Note: This page summarizes the rationale behind a GiveWell grant to Ansh as of December 2024, when we made the grant. Ansh staff reviewed this page prior to publication.
In a nutshell
In December 2024, we recommended a $1,976,578 grant to Ansh to support their Kangaroo care (KC), also known as Kangaroo mother care (KMC), program in Rajasthan, India across 2025 and 2026. Ansh’s KC program involves skin-to-skin contact between low birth weight newborns and caregivers, exclusive breastfeeding, and monitoring of danger signs for early detection and treatment. We believe this program reduces neonatal mortality. Ansh works with implementation partners to staff nurses in hospitals, who support new caregivers with providing KC to at-risk newborns.
We are recommending this grant because:
- We think the direct benefits of this program are likely to be highly cost-effective. We estimate that this program is around 13 times as cost-effective as GiveWell’s benchmark, primarily due to reductions in neonatal mortality.
- We believe we will learn valuable information from this grant that will inform our future funding decisions and could lead to further cost-effective grant opportunities to KC programs. Through this grant we expect to learn about the cost and number of children reached at scale and Ansh’s ability to maintain program quality as they expand. If we continue to believe the grant is cost-effective, we may fund Ansh’s program in the future, or consider whether similar program models may be promising in other locations.
- Ansh is a new organization, and by providing funding now, we think we can give them the opportunity to invest in their organizational development, rather than focusing their core capacity on fundraising.
Our main reservations about this grant are:
- We’re quite uncertain about our counterfactual mortality estimates, and this grant is unlikely to help us address this uncertainty. We have estimated the mortality rate of low birth weight and preterm newborns in Rajasthan, but it is unlikely that we will be able to collect primary data on this. We could be overestimating the cost-effectiveness of the program if our estimates of counterfactual mortality are too low.
- Because Ansh is a new organization, their program is relatively unproven. We therefore have some uncertainty about whether they will be able to scale at their planned rate while maintaining program quality. We also think if Ansh is unable to hire key staff or integrate their monitoring and evaluation (M&E) systems into decision-making, then there may be a risk that the program could be less scalable and sustainable than we have assumed.
- Because we’ve previously done work investigating KC programs and because our key questions about Ansh were qualitative, we spent a relatively limited amount of time on desk research for this grant compared to other similarly sized grants. This could mean we have missed key considerations that we otherwise would have noticed.
Published: March 2025
The organization
Ansh is an organization launched in April 2023, incubated by Ambitious Impact (AIM).1 Ansh operates in Rajasthan, India, to support kangaroo care (KC) in postnatal care wards and special newborn care units (SNCUs). Ansh has been implementing their program since January 2024.
The intervention
Kangaroo care, also known as "kangaroo mother care" (KMC), is an intervention targeted at low birth weight infants that primarily involves skin-to-skin contact between a caregiver and the child. Many KC programs, including Ansh’s, also promote exclusive breastfeeding and monitoring of danger signs for early detection and treatment.2 There is strong evidence that KC reduces neonatal mortality compared to conventional neonatal intensive care. Some hypothesized reasons for this include reductions in hypothermia, increased maternal milk production, and a reduction in infections.3 See our intervention report for more detail on KC.
While GiveWell believes KC to be a highly promising intervention, we have had difficulty finding programs that implement KC in a scalable way. We believe Ansh's implementation model has the potential to be quite scalable because it is focused on enabling KC through expanded staff capacity at health facilities and training with established materials and procedures, but does not include heavy investments in hospital infrastructure (such as outfitting specialist wards) or conducting additional home-based services. We think this approach might be promising in certain contexts.
The grant
This $1,976,578 grant will support Ansh's KC program in Rajasthan, India.4 As of December 2024, Ansh works across four hospitals in Rajasthan and plans to expand to an additional 11 during 2025. Ansh has raised funds to cover most, but not all, of its 2025 expenses. This grant would cover Ansh’s remaining funding gap for 2025, as well as cover their full program expenses in 2026.5
Ansh’s program specifically involves working with implementation partners to provide nurses at hospitals who support new parents with skin-to-skin contact, breastfeeding, danger sign and distress sign monitoring, and other counseling.6 The specific activities that Ansh supports include:7
- Recruiting and paying the health workers (nurses) that deliver the program8
- Providing chairs, privacy screens and backrests for KC
- Providing necessary supplies (thermometers, weighing scales, tablets, temperature monitoring watches, KC slings, etc.)
- Providing training to nurses and coordinators
- Providing training program and materials
- Providing program monitoring tools and a data collection platform
Ansh does not outfit a separate space to support KC in the facilities they support, but they do provide KC chairs in SNCUs, post-operative rooms, and postnatal care wards.9
Among other activities,10 Ansh conducts four or more follow-up calls to caregivers after they are discharged from health facilities.11 While they do not conduct home visits, they encourage caregivers to arrange home visits by Accredited Social Health Activists (ASHAs),12 and in some cases, call the local ASHAs themselves.13
Budget for grant activities
The $1,976,578 grant includes:14
- $256,475 to cover their remaining funding gap for 2025. This is the remaining portion of $1,164,475 for program running costs in 2025 (of which $908,000 has been fundraised for). Of these program running costs:
- $1,322,230 for program running costs in 2026
- $397,873 for contingency and inflation costs18
The case for the grant
Cost-effectiveness
Direct benefits
Our best guess is that the direct benefits of the grant are around 13 times as cost-effective as cash transfers.19 The main benefit we model in our analysis is a reduction in neonatal mortality, and our best guess is that the grant will avert a neonatal death for around $2,500. However, we’re quite uncertain about our estimate of baseline mortality and somewhat uncertain about whether Ansh’s program will be able to maintain the same impact at a larger scale (more). Accordingly, we think it is possible we have over- or under-estimated the cost-effectiveness of this program. Under different plausible assumptions about baseline mortality rate and program effect size, we think the cost-effectiveness of the program could range from 6x – 19x.20
We think this program is cost effective because:
- The program is estimated to cost around $85 per infant. This is based on Ansh’s budget and estimated number of births occurring at the facilities they are working in.21
- We think that neonatal mortality is high among low birth weight infants in Rajasthan, India. Our best guess is that the neonatal mortality rate in this group is around 12%, although this estimate is highly uncertain.22
- We believe that KC can lead to large reductions in neonatal mortality.
- There is strong evidence that KC reduces neonatal mortality compared to conventional neonatal intensive care.23
- Moreover, program data indicates that neonatal mortality for low birth weight infants who received Ansh’s program is much lower than our best guess of their counterfactual mortality rate (i.e., what the mortality rate would be for those same infants if they were not enrolled in the program).24
- We also apply a rough 25% upward adjustment to account for other benefits. This is based on our existing KC work.
A sketch of our cost-effectiveness analysis and level of uncertainty is in the table below. This analysis requires taking a stand on several uncertain parameters that could change cost-effectiveness substantially. These key uncertainties are highlighted in yellow.
| Best guess | Confidence intervals (25th - 75th percentile) | Implied cost-effectiveness | |
|---|---|---|---|
| Grant size | $1,976,578 | ||
| Population reached | |||
| Cost / patient | $85 | ||
| Number of infants reached | 23,181 | ||
| Child mortality benefits | |||
| Estimated neonatal mortality (in target population) | 12% | ||
| Estimated neonatal mortality rate for infants in program | 6% | ||
| Mortality reduction, without adjustments | 47% | ||
| Mortality reduction, with IV/EV adjustments | 28% | 9% / 43% | 4-19x |
| Number of deaths averted | 789 | ||
| $ / death averted | $2,505 | ||
| Moral weight - neonatal mortality | 84 | ||
| Initial cost-effectiveness estimate | 10 | ||
| Benefit streams (% of modeled total) | |||
| Neonatal mortality benefits | 80% | 70% / 100% | 10-14x |
| Development effects | 12% | ||
| Morbidity benefits | 8% | ||
| Additional benefits and adjustments | 1.3 | ||
| Final cost-effectiveness (after all adjustments) | 13 |
You can see our cost-effectiveness analysis for the program here.
Option value of potential room for more funding
While we think this grant is cost-effective based on direct benefits alone, it also provides additional benefit through the value of the information we expect to gain about Ansh's program and model. Learnings through this grant could enable us to make larger future investments in a cost-effective program area if this grant increases our confidence that Ansh's KC model is impactful and scalable.
Specific parameters we expect to learn about through the grant include quantitative inputs such as the number of infants reached, the costs per hospital, and the mortality rate within the program as it scales. We also expect to learn about key qualitative considerations, such as Ansh’s ability to develop their internal management team and whether they can scale at their planned speed. When we include the value of information from the learnings during this grant, we model the cost-effectiveness of this grant to be around 15x.
Qualitative factors
Ansh is a relatively young organization, incubated through AIM. Our understanding is that they have placed more focus on program development than organizational strengthening.25 It seems plausible that by providing some runway of funding, this grant will allow Ansh to focus on activities such as the hiring of core leadership roles and developing robust M&E systems.
Additionally, Ansh has indicated that they currently have buy-in for their program from state and district officials. Funding Ansh now, when they can leverage the momentum and support of this political buy-in and good will, may give them an increased likelihood of successful program implementation at scale. During a site visit, Laliteswar Kumar, acting as a consultant on behalf of GiveWell, also observed that there was enthusiasm from district hospital officials for the program.
Risks and reservations
Our main reservations about this grant include the following.
- We’re quite uncertain about our counterfactual mortality estimates, and this grant is unlikely to help us address this uncertainty. For this grant, baseline mortality among the target demographic (preterm and low birth weight infants in the hospitals where Ansh will work) is a key uncertainty. We would ideally want to collect good data on mortality rates for low birth weight infants in facilities before Ansh begins working in them, to establish a baseline mortality estimate, but we don’t expect to be able to collect that data as part of this grant. Before GiveWell would fund KC programs at a larger scale, we will likely want to consider conducting more rigorous data collection on counterfactual mortality.
- Ansh’s program is relatively unproven. Ansh is a relatively new organization, which means they don't have a long track record of program implementation. As a result, we have some uncertainty about their ability to scale at their planned rate and maintain program quality while doing so. This is something we hope to learn about during the grant period. Specific things we will be looking at include whether Ansh is able to identify and work with local partner organizations and understanding the reasons behind any potential delays in scaling.
- We have reservations related to Ansh's currently underdeveloped organizational structure. As noted above, Ansh is a relatively young organization, and our understanding is that they have placed more focus on program development than organizational strengthening. As a result, there may have been a lack of investment in the core leadership team beyond the CEO. Ansh told us that funding constraints made hiring difficult. It is unclear to us whether Ansh will be able to hire effectively and build a robust, stable core team. We also think that Ansh could improve their systems for using their M&E data, as well as their governance structure.26 We think it is plausible that these issues are characteristic challenges of an early stage organization that are resolvable. GiveWell's follow up on this grant will monitor core team development, use of M&E data, and organizational governance improvements.
- We limited our desk research for this investigation more than usual, so it is possible that we may have missed important considerations that we would have otherwise noticed. Our investigation in this case focused on qualitative considerations about Ansh’s program, relationships with stakeholders, organizational maturity, and ability to scale, and was informed by takeaways shared by Laliteswar Kumar, acting as a consultant on behalf of GiveWell, who visited Ansh’s program in Rajasthan, India on behalf of GiveWell. This reflects our key uncertainties about this opportunity prior to our investigation. Because we have already conducted a moderate level of research into KC through our existing work, and our key questions were about Ansh’s implementation and ability to scale, we spent a limited amount of time on desk research for this grant investigation compared to other similarly sized grants. However, because of this it is possible that we missed important considerations that have led us to overestimate the expected impact or cost-effectiveness of this program.
Plans for follow up
- We will hold calls with Ansh at least every 6 months and may request additional ad hoc or written check-ins, if needed.
- We will be specifically tracking program outcomes detailed in a shared learning agenda (unpublished).
- We plan to revisit our take on Ansh's program in early 2026, as part of considering whether we will extend or exit funding.
Internal forecasts
For this grant, we are recording the following forecasts:
| Confidence | Prediction | By time | Resolution |
|---|---|---|---|
| 50% | We will make an additional grant to Ansh. | By December 2026 | |
| 70% | Ansh will successfully onboard a senior operations hire by June 2025. | End of June 2025 | |
| 55% | Ansh will be operating in 15 facilities. | End of 2025 | |
| 60% | Ansh’s program will cost ≤$86 per infant once they are operating in 15 facilities. | By end of 2025 or whenever they complete expansion phase | |
Our Process
- We had three conversations with Supriya Bansal, the CEO of Ansh.
- Ansh shared, and we reviewed, a range of materials detailing their program.
- Laliteswar Kumar, acting as a consultant on behalf of GiveWell, visited Ansh in Rajasthan. He spent 3 days visiting facilities Ansh worked in, speaking with Supriya Bansal, Ansh’s nurses, non-Ansh hospital workers, and hospital officials.
- We created a BOTEC (back-of-the-envelope-calculation) and a Value of Information BOTEC to estimate the cost-effectiveness of the program and the learning value of this grant.
- We drafted a learning spreadsheet to share with Ansh to track learnings throughout the grant period.
Sources
- 1
formerly Charity Entrepreneurship
- 2
- See the “What is the program” section of our intervention report on KC.
- Ansh, PowerPoint presentation on Ansh’s KC program in Rajasthan, “The Solution” slide.
- Supriya Bansal, comments on a draft of this page, February 24, 2025.
- 3
See the “What is the program” and “Main findings of Cochrane meta-analysis on mortality and morbidity” sections of our intervention report on KC.
- 4
- Ansh’s total budget in 2025 and 2026 is $2,884,578.
- Ansh’s fundraising from other sources will cover a portion of this budget.
- GiveWell’s grant covers the remaining ~69% of Ansh’s budget for the two year period ($1,976,578 / $2,884,578 = 0.685).
- Ansh does not currently run programs outside of Rajasthan, India.
- 5
Ansh’s total budget for this two year period is $2,884,578, so this grant covers 69% of their total budget over this period.
- 6
- Ansh, Program Implementation Guideline, 2024.
- “The implementation of this program is driven by a collaborative partnership between Ansh and a designated implementing partner. Under the supervision of Ansh, the implementing partner executes all activities detailed in this program design. Ansh plays a crucial role in the daily operations of the program and also conducts ongoing Monitoring and Evaluation to ensure effectiveness and adherence to the standards.” p. 2.
- Ansh, All Data and Processes (unpublished).
- Ansh, PowerPoint presentation on Ansh’s KC program in Rajasthan, “The Process” slide.
- Ansh, Program Implementation Guideline, 2024.
- 7
- Ansh, PowerPoint presentation on Ansh’s KC program in Rajasthan, “The Set Up” slide.
- Ansh, Program Implementation Guideline, 2024.
- “Our program requires hiring one program coordinator and a team of nurses (the number of nurses is dependent on the size of the hospital and the prevalence of [low birth weight] infants).” p. 6
- “We provide backrests, plastic recliner chairs, a sling, cap, mittens, and gloves as well as a temperature monitoring watch for the baby, to facilitate ease of providing Kangaroo Mother Care to the baby.” p. 5
- “Training of the nurses will be arranged by Ansh and will cover the nursing guideline booklet. Ansh has an in-house trainer that will be training all the nurses in the facilities, will help nurses with troubleshooting, and hold regular online weekly counselling sessions.” p. 7.
- Ansh, M&E Coordinator Guideline, 2023 (unpublished)
- 8
Our understanding is that Ansh’s program is unlikely to be poaching or displacing health workers from government roles, because there is a high supply in Rajasthan of people with nursing skills, and relatively few government job openings. This understanding is based primarily on feedback from Laliteswar Kumar, acting as a consultant on behalf of GiveWell, who visited Ansh’s program and spoke with government officials.
- 9
Supriya Bansal, Co-founder and CEO, Ansh, Call with GiveWell, July 17, 2024 (unpublished)
- 10
Ansh told us that they also conduct group counseling sessions across different wards to provide comprehensive education on basic neonatal practices, and leverage their full-time presence in district hospitals to support policy related to discharge, neonatal care, Infection Prevention and Control, and overall maternal and newborn care. However, GiveWell did not investigate these components closely, as they were not a key part of this grant decision.
- 11
- Ansh, KC Nurse Guidelines, 2024 (unpublished)
- “After the beneficiary has been discharged, a nurse is expected to follow up with the beneficiary via a phone call, to check on the health of the baby, provide ad hoc support and repeated counselling. Every beneficiary receives 4 follow up calls:
- 2 days after discharge
- 7 days after discharge
- 14 days after discharge
- When the baby turns 28 days old
- “After the beneficiary has been discharged, a nurse is expected to follow up with the beneficiary via a phone call, to check on the health of the baby, provide ad hoc support and repeated counselling. Every beneficiary receives 4 follow up calls:
- Supriya Bansal, Co-founder and CEO, Ansh, Call with GiveWell, July 17, 2024 (unpublished)
- Ansh, KC Nurse Guidelines, 2024 (unpublished)
- 12
- ASHAs are trained female community health activists: “ASHAs support maternal care and immunization for children against vaccine-preventable diseases; tuberculosis, neglected tropical diseases, communicable disease prevention and control; and core areas of health promotion for nutrition, sanitation, and healthy living.” WHO, “India’s ASHA workers win Global Health Leaders Award.”
- Supriya Bansal, Co-founder and CEO, Ansh, Call with GiveWell, July 17, 2024
- 13
Ansh, All Data and Processes (unpublished)
- 14
See our back-of-the-envelope calculations (BOTEC) for the Ansh Kangaroo Care program, “2 year budget” sheet.
- 15
Adds together totals from each of the “Launch” and “Running” rows in Ansh’s 2025 budget.
- 16
Adds together totals from each of the “M&E” rows in Ansh’s 2025 budget.
- 17
$107,143 for M&E coordinators and $6,429 for fiscal sponsorship (total comes out to $113,571, rather than $113,572 due to rounding).
- 18
$248,671 for contingency costs and $149,202 to roughly account for possible inflation.
- 19
- 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.
- 20
Specifically, we estimate the following alternative scenarios:
- Our best guess is 13x. This uses our best guess of counterfactual mortality, Ansh’s estimate of mortality within their program, and a strict -40% IV adjustment because we are very uncertain about the true baseline mortality in this population.
- Our cost-effectiveness estimate would be 9x if we use our best guess of counterfactual mortality, and apply our estimate of the impact of KC (based on the literature) to estimate the mortality rate that would occur during the program.
- Our cost-effectiveness estimate would be 6x if we use Ansh’s estimate of mortality within their program, and apply our estimate of the impact of KC (based on the literature) to estimate the mortality rate that would occur without the program.
- Our cost-effectiveness estimate would be 11x if we use our estimate of counterfactual mortality assuming a higher mortality rate in Ansh facilities than statewide, based on IHME’s upper bound estimates. We apply our estimate of the impact of KC (based on the literature) to this estimate the mortality rate that would occur during the program.
- Our cost-effectiveness estimate would be 19x if we use our ‘best guess’ approach, but use an estimate of counterfactual mortality assuming a higher mortality rate in Ansh facilities than statewide, based on IHME’s upper bound estimates.
- 21
See our summary sheet for Ansh’s budget in our BOTEC for this grant.
- 22
- Our 12% estimate of mortality among low birth weight infants in Rajasthan is based on a number of modeled inputs from the global burden of disease, and we have not dug into how these inputs were derived (additional details in our BOTEC here).
- It’s possible that the counterfactual mortality rate of infants enrolled in Ansh’s program (i.e., what the mortality rate would be for those same infants if they were not enrolled in the program), is higher than the mortality rate for low birth weight infants in Rajasthan as a whole. This could occur if parents with the highest risk infants intentionally deliver or bring their infants to facilities where Ansh works, or if Ansh targets districts within Rajasthan with higher mortality rates than the average. In addition, Ansh doesn’t enroll all low birth weight infants in their program: their inclusion criteria specifically focuses on stable newborns with a birth weight of 2,200g or below, as well as babies admitted to the intensive care unit with a birth weight of 2,500g or below. This means that stable low birth weight newborns weighing between 2,200g and 2,500g are not enrolled in the program. This means that the counterfactual mortality of infants enrolled in Ansh’s program may be higher than for all low birth weight infants.
- Ansh told us via email that they specifically work, or plan to work, in high mortality districts in Rajasthan. However, not all of these districts have been selected yet, so we have modeled the cost-effectiveness of their program using aggregate data for Rajasthan. Ansh, Email to GiveWell, November 15, 2024 (unpublished).
- If the counterfactual mortality rate of Ansh’s patients were higher than 12%, this would suggest that we are underestimating the cost-effectiveness of the program.
- On the other hand, if we believe that the counterfactual mortality rate of Ansh’s patients is 12% (or even higher), and we believe Ansh’s measured mortality data, this would imply that Ansh’s program averts significantly more neonatal deaths than what is estimated by studies of other KC programs. We’re not certain whether that conclusion is plausible; it could suggest that baseline mortality is actually lower than we think.
- 23
See our intervention report on KC.
- 24
Data Ansh shared with us from their program, January to August 2024, implies a neonatal mortality rate of 6.3% for low birth weight infants who received their program. Ansh, Email to GiveWell, October 11, 2024 (unpublished).
- 25
Supriya Bansal, Co-Founder and CEO, Call with GiveWell, October 30, 2024 (unpublished)
- 26
- Regarding M&E improvements: While Ansh has been able to share detailed data with us and is collecting extensive data on the program (they embed an M&E coordinator at all facilities where they work), Laliteswar Kumar, acting as a consultant on behalf of GiveWell who visited the program, noted that there seem to be gaps in data analysis and in how M&E data was incorporated into decision-making.