42 ), combined via the following calculation:
    • National child malaria mortality rate (weighted average of four data sources): 0.258%
    • Nampula child malaria mortality rate (weighted average of three data sources): 0.38% or ~50% higher than our national estimate
    • Regression of subnational data towards the mean: 0.351%
      • This calculation means we're regressing subnational data 30% towards the national mean due to our sense that the difference is due to noisy sub-national data.
  • We place most weight on IGME and GBD (for national burden) and MAP (for subnational burden) due to our assessment of their modeling approach and the fact that they directly estimate malaria mortality. Less weight is provided to DHS because it doesn't directly measure malaria burden and COMSA because it's a newer source whose methodology and accuracy we haven't confirmed. Our final estimate is sensitive to these assumptions. For example, if we placed 100% weight on COMSA's subnational data, we would estimate an annual under-5 malaria mortality rate of 0.25% in Nampula; the most conservative approach would be taking GBD national data at face value and applying no subnational adjustment, which would imply a mortality rate of 0.2%.

Plans for follow up

• We will continue to have regular monthly calls with Malaria Consortium to discuss their work.
• Malaria Consortium will continue to submit spending reports and coverage survey reports for its Mozambique program, as it does for all campaigns, on the basis of which we will develop updated cost-effectiveness estimates.
  • We typically receive these in April each year, meaning that campaigns in Mozambique get split across reporting years due to the timing of the SMC season there. By Q3 of 2025, we will ask to see a separate spending report for Mozambique for August 2024–July 2025 (the year of the cancelled campaign). We will use this to determine how much to disburse in the second disbursement of this grant in January 2026 (that number might be $0 in the event Malaria Consortium has enough or more than enough to conduct the 2026-2027 campaign).
• In 2025, we plan to commission a third party review of Malaria Consortium’s routine coverage monitoring of its SMC programs, similar to what we’ve done recently for AMF (and are planning for other Top Charities). While we judge Malaria Consortium’s coverage monitoring to be of relatively high quality, this is based on a series of rather shallow reviews, and we expect a deeper dive by M&E experts to surface areas for further improvement. Our expectation is that any action items resulting from that review could be implemented in time for the ‘26-’27 SMC campaign.
• We are currently evaluating research opportunities to learn more about SMC drug resistance outside the Sahel and the relationship between resistance, efficacy, and effectiveness. Our best guess is that we will make at least one grant along these lines in 2025.

Internal forecasts

Confidence	Prediction	By time	Resolution
60%	We will renew support for SMC in Nampula beyond the '26-'27 season (not in the form of an exit grant).	Jan 2027	-
20%	We will make a grant for SMC in Mozambique in a province other than Nampula.	Jan 2027	-
30%	We will make an exit grant for SMC in Nampula.	Jan 2027	-
65%	The first cycle of the '25-'26 round will take place by Jan 2026.	Jan 2026	-
15%	The '25-'26 round will have more than zero but fewer than four cycles.	Aug 2026	-
10%	The '25-'26 round will be cancelled entirely.	Aug 2026	-
45%	Contingent on at least one cycle taking place, last-cycle coverage in the '25-'26 campaign will be >80%.	May 2027	-
45%	Contingent on at least one cycle taking place, last-cycle coverage in the '25-'26 campaign will be between ≥70% and ≤80%.	May 2027	-
15%	Contingent on at least one cycle taking place, last-cycle coverage in the '25-'26 campaign will be <70%.	May 2027	-
75%	We will make a grant to a third party to review Malaria Consortium's monitoring and evaluation.	Jan 2026	-
70%	We will make at least one VOI grant to explore the question of SMC drug resistance and effectiveness outside the Sahel.	Jan 2026	-
20%	We will make a grant for resistance or efficacy monitoring in Mozambique.	Jan 2026	-

Our process

• We had a series of calls with Malaria Consortium global and country team leadership and the Global Fund, and two members of GiveWell staff went to Nampula province in March 2024 to observe implementation of the 2023-2024 SMC campaign. They spoke with Malaria Consortium global and country team leadership, Mozambique central, provincial, and district health authorities, the Global Fund and PMI, and campaign staff.
• We received a funding proposal and budget for this grant.
• We updated our cost per SMC cycle analysis for ‘25-’26 and ‘26-’27 using cost projections and coverage data from the ‘23-’24 round.
• A Senior Researcher not otherwise involved in this grant spent several weeks conducting desk research and speaking with experts to explore the question of SMC drug resistance outside the Sahel.
• We reviewed Malaria Consortium’s End of Round Survey report from the ‘23-’24 campaign in Mozambique to update our shallow review of coverage monitoring methodology.
• We updated our burden estimates for Mozambique after a review of available data sources, and this was reviewed by two researchers not otherwise involved in this grant investigation.

Sources

Document	Source
Amnesty International, Mozambique: Authorities must investigate reports of more than 300 unlawful killings during post-election protest crackdown, 2025	Source (archive)
Demographic and Health Surveys Program, 2022-2023 DHS, Mozambique	Source (archive)
Demographic and Health Surveys Program, Homepage	Source (archive)
GiveWell, 2020 moral weights	Source
GiveWell, All Content on Malaria Consortium's Seasonal Malaria Chemoprevention Program	Source
GiveWell, Analysis of Malaria Consortium's cost per SMC cycle administered, 2024 (public)	Source
GiveWell, Counterfactual Value of Global Fund Spend	Source
GiveWell, GiveDirectly’s Cash for Poverty Relief Program	Source
GiveWell, GiveWell's Cost-Effectiveness Analyses	Source
GiveWell, GW's analysis of subnational mortality rates for SMC [2025]	Source
GiveWell, IDinsight — Review of AMF’s Monitoring (March 2024)	Source
GiveWell, Malaria Consortium – Seasonal Malaria Chemoprevention	Source
GiveWell, Malaria Consortium – Seasonal Malaria Chemoprevention, Nampula, Mozambique (February 2022)	Source
GiveWell, Malaria Consortium — SMC in Karamoja, Uganda in 2024 and Nampula, Mozambique in 2024-2025 (November 2023 - January 2024)	Source
GiveWell, Malaria mortality adjustments due to vaccine rollout (November 2024)	Source
GiveWell, Malaria Vaccines	Source
GiveWell, Our Top Charities	Source
GiveWell, Projection of Malaria Consortium's cost per SMC cycle administered in Mozambique in 2025/26-2026/27	Source
GiveWell, Re-evaluating the Impact of Unconditional Cash Transfers, 2024	Source
GiveWell, Seasonal Malaria Chemoprevention, 2024	Source
GiveWell, Summary of Malaria Consortium SMC Coverage Surveys [2025]	Source
GiveWell, The fungibility question: How does GiveWell’s funding affect other funders?, 2024	Source
GiveWell, GiveWell's CEA of seasonal malaria chemoprevention (SMC) — 2025 version 2	Source
Institute for Health Metrics and Evaluation - Global Burden of Disease	Source (archive)
Inter-agency Group for Child Mortality Estimation (UN IGME), Homepage	Source (archive)
International Crisis Group, Mozambique	Source (archive)
Macicame et al., 2023 (COMSA)	Source
Malaria Atlas Project, Homepage	Source (archive)
Malaria Consortium, 2022 SMC Philanthropy Report	Source
Malaria Consortium, 2023 SMC Coverage Report	Source
Malaria Consortium, 2023 SMC Philanthropy Report	Source
Malaria Consortium, End of Round Survey for 2024 SMC Campaign in Mozambique	Source
Malaria Consortium, Grant request: Supporting SMC in Mozambique, 2025/26-2026/27 November 2024	Source
Malaria Consortium, Home page	Source (archive)
Malaria Consortium, Lessons from using SALAMA health campaign management platform to implement and optimise SMC in Mozambique, 2024	Source (archive)
Malaria Consortium, Mozambique SMC Budget 2025-26 and 2026-27	Source
Nuwa et al. 2023	Source
UNICEF, Mozambique introduces the lifesaving R21 malaria vaccine, a critical step forward to revitalize malaria progress and improve child survival, 2024	Source (archive)
Wikipedia, Noisy Data	Source (archive)
World Health Organization, Country Disease Outlook, Mozambique, 2023	Source (archive)
World Health Organization, Updated WHO recommendations for malaria chemoprevention among children and pregnant women, 2022	Source (archive)

• 1
  • WHO Country Disease Outlook, Mozambique, 2023: “The entire population is at risk of malaria, which causes 22,000 deaths annually, with a burden of >10 million cases. This is high relative to GTS targets and WHO supports a high burden, high impact approach to malaria control.”
  • Malaria Consortium SMC Program Philanthropy Report 2023: “Malaria continues to be highly endemic across the entire country, with the highest prevalence in the north and along the coast. Unlike in all other SMC-implementing countries supported by Malaria Consortium, the high malaria transmission season in northern Mozambique cuts across calendar years, approximately between December and March.”

• 2

  From the Malaria Consortium, 2022 SMC Philanthropy Report:

  • “An insight brief summarising lessons learnt from the first phase of the project during the 2020/21 season, when SMC was implemented in two districts of Nampula, targeting 70,000 children, was published on Malaria Consortium’s website....The second phase of the project involved SMC delivery to 110,000 children during the 2021/22 season.” p. 42
  • “The SMC study in Mozambique was designed as a two‐phase hybrid effectiveness‐implementation study. The first phase focused on acceptability and feasibility, followed by more rigorous assessments of the effectiveness of the intervention and chemoprevention efficacy of [Sulfadoxine-Pyrimethamine with Amodiaquine (SPAQ)] in phase 2…The study was conducted in collaboration with the PNCM and the Centro de Investigação em Saúde de Manhiça. It was co‐funded by the Bill & Melinda Gates Foundation.” p. 73

• 3

  “In 2012, the World Health Organization (WHO) recommend the scale-up of SMC in areas where malaria transmission is highly seasonal and the therapeutic efficacy of SP and AQ is above 90%. Consequently, the Sahel region was prioritized for SMC, as prevalence of resistance markers for SP is high in much of east and southern Africa.” Nuwa et al. 2023, p. 2

• 4

  Note that the 2024-2025 campaign was cancelled by the Mozambican government in January 2025 due to concerns related to post-election violence – more.

• 5

  See our coverage projections for the 2025-26 and 2026-27 campaigns here.

• 6
  • In 2024. GiveWell made a $14,574,211 grant to support Malaria Constortium’s SMC campaign for 2024-25 in Nampula province, Mozambique. Email from Malaria Consortium, Jan 8, 2025 (unpublished).
  • In January 2025, Malaria Consortium alerted GiveWell that the Mozambican government had suspended the 2024-2025 campaign due to concerns about post-election violence. The campaign, which was scheduled to start in late 2024, had previously been delayed due to the security situation and administrative issues, specifically around payment issues for the previous 2023/24 round

• 7

  Malaria Consortium stated to us that the primary costs associated with the cancelled campaign were the procurement of drugs used for SMC campaigns, that will now be used in the 2025-2026 cycle, and therefore did not add to overall costs, and staff costs, which did result in ~800,000 more in overall costs.

• 8

  See our summary of the budget for Malaria Consortium’s 2025-26 and 2026-27 campaigns here.

• 9

  See our summary of these costs here.

• 10
  • To date, GiveWell has used GiveDirectly's unconditional cash transfers as a benchmark for comparing the cost-effectiveness of different funding opportunities, which we describe in multiples of "cash” (more). In 2024, we re-evaluated the cost effectiveness of direct cash transfers as implemented by GiveDirectly and we now estimate that their cash program is 3 to 4 times more cost-effective than we’d previously estimated. (more)
  • For the time being, we continue to use our estimate of the effectiveness of unconditional cash transfers prior to the update to preserve our ability to compare across programs, while we reevaluate the benchmark we want to use to measure and communicate cost-effectiveness.
  • Note that a) our cost-effectiveness analyses are simplified models that are highly uncertain, and b) our cost-effectiveness threshold for directing funding to particular programs changes periodically. See GiveWell’s Cost-Effectiveness Analyses webpage for more information about how we use cost-effectiveness estimates in our grantmaking.

• 11

  See more on how we use cost-effectiveness in our grantmaking, why we use unconditional cash transfers as our benchmark, and our funding bar for charities here.

• 12

  In the referenced spreadsheet, we calculate the implied annual under-5 malaria mortality rate in Mozambique for non-IHME data sources (DHS, COMSA, UN IGME) relative to IHME. We do this to standardize measures across data sources as each data source reports malaria-specific and all-cause mortality figures differently. Using the data sources in the spreadsheet, GBD 2021 estimates an annual under-5 malaria mortality rate in Mozambique of 209 per 100,000. We calculate that UN IGME 2021’s figures imply a rate 53% higher than IHME’s (320), DHS 2022-23 implies a rate 8% higher than IHME’s (226), and COMSA 2019-20 implies a rate 34% higher than IHME’s (280).

• 13

  See this row of our CEA for estimated malaria-attributable mortality rates across all locations that we’ve investigated.

• 14

  See this UNICEF release on the rollout. “Through Gavi, the Alliance for Vaccines, and co-financing from the Government of Mozambique, the country has around 800,000 doses of the malaria vaccine for the Expanded Programme of Immunization, with the aim of immunizing around 300,000 children in the first phase, which begins on 5 August 2024. The vaccine will be distributed in 22 districts in the province of Zambézia will be administered in a 4-dose schedule, the first dose will cover children aged 6 to 11 months.”

• 15

  In Mozambique, the 17% mortality reduction is based on an assumption that third-dose malaria vaccine coverage will reach 61% in 2026 and 65% in 2027. We are highly uncertain about what rollout will actually look like. For more on malaria vaccines, see our intervention report.

• 16

  In light of the cancellation of the ‘24-’25 campaign, Malaria Consortium told us it expects to need to spend roughly $800k in the first half of 2025 for staff costs for core staff, severance for non-core staff, etc. We're deciding to incorporate this expenditure into our forward-looking cost estimates given that it hasn't yet been incurred.

• 17

  We believe that Malaria Consortium's forward-looking budgets tend to be slightly conservative based on our cursory review of past data on actual expenditure compared to budget projections. We adjust their projections downwards by 10% to account for this.

• 18

  We assume that beyond Malaria Consortium's own costs, there is an additional ~12% of costs incurred by the government to support the program.

• 19

  Our coverage estimates include the fact that not all targeted children are reached, and not all those who are reached receive the full number of cycles. Malaria Consortium’s 2024 End of Round Survey found average coverage of 74.6% across the four SMC cycles in Nampula in the ‘23-’24 round; we assume the same level of coverage in the ‘25-’26 round. In the ‘26-’27 round, we assume a 5pp improvement in coverage. In both rounds, we then apply a 15% downward adjustment for self-report bias, as coverage figures reported in Malaria Consortium’s M&E are dependent on caregiver recall several months after the fact (more) and are therefore, we think, slightly overstated.

• 20

  Each cycle of SMC involves a three-day regimen, with the first day’s drugs being administered by a campaign distributor, and the second and third day’s drugs being administered by the child’s caretaker without direct observation. We apply a ~6% penalty as our best-guess adjustment to account for children who receive the first day’s drugs but not the following days’.

• 21

  Malaria Consortium noted to us that the Nampula program is reaching a phase of maturity where supervision costs are expected to fall substantially, certain functions are being transitioned to the government, and less money is spent on securing authorities’ and communities’ buy-in and providing health authorities' staff with knowledge of SMC. Through our discussions with Malaria Consortium as part of this investigation, we found the overarching narrative for the cost reductions intuitively plausible.

• 22

  In the Sahel, our estimate is that 70% of annual malaria cases happen during the season when SMC is delivered. See this section of our SMC intervention report for more.

• 23

  See this section of our SMC intervention report for more.

• 24

  SP (sulfadoxine pyrimethamine) is one of the two drugs used together for SMC. The other is AQ (amodiaquine).

• 25

  See this WHO recommendation which notes higher levels of resistance to SP and AQ outside of the Sahel. “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.”

• 26

  See here for our reasoning behind this assumption.

• 27

  Our understanding of the mechanism by which this would happen is roughly: SPAQ is less efficacious when parasites have high resistance to it → administering SPAQ in high-resistance places doesn’t clear all parasites from the bloodstream → those parasites left behind are the most resistant against SPAQ → each time SPAQ is used, a generation of slightly more resistant parasites is left to propagate, exerting selection pressure with each cycle. We discuss our understanding of the development of drug resistance further in this section of our report on SMC.

• 28
  This estimate of GiveDirectly's value per dollar donated is out of date as of 2024. We are continuing to use this outdated estimate for now to preserve our ability to compare across programs, while we reevaluate the benchmark we want to use to measure and communicate cost-effectiveness.
  
• 29

  See Malaria Consortium’s report on lessons learned from rolling our SALAMA in Mozambique here.

• 30

  This is our understanding from multiple conversations with stakeholders.

• 31

  See our shallow 2024 review of Malaria Consortium’s coverage survey methodology here.

• 32

  In effect, this means we think 40% of our funding in this grant is causing Niassa to be funded. We translate that into a 35% value on the funging risk parameter because, based on cursory calculations, we think SMC in Niassa is likely to be significantly less cost-effective than in Nampula, but slightly more cost-effective than our current estimate for the Global Fund’s marginal dollar:

  • Using under-5 malaria prevalence in the 2022-2023 DHS as a rough proxy, malaria burden in Niassa seems to be about 60% of that in Nampula;
  • With Nampula around 11x cash without any leverage and funging adjustment, applying that burden adjustment suggests (crudely) that Niassa would be around 6.6x;
  • We expect that cost per child will be higher in Niassa than in Nampula because of its smaller population. We roughly guess this would reduce cost-effectiveness by ~20%, bringing our estimate of cost-effectiveness of SMC in Niassa to 5.3x. Our model currently estimates the value of the Global Fund’s marginal dollar at 4.6x.
  • After applying a proportional adjustment to the funging penalty (which would be 40% if Niassa SMC were equally cost-effective as the GF's marginal dollar), our best guess for the funging parameter is 35%.

• 33
  • In the ‘22-’23 round, coverage in the final cycle in Mozambique was 77%, whereas for all other countries the corresponding figure was above 90% (save for Togo with 89%) – see Malaria Consortium's 2023 SMC Coverage Report, table 22.
  • In the ‘23-’24 round, coverage in the final cycle in Mozambique was even lower at 71% (see Mozambique 2024 End of Round Survey Report, table 4). We don’t have corresponding figures from other countries yet, but we don’t have reason to believe that they’ll be much different from previous years (i.e. coverage rates in the nineties).

• 34

  This is based on unpublished conversations with Malaria Consortium. Exceptions to this rule apply in the case of extenuating circumstances, none of which applied to Mozambique in the years in question.

• 35

  Malaria Consortium has told us that they’ve had issues reconciling records of work done by community distributors in Mozambique which has led to delayed payments and demotivated distributors. They’ve acknowledged this as a priority area for improvement and they plan to roll out new digital tools to manage payments in the coming years. They’ve also highlighted the need for robust planning that defines responsibilities and timelines between Malaria Consortium and local authorities, as well as within Malaria Consortium (e.g. programmes and finance).

• 36

  See, for example, this Amnesty International article on Mozambique’s political unrest.

• 37

  The International Crisis Group cover ongoing insurgency in Cabo Delgado here.

• 38

  The United Nations Inter-agency Group for Child Mortality Estimation.

• 39

  The Institute for Health Metrics and Evaluation’s Global Burden of Disease database.

• 40

  The Demographic and Health Surveys, last conducted in Mozambique in 2022-2023.

• 41

  COMSA is a government-led effort (supported by Johns Hopkins University) to a) create a unique sample vital registration system that samples a large number of households (~200,000 across the country) on which it continuously collects vital statistics, and b) model causes of death through verbal autopsy algorithms (calibrated at the national level using a small set of tissue samples).

• 42

  The Malaria Atlas Project.