IRD Global — Mobile Conditional Cash Transfers for Immunizations (February 2023)

Note: This page summarizes the rationale behind a GiveWell grant to IRD Global. It reflects our understanding at the time we recommended the grant in February 2023. IRD Global staff reviewed this page prior to publication.

Published: March 2024

Summary

What we think this grant will do

This grant is a one-year extension (covering October 2024 to September 2025) of a three-year grant of up to $25 million that GiveWell recommended in October 2021.

This grant will fund IRD's continued technical support to the government to run the Zindagi Mehfooz electronic immunization registry in Sindh. It will also fund IRD staff to send SMS reminders for vaccination across Sindh province and will pay for the costs of sending those messages. Finally, this grant will fund IRD staff to operate the mCCT program to incentivize immunizations in seven districts in Sindh. The costs of the incentives themselves are covered by our previous grant. (more)

Why we recommended this grant

We estimate that this grant is 9.6 times as cost-effective as unconditional cash transfers, the benchmark to which we compare funding opportunities, (“9.6x”) when accounting only for direct benefits. We expect the program’s cost-effectiveness to increase from 9.6x to 13x in the next five years as IRD leverages the ZM platform to deliver additional programs. We expect the cost-effectiveness of this grant opportunity to be 26.4x when including opportunities for future programming that this grant creates.

In simple terms, we think this grant will be cost-effective because we think:

• This grant will increase child vaccination rates, which will lead to a reduction in child mortality from vaccine-preventable diseases. We expect vaccinations to be effective at preventing child deaths in Pakistan, and we estimate the incentives provided by IRD's program will increase coverage by roughly 10 percentage points in districts of Sindh province with low baseline vaccination rates. This is based on our review of available evidence, including preliminary results from the impact evaluation that we funded. We also expect SMS reminders for immunization to increase vaccination rates by roughly 3 to 4 percentage points, based on results from the Chandir et al. 2022 trial. (More)
• IRD can increase the cost-effectiveness of its program in the next five years by layering additional programs to the ZM platform. Since the ZM platform includes a registry of births and pregnancies in some health facilities, IRD plans to pilot using it to send SMS reminders for antenatal care (ANC), postnatal care (PNC), and breastfeeding. We incorporate a rough estimate of the benefits of ANC, PNC, and breastfeeding into our model. IRD is also considering using the platform for other programs. Since we don't have precise estimates of how each individually layered program may affect the overall program's effectiveness, we account for this through a rough 20% upward adjustment in benefits. (More)
• IRD will be able to increase the scale of its program in the next five years, to reach a size of $10 million per year. This includes expansion within and beyond Pakistan. This is based on a rough subjective guess. (More)

Recommending this grant also preserves optionality for our future grantmaking. We expect we will be able to improve our estimate of the program’s cost-effectiveness in the next year through the impact evaluation we funded to evaluate the effect on vaccination coverage due to IRD's program. At the time of making this grant, we expect final results by October 2024. This grant preserves the option to continue to fund this program in the future, when we may have more evidence of its cost-effectiveness. (More)

Main reservations

Our main reservations about the grant's cost-effectiveness are:

• What is the effect of mCCTs on childhood vaccination? Our estimate puts significant weight on the preliminary results from the impact evaluation we funded. However, those results are not statistically significant at conventional levels for four of the six vaccines; for a fifth vaccine (BCG, which makes up 30% of total benefits), the results are considerably higher than what was observed in a previous trial, and what researchers expected (and we don’t fully understand the reason for the discrepancy). The weight we put on these results increased our cost-effectiveness estimate by 25%, compared to our 2021 estimate. (More)
• How much of an increase in cost-effectiveness could be achieved by adding new programs to the ZM platform? Our calculation is based on a preliminary estimate of the cost-effectiveness of SMS reminders for antenatal, postnatal care and breastfeeding (which IRD is interested in adding to the platform in the nearer term), and a rough subjective guess of the cost-effectiveness of additional programs that could be added beyond that (which we express in a 20% upward adjustment). Both estimates are uncertain. (More)
• What are the baseline rates of vaccination coverage? Our cost-effectiveness estimate is highly sensitive to this input, as lower baseline coverage means the program could reach a higher number of otherwise unvaccinated infants. We estimate baseline coverage on the basis of the latest coverage estimates before program rollout. We add a subjective 25% downward adjustment for reduced cost-effectiveness due to increases in vaccination coverage and reduction in vaccine-preventable disease over time, but we are uncertain about the size of that adjustment. It is possible we are still underestimating baseline coverage and therefore overestimating the cost-effectiveness of the program. (More)
• How much room for more funding will IRD have at scale? We estimate that IRD will be able to use an additional $10 million per year by 2028, based on a rough subjective guess, informed by our rough understanding of the size of the locations that IRD told us it would be interested in expanding to. We're unsure whether IRD will reach that scale and, if it does, how long it will take to get there. (More)
• It’s possible we won’t learn much about the effectiveness of the program from the impact evaluation. At the time of recommending the grant to support the study, we had concerns about low statistical power to detect effects of the program. Based on preliminary results, we’re still unsure about how much the final results will update our understanding of the program’s effectiveness, due to low statistical power and disruptions to the program caused by floods. If the final results are not as informative as we expect, the optionality gained from this grant would be less valuable than we are estimating. (More)

The organization

IRD is a global health-focused organization with offices and programs in several countries.1 GiveWell has recommended funding to IRD programs previously, including a three-year grant of up to $25 million in October 2021 for its mobile phone-based conditional cash transfer (mCCT) program and support for the Sindh Electronic Immunization Registry (Zindagi Mehfooz) in Pakistan, and a planning grant for a randomized controlled trial of its Zero TB program.

The intervention

We model the core benefit of IRD's ZM platform and mCCT program as increasing child vaccination rates, leading to a reduction in child mortality from vaccine-preventable diseases. See our intervention report for more details on the evidence for effectiveness and how we model the cost-effectiveness of this program. We have also made updates to our cost-effectiveness analysis for this program, described in more detail below.

The grant

This grant covers a fourth year of funding for three activities funded through our previous grant: (1) technical assistance and implementation of a mobile-based electronic immunization registry called Zindagi Mehfooz (ZM) throughout Sindh Province, Pakistan, (2) sending SMS reminders to caregivers to bring children in for routine childhood vaccinations, using information from the ZM platform throughout Sindh, and (3) support for an mCCT program for immunizations that leverages the ZM platform in seven high-risk, low immunization coverage districts in Sindh.

ZM captures data on demographics and child vaccination status, and helps vaccinators determine the appropriate vaccines to offer at child visits. In some health facilities, it includes a pregnancy and birth registry to capture child births. IRD also employs the platform to deliver Short Message Service (SMS) reminders for immunizations as well as its mCCT program. The mCCT program provides cash incentives for caregivers who bring their children in for routine vaccines, which are provided by the Sindh government free of charge. Full details on the program and intervention are discussed in our 2021 grant page.2

Budget for grant activities

The budget for the fourth year of this program is $3,440,403.3 IRD plans to rollover $728,925 from estimated savings in implementation costs from the first three years of the program. This grant will cover the remaining costs.4

This grant does not cover the costs of the incentives provided in the mCCT program, since IRD expects to be able to rollover incentive funds received in 2021.5

The budget breaks down as follows:6

• Personnel (technical support, data analytics, partnership engagement, field implementation staff): $1,740,543
• Procurement (including program SMS costs and program technology costs): $462,216
• Operations: $380,525
• Indirect costs (12%): $368,615
• Local travel and fieldwork: $254,590
• Mobile van operations: $118,111
• Communications: $54,167
• International travel: $31,223
• Contingency (1%): $30,414

The case for the grant

Cost-effectiveness

We are recommending the grant because we believe it is a cost-effective use of funding, comparable to that of our top charities. We model the core benefit of IRD's technical support for the ZM platform, SMS reminders, and mCCT program as increasing child vaccination rates, leading to a reduction in child mortality from vaccine-preventable diseases. We estimate the grant’s cost-effectiveness to be 9.6 times as cost-effective as unconditional cash transfers (“9.6x”) when only accounting for direct benefits and 26.4x when including opportunities for future programming that this grant creates.7

The key drivers of this grant's cost-effectiveness are:

• We think this grant will increase child vaccination rates, which will lead to a reduction in child mortality from vaccine-preventable diseases. We expect vaccinations will reduce child deaths in Sindh by roughly 1 percentage point. This is based primarily on meta-analyses of the effects of vaccines provided as part of routine immunization in Pakistan.8

  We think small cash incentives can effectively incentivize caregivers to immunize their children. In Sindh, we estimate incentives of $1.26 per visit increase overall vaccine coverage by roughly 10 percentage points (15 percentage points for the BCG vaccine, and 6 to 12 for the PENTA sequence).9 This is based on the effect size observed in the Chandir et al. 2022 trial with some adjustments for scale, the effect size we would expect based on our review of New Incentives (a conditional cash transfer program operating in Nigeria that is one of our top charities), and preliminary results from the impact evaluation that we funded.

  We also expect SMS reminders for immunization to somewhat increase vaccination rates (by about 4 percentage points). This is based on the effect size observed in the Chandir et al. 2022 trial.

• We expect IRD to be able to increase the cost-effectiveness of its program in the next five years, by layering additional programs onto the ZM platform. By 2028, we expect the program’s cost-effectiveness to increase to 13x, as IRD leverages the ZM platform to deliver additional programs (in addition to mCCTs and SMS reminders for immunization).

  IRD is planning to pilot the addition of SMS reminders for antenatal care (ANC), postnatal care (PNC), and breastfeeding in the near term. Additional programs may be added in the future, such as sending SMS reminders to increase the likelihood that pregnant people deliver in health facilities and using the ZM platform to manage the stock of vaccines.10 We use a cost-effectiveness model we had developed to estimate the cost-effectiveness of ANC, PNC, and breastfeeding as a case study to roughly estimate IRD’s cost-effectiveness in the long run with those programs added.11 We make a rough upward adjustment of 20% to account for the benefits of programs beyond ANC, PNC, and breastfeeding in our model.

  After 2028, we estimate that:

  • 73% of the benefits of IRD’s work will come from mCCTs and immunization SMS reminders,
  • 10% will come from SMS reminders for antenatal care, postnatal care, and breastfeeding, and
  • 17% will come from other components that could be layered onto the platform.
• We expect IRD to be able to increase the scale of its program in the next five years, to reach a size of $10 million per year. This is a subjective guess based on our understanding of where IRD is interested in expanding within and beyond Pakistan.12

Optionality

• We expect we will be able to improve our estimate of the program’s cost-effectiveness in the next year due to the impact evaluation we funded. By funding the program to continue operating until we receive the impact evaluation's updated results, we are preserving the option to fund a highly cost-effective program in future years. In particular, we expect to receive updated results which will be more likely to be statistically significant at conventional levels. The preliminary results we used in our cost-effectiveness analysis are not statistically significant at conventional levels. We expect that future results will have a higher chance of being statistically significant because they will use individual level data rather than aggregate data.

Risks and reservations

Our estimate of the grant's cost-effectiveness is sensitive to some highly uncertain assumptions. The most significant of these uncertain assumptions are:

• How much weight should we put on the preliminary results from the UChicago evaluation of IRD's program? In our estimate of the effect of mCCTs on childhood vaccination rates, we place 40% weight on the preliminary results of an impact evaluation we funded.13 The results indicate the program is about 25% more effective at increasing vaccination coverage than we would have estimated using our 2021 methodology.14

  However, the preliminary results are not statistically significant at conventional levels for four of the six vaccines (Penta 2, Penta 3, Measles 1, and Measles 2).15

  We're particularly unsure about the size of the program’s effect on BCG, the first vaccine in the sequence, which is responsible for ~30% of program benefits. In particular, the result for this vaccine is considerably higher than it was in a three-year trial of a similar mCCT program in Korangi town in Sindh Province. (More on this trial’s results in our intervention report on IRD's program). It's also higher than what we would expect if, as the evaluation researchers hypothesize, most caregivers learn about the program at the first immunization visit. We're unsure what explains this result. It's possible that caregivers are learning about the program at birth, or through communications campaigns. However, it's also possible that the real size of the effect is smaller than we're currently estimating. We expect to get more data on this question in later evaluation data analyses.

• How much will cost-effectiveness increase when new programs are added to the ZM platform? We estimate that IRD’s program will reach 13x in five years by leveraging the ZM platform to deliver additional programs (aside from mCCTs and SMS reminders for vaccinations).16 (More here.) However, our subjective guess about the effect of other components on cost-effectiveness is highly uncertain.
• What are baseline vaccination coverage rates in the areas where IRD operates?17 Our model is highly sensitive to this input.18 A 7% decrease in baseline coverage estimates since 2021 increased our cost-effectiveness estimate by roughly 15%.19 We use the latest coverage estimates before program rollout as a proxy for baseline coverage in 2025, based on data from IRD. We then include a 25% discount to account for a decline in cost-effectiveness due to increases in vaccination coverage and reduction in vaccine-preventable disease over time. However, we did not explicitly model this, and it's possible we're underestimating real coverage and thus overestimating cost-effectiveness.
• How much room for more funding will IRD have at scale? We estimate that IRD will be able to use $10 million per year by 2028, based on a rough subjective guess. However, we are unsure about IRD’s ability to reach that scale and how long it might take.
  
• It’s possible we won’t learn much about the effectiveness of the program from the impact evaluation. At the time of making the grant to support the study, we had concerns about low power. Based on preliminary results, we’re still unsure about how much the results will update our understanding of the program’s effectiveness, due to low statistical power and disruptions to the program caused by floods.20 If the final results are not as informative as we expect, the optionality gained from this grant would be less valuable than we are estimating.

Plans for follow up

We will consider investigating a renewal grant in 2024. Before then, we plan to:

• Review results from the program evaluation. We plan to continue regular check-ins with the evaluation research team and follow up on their results and progress.
• Update the cost-effectiveness model with recent program data, including:
  • Total program costs
  • Percentage of children eligible for mCCTs
  • Updates on anti-fraud checks
  • Updates on cold chain checks
• Investigate mechanisms through which caregivers learn about incentives. For example, do caregivers learn about the program through communications campaigns, at birth, or during the first vaccination visit? We will ask the evaluation team and IRD to share data that would help us answer these questions. This will help us sense-check results and understand whether the effect on BCG vaccination rates is plausible.
• Investigate the likelihood of an increase in cost-effectiveness from additional components, and potential timelines for adding them to the platform. We plan to work with IRD to identify the most promising additional programs, and focus on those.
• Investigate room for more funding for this program at a larger scale, and timeline for expansion. This could include expansion in both Pakistan and in other countries. This work includes estimating where the program could be run cost-effectively.

Internal forecasts

For this grant, we are recording the following forecasts:

Confidence	Prediction	By time
60%	The final results of the program evaluation indicate mCCTs are at least as effective as initially estimated in our 2021 cost-effectiveness analysis	February 2024
40%	With GiveWell funding, IRD has scaled up SMS reminders for ANC, PNC, and breastfeeding across Sindh province, Pakistan	December 2024
30%	The program evaluation identifies at least one program variation which is more cost-effective than IRD’s main program	December 2025
65%	We estimate the IRD program is above the 8x bar (using current methodology) and has at least $10 million in room for more funding	February 2028
55%	We estimate the IRD program is above the 10x bar (using current methodology) and has at least $10 million in room for more funding	February 2028
10%	We estimate the IRD program is above the 12x bar (using current methodology) and has at least $10 million in room for more funding	February 2028

Our process

• In addition to ongoing quarterly check-ins for the 2021 grant, we had two calls with IRD to discuss this one-year extension and our open questions, and to answer their questions about our investigation and process.
• We followed up with questions via email and asked for IRD’s answers on our investigation plan.
• We asked IRD for a detailed budget. We've received permission to publish a category-level summary of this budget here.
• We have made a number of adjustments to our 2021 cost-effectiveness analysis, including adding preliminary results from the program evaluation by a team at the University of Chicago21 and updating the program model using new data from IRD.22 We also accounted for the long-term (beyond five years) benefits of additional ZM components which we expect would be added by 2028.23

Sources

Document	Source
Chandir et al. 2022	Source
Gavi, "Pentavalent vaccine support"	Source (archive)
GiveWell, “IRD Global — Mobile Conditional Cash Transfers for Immunizations (October 2021)”	Source
GiveWell, "University of Chicago — Evaluation of Mobile Conditional Cash Transfers (November 2021)"	Source
GiveWell, 2023 Cost-effectiveness analysis of IRD's ZM and mCCT program in Sindh	Source
GiveWell, Cost-effectiveness analysis of IRD’s mCCT program in high-risk Sindh districts, 2021 (public)	Source
GiveWell, IRD's ZM and mCCTs program value of information BOTEC	Source
GiveWell's non-verbatim summary of a conversation with IRD Global, August 17, 2021	Source
GiveWell's non-verbatim summary of a conversation with IRD Global, June 7, 2022	Source
Glennerster, mCCT Estimates Preliminary Results, 2023	Source
IRD, "About Us"	Source (archive)
IRD, "Global Footprint"	Source (archive)
IRD, Budget for mCCT scale-up (Sindh low-coverage districts), 2021 (redacted)	Source
IRD, Year 4 Summary Budget: mCCTs scale up in low-coverage districts, 2023	Source
World Health Organization, "BCG vaccine"	Source (archive)

• 1
  • "IRD is an international not-for-profit organization based in Singapore, with IRD country affiliates registered in Bangladesh, Indonesia, Nigeria, Pakistan, Philippines, South Africa, Vietnam, and Zimbabwe. We leverage process and technology innovations to develop, implement, and scale cost-effective, evidence-based, and high-impact solutions to stubborn health challenges and health inequities." IRD, "About Us"
  • For a map of IRD's country offices and programs, see IRD, "Global Footprint"

• 2

  See "The program" section of GiveWell, “IRD Global — Mobile Conditional Cash Transfers for Immunizations (October 2021)”

• 3

  IRD’s annual budget period is October to September. See IRD's budget for the first three years of the program here, and a summary of the year four budget here.

• 4

  The full grant amount was $2,712,552. Note that this is slightly higher than the funding gap ($3,440,403 - $728,925 = $2,711,478) as one of the donors supporting this grant rounded up their donation.

• 5
  • In 2021, we recommended up to $15 million for incentives: “Because the total cost for incentives is unknown (it will depend on enrollments in the program), we recommended that Open Philanthropy approve up to $15,066,861 overall across the three years for incentives.” GiveWell, “IRD Global — Mobile Conditional Cash Transfers for Immunizations (October 2021)”
  • See the rollover in the incentives budget here.

• 6

  IRD, Year 4 Summary Budget: mCCTs scale up in low-coverage districts

• 7

  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. As of early 2023, our bar for directing funding is about 10 times as cost-effective as unconditional cash transfers. See GiveWell’s Cost-Effectiveness Analyses webpage for more information about how we use cost-effectiveness estimates in our grantmaking.

• 8

  See this section of our intervention report for IRD's mCCT program.

• 9

  The BCG (bacille Calmette-Guérin) vaccine protects against tuberculosis, and the pentavalent vaccine (PENTA) is a five-in-one combination vaccine that protects against diphtheria, tetanus, pertussis, hepatitis B, and haemophilius influenzae type-B (HiB). See World Health Organization, "BCG vaccine" and Gavi, "Pentavalent vaccine support" for details.

• 10
  • “These benefits would all be layered onto ZM - ANC and PNC is doable right away. For nutrition, [IRD would] have to work with the government; the government is willing to. [IRD is] already doing ANC and PNC in Sindh in selected areas.”

  GiveWell's non-verbatim summary of a conversation with IRD, February 15, 2022 (unpublished)

  • Aside from reminders for breastfeeding, ANC, and PNC, in the long run IRD is considering using SMS reminders for:
    • Maternal tetanus immunizations.
    • Awareness messages to increase institutional delivery rates.
    • Breastfeeding initiation and continuity.
    • Guidance on family planning and linkage with contraceptive services.
    • Nutritional advice for women, newborns and children
    • Awareness on antenatal and postnatal depression and linkage with mental health care.
    • Awareness messages to improve newborn healthcare.
    • Screening services including Tuberculosis, Hepatitis, and HIV, especially in high prevalence areas, to prevent mother-to-child transmission.

  IRD, Answers to GiveWell Questions, December 2022 (unpublished).

• 11

  This model is based on rough, internal calculations and is thus currently unpublished. See this row in our cost-effectiveness analysis.

• 12

  IRD told us in June 2022 that the electronic immunization registry was launched in a new province in Pakistan (Khyber Pakhtunkhwa) and that there was interest in launching the registry in additional provinces as well. In previous conversations, IRD also expressed interest in expanding the ZM platform to countries outside of Pakistan. GiveWell's non-verbatim summary of a conversation with IRD Global, June 7, 2022; GiveWell's non-verbatim summary of a conversation with IRD Global, August 17, 2021

• 13

  The UChicago evaluation team shared preliminary results on the effect of IRD's program on vaccination coverage in February 2023. We expect updated results at the end of 2023, which will be based on a larger quantity of data.

• 14
  • Note that our estimate of the program’s effect size using 2021 methodology is different from our actual 2021 estimate, since our best guess depends on baseline coverage estimates, which we have updated.
  • Percentage point effect from trial (after adjustments): 12
  • Percentage point effect if we place 0% weight on the preliminary results: 9.7. See this section of the "Effect size - mCCTs" sheet of our cost-effectiveness analysis for these weights.
  • (12 - 9.7) / 9.7 = 23.7%

• 15

  See the confidence intervals in the "Percentage Point Effect" column of Table 2 in this document.

• 16

  This is based on (i) our estimate of cost and benefits from SMS reminders for antenatal care, postnatal care, and breastfeeding, which IRD told us it is interested in implementing (+10% increase in cost-effectiveness), and (ii) a subjective guess of how other components might increase the program’s cost-effectiveness (+20% increase in cost-effectiveness). However, the SMS model is still under review, and relies on IRD inputs we have not investigated deeply.

• 17

  Our best guess of counterfactual coverage in the near future for BCG is 78% in high coverage districts, and 62% in low coverage districts, where IRD plans to operate the mCCT incentive program. See our estimates of counterfactual coverage for the full schedule of vaccinations here.

• 18

  Lower baseline coverage increases cost-effectiveness because: (a) lower baseline coverage means providing incentives to fewer "always-takers," leading to lower costs, and (b) a higher share of unvaccinated infants means more caregivers could be affected by incentives, leading to a higher effect on coverage. As a result, our estimate of the program’s cost-effectiveness is highly sensitive to this input (for example, a 7% decrease in baseline coverage would lead to a 15% increase in cost-effectiveness).

• 19

  This 15% figure is based on a rough, internal analysis.

• 20

  Since making the grant, the evaluation has been substantially disrupted by floods in Pakistan in 2022. At the time of making this grant, the evaluation team is unsure how long it will take for flooding impacts to subside. It is possible those will affect the reliability of the study results.

• 21
  • In November 2021, we recommended a grant of $1.6 million to a team of University of Chicago researchers led by Dr. Rachel Glennerster to support an evaluation of IRD’s mCCT program in Sindh.
  • In October 2022, we received preliminary results from this evaluation. See the preliminary results here.

• 22

  We made these adjustments based on updated data from IRD:

  • Update in cost per child eligible for mCCTs from $9.50 to $10.76 (~15% decrease in cost-effectiveness based on a rough internal analysis)
  • Decrease in baseline coverage by an average of 7%. (~15% increase in cost-effectiveness)
  • Update in fraud adjustment from 10% to 5% (~2% increase in cost-effectiveness)
  • Update in vaccine efficacy adjustment from 0.83 to 0.8 (~8% decrease in cost-effectiveness). This change was a result of unpublished data shared by IRD on cold chain functionality, indicating that fridges were found not to be functional in roughly 27% of data collected. We believe the cold chain to be one of the main drivers of vaccine efficacy, and we judge this data to provide some evidence of cold chain malfunction in the intervention areas. We therefore applied an additional downward adjustment to vaccine efficacy.

• 23
  • In the near term, we believe IRD can pilot using the ZM platform to send SMS reminders for breastfeeding, antenatal care visits, and postnatal care visits. See the adjustment we make for these additional SMS reminders here in our cost-effectiveness analysis.
  • We expect that IRD may be able to leverage the ZM platform to deliver additional programs. We subjectively guess these might lead to a 20% increase in benefits (leading to a 17% increase in cost-effectiveness), compared to the version of the program including SMS reminders for immunization, breastfeeding, ANC, PNC, and mCCTs. However, this is a rough subjective guess about which we are uncertain. Potential programs to layer onto the ZM platform include:
    • Maternal tetanus immunizations.
    • Awareness messages to increase institutional delivery rates.
    • Breastfeeding initiation and continuity.
    • Guidance on family planning and linkage with contraceptive services.
    • Nutritional advice for women, newborns and children
    • Awareness on antenatal and postnatal depression and linkage with mental health care.
    • Awareness messages to improve newborn healthcare.
    • Screening services including Tuberculosis, Hepatitis, and HIV, especially in high prevalence areas, to prevent mother-to-child transmission.

  IRD, Answers to GiveWell Questions, December 2022 (unpublished).