The GiveWell Blog

December 2016 open thread

Our goal with hosting quarterly open threads is to give blog readers an opportunity to publicly raise comments or questions about GiveWell or related topics (in the comments section below). As always, you’re also welcome to email us at info@givewell.org or to request a call with GiveWell staff if you have feedback or questions you’d prefer to discuss privately. We’ll try to respond promptly to questions or comments.

If you have questions related to the Open Philanthropy Project, you can post those in the Open Philanthropy Project’s most recent open thread.

You can view our September 2016 open thread here.

AMF and Population Ethics

One important input we consider in our top charity recommendations is our quantitative cost-effectiveness estimate of each charity: an estimate of how much good the charity accomplishes per dollar. We have written before about the many challenges of constructing and interpreting such cost-effectiveness estimates. One such challenge is the problem of how to assign a numerical “score” to various good outcomes such as averted deaths or increased incomes, so that they can all be compared on the same scale. This involves answering questions such as:

  • What would I do if I had a choice between doubling the income of three individuals in extreme poverty for a year and averting the death of a child under the age of 5?
  • How many deaths of children under the age of 5 would I need to avert to accomplish as much good as averting the death of one adult?

GiveWell staff members enter their estimates for such values-based comparisons in our cost-effectiveness analysis (CEA). Along with empirical estimates of program costs and outcomes, this forms the basis for the numerical comparisons we make between our top charities.

Recently, a post on the Effective Altruism Forum raised concerns about our recommendation of AMF specifically. It argues that the value of averting deaths for children under five depends on one’s view of population ethics – a branch of philosophy that asks questions like “Is it good to avert a death if it has no long-run impact on the total population?” – and that some approaches to population ethics would imply a substantial discount to our cost-effectiveness estimates. We’ve chosen to respond to this argument at length because we think it is interesting, serves as a good example of the thorny issues we grapple with in estimating cost-effectiveness, and gives the opportunity to explain some aspects of our 2016 cost-effectiveness model that are not widely understood.

In this post, I will

  • Explain basic concepts in population ethics and how they inform the way people think about the value of averting death
  • Summarize arguments from the post mentioned above, which argues that people with certain views on population ethics should substantially discount our cost-effectiveness estimate of the Against Malaria Foundation (AMF) to better reflect their values
  • Walk through the reasoning behind our current estimate of AMF’s cost-effectiveness and explain why I believe it’s compatible with most plausible accounts of population ethics, so discounts as aggressive as those suggested in the linked post are likely inappropriate.

Continue reading “AMF and Population Ethics”

Staff members’ personal donations for giving season 2016

For this post, GiveWell staff members wrote up the thinking behind their personal donations for the year. We made similar posts in 2013, 2014, and 2015. After Elie and Holden, staff are listed in order of their start dates at GiveWell.
Continue reading “Staff members’ personal donations for giving season 2016”

Why I mostly believe in Worms

The following statements are true:

  • GiveWell is a nonprofit dedicated to finding outstanding giving opportunities through in-depth analysis. Thousands of hours of research have gone into finding our top-ratepd charities.”
  • GiveWell recommends four deworming charities as having outstanding expected value. Why? Hundreds of millions of kids harbor parasitic worms in their guts[1]. Treatment is safe, effective, and cheap, so much so that where the worms are common, the World Health Organization recommends administering pills once or twice a year to all children without incurring the cost of determining who is infected.
  • Two respected organizations, Cochrane and the Campbell Collaboration, have systematically reviewed the relevant studies and found little reliable evidence that mass deworming does good.

That list reads like a logic puzzle. GiveWell relies on evidence. GiveWell recommends mass-deworming charities. The evidence says mass deworming doesn’t work. How is that possible? Most studies of mass deworming track impact over a few years. The handful that look longer term find big benefits, including one in Kenya that reports higher earnings in adulthood. So great is that benefit that even when GiveWell discounts it by some 99% out of doubts about generalizability, deworming charities look like promising bets.

Still, as my colleagues have written, the evidence on deworming is complicated and ambiguous. And GiveWell takes seriously the questions raised by the Cochrane and Campbell evidence reviews. Maybe the best discount is not 99% but 100%. That would make all the difference for our assessment. This is why, starting in October, I delved into deworming. In this post and the next, I will share what I learned.

In brief, my confidence rose in that Kenya study’s finding of higher earnings in adulthood. I will explain why below. My confidence fell in the generalizability of that finding to other settings, as discussed in the next post.

As with all the recommendations we make, our calculations may be wrong. But I believe they are reasonable and quite possibly conservative. And notice that they do not imply that the odds are 1 in 100 that deworming does great good everywhere and 99 in 100 that it does no good anywhere. It can instead imply that kids receiving mass deworming today need it less than those in the Kenya study, because today’s children have fewer worms or because they are healthy enough in other respects to thrive despite the worms.

Unsurprisingly, I do not know whether 99% overshoots or undershoots. I wish we had more research on the long-term impacts of deworming in other settings, so that we could generalize with more nuance and confidence.

In this post, I will first orient you with some conceptual and historical background. Then I’ll think through two concerns about the evidence base we’re standing on: that the long-term studies lack design features that would add credibility; and that the key experiment in Kenya was not randomized, as that term is generally understood.

Continue reading “Why I mostly believe in Worms”

Our updated top charities for giving season 2016

We have refreshed our top charity rankings and recommendations. We now have seven top charities: our four top charities from last year and three new additions. We have also added two new organizations to our list of charities that we think deserve special recognition (previously called “standout” charities).

Instead of ranking organizations, we rank funding gaps, which take into account both charities’ overall quality and cost-effectiveness and what more funding would enable them to do. We also account for our expectation that Good Ventures, a foundation we work closely with, will provide significant support to our top charities ($50 million in total). Our recommendation to donors is based on the relative value of remaining gaps once Good Ventures’ expected giving is taken into account. We believe that the remaining funding gaps offer donors outstanding opportunities to accomplish good with their donations.

Our top charities and recommendations for donors, in brief

Top charities

We are continuing to recommend the four top charities we did last year and have added three new top charities:

  1. Against Malaria Foundation (AMF)
  2. Schistosomiasis Control Initiative (SCI)
  3. END Fund for work on deworming (added this year)
  4. Malaria Consortium for work on seasonal malaria chemoprevention (added this year)
  5. Sightsavers for work on deworming (added this year)
  6. Deworm the World Initiative, led by Evidence Action
  7. GiveDirectly

We have ranked our top charities based on what we see as the value of filling their remaining funding gaps. We do not feel a particular need for individuals to divide their allocation across all of the charities, since we are expecting Good Ventures will provide significant support to each. For those seeking our recommended allocation, we recommend giving 75% to the Against Malaria Foundation and 25% to the Schistosomiasis Control Initiative, which we believe to have the most valuable unfilled funding gaps.

Our recommendation takes into account the amount of funding we think Good Ventures will grant to our top charities, as well as accounting for charities’ existing cash on hand, and expected fundraising (before gifts from donors who follow our recommendations). We recommend charities according to how much good additional donations (beyond these sources of funds) can do.

Other Charities Worthy of Special Recognition

As with last year, we also provide a list of charities that we believe are worthy of recognition, though not at the same level (in terms of likely good accomplished per dollar) as our top charities (we previously called these organizations “standouts”). They are not ranked, and are listed in alphabetical order.

Below, we provide:

  • An explanation of major changes in the past year that are not specific to any one charity. More
  • A discussion of our approach to room for more funding and our ranking of charities’ funding gaps. More
  • Summary of key considerations for top charities. More
  • Detail on each of our new top charities, including an overview of what we know about their work and our understanding of each organization’s room for more funding. More
  • Detail on each of the top charities we are continuing to recommend, including an overview of their work, major changes over the past year and our understanding of each organization’s room for more funding. More
  • The process we followed that led to these recommendations. More
  • A brief update on giving to support GiveWell’s operations vs. giving to our top charities. More

Continue reading “Our updated top charities for giving season 2016”

New Incentives update

We’re planning to release updated top-charity recommendations in mid-November, and one of the questions our staff has been debating recently is whether to recommend New Incentives as a top charity.

We’ve decided that New Incentives doesn’t currently meet our criteria for a top charity because its program doesn’t have sufficient evidence supporting it. However, we have been extremely impressed with and think very highly of New Incentives’ staff and are considering how best to support them in the future and incentivize others to found an organization like they did.

In this post, we summarize the answers to the key questions we asked to determine whether New Incentives meets our criteria for a top charity recommendation and the options we’re considering for future support.

Background

New Incentives operates a conditional cash transfer (CCT) program in Nigeria to incentivize pregnant women to deliver in a health facility. New Incentives originally intended its CCT program to focus primarily on prevention of mother-to-child transmission (PMTCT) of HIV. However, under this model the program did not reach enough HIV-positive pregnant women to justify its operating costs, and in 2015, New Incentives expanded its program to target both HIV-positive women and HIV-negative women.

New Incentives was the first organization we supported as part of our experimental work to support the development of future top charities. It has been about two and a half years since New Incentives received its initial grant, and it now has a long enough track record implementing its program to be considered for a top charity designation.

Is New Incentives’ intervention evidence-backed?

New Incentives’ impact is made up of three components: (a) delivering cash to very poor people, (b) incentivizing HIV-positive pregnant women to deliver in clinics and get the medicines that prevent mother-to-child transmission of HIV, and (c) incentivizing pregnant women to deliver their babies in a health facility.

Because a relatively small portion of New Incentives’ beneficiaries are HIV-positive, because it costs New Incentives more than GiveDirectly to deliver each dollar, and because it is likely reaching individuals with higher incomes than GiveDirectly does, the impact that has the dominant effect on our view about whether or not New Incentives meets the standard we have for a top charity’s cost-effectiveness is the impact of facility delivery on neonatal mortality.

The evidence we have for the impact of facility delivery comes from (1) relevant randomized controlled trials (RCTs), (2) monitoring that New Incentives carries out, and (3) non-RCT evidence on the impact of facility delivery.

Overall, the evidence from the RCTs increases our confidence that an intervention that offers improved neonatal care could have a significant impact on neonatal mortality, but the evidence we have seen and New Incentives’ current monitoring of its program is insufficient to convince us that increasing the number of women who deliver at facilities has a similar impact.

Randomized controlled trial evidence

Two RCTs of low-intensity training programs for traditional birth attendants found significant (30-45%) reductions in neonatal mortality. These interventions are different than New Incentives’ intervention but may have a similar effect since they aim to increase the knowledge of traditional birth attendants so that they offer similar care to that which is offered in health facilities. We did not find any RCTs on facility delivery itself; these two RCTs are the most similar ones to New Incentives’ program that we identified. The interventions varied:

  • In Gill et al. 2011, the intervention group received training and supplies related to common practices to reduce neonatal mortality immediately following birth. The study observed significant differences between the treatment and control group on practices such as drying the baby with a cloth and then wrapping it in a separate blanket (as opposed to using the same blanket), clearing the baby’s mouth and nose with a suction bulb (instead of a cloth), and using a pocket resuscitator (instead of mouth to mouth) (see Table 5, Pg. 8).

    We have not closely vetted this study but note some significant-seeming differences between the treatment and control birth attendants–in particular, the treatment group had significantly more education than the control group (see Table 1, Pg. 4).

  • In Jokhio et al. 2005, the intervention group received supplies and 3 days of training focused on antepartum, intrapartum, and postpartum care, including activities such as: “how to conduct a clean delivery; use of the disposable delivery kit; when to refer women for emergency obstetrical care; and care of the newborn.” The intervention group was “asked to visit each woman at least three times during the pregnancy (at three, six, and nine months) to check for dangerous signs such as bleeding or eclampsia, and to encourage women with such signs to seek emergency obstetrical care.”

New Incentives’ monitoring

New Incentives’ staff interviews a nurse and conducts additional inspection at each health facility it considers working with. New Incentives reports the results of these interviews. Two questions are most relevant to our assessment of the similarity between the interventions studied in the RCTs discussed above and the care offered in facilities New Incentives works with.

New Incentives asks nurses at each health facility: 1) “What multiple steps do you take immediately after delivery?” and 2) “What are the essential steps immediately after birth in ensuring that the baby can breathe and is warm?”

For the first question, New Incentives counts how many of the following steps nurses say they take (without being prompted by the New Incentives staff member asking the question): a) Dry baby with cloth, b) Slightly rub baby, c) Clear airways, d) Use air mask if necessary, e) Regulate temperature (put on mother’s belly), f) Don’t know/refused to answer. For the second question, New Incentives captures a free form answer.

We have limited information about the differences in practices between the intervention and control groups in Jokhio et al. 2005, but we do have this information for Gill et al. 2011. (See Gill et al. 2011, Table 5, Pg. 8.) It does not appear that the way New Incentives evaluates answers to its first question can tell us whether nurses in the facilities with which it works follow the improved practices from Gill et al. 2011.

We aggregated the answers to the second question, and 17 of 54 answers explicitly mentioned using a bulb syringe or mucus extractor, which we would guess is equivalent to clearing the baby’s mouth and nose with a suction bulb in Gill et al. 2011 (another 11 mentioned ‘clear airways’ or ‘suck’ which might refer to the procedure used in Gill et al. 2011). We were not able to get additional relevant information from nurses’ answers to the second question.

New Incentives does not appear to ask questions that fully address the other major difference between the intervention and control groups in Gill: use of a resuscitation intervention.

The intervention offered by Jokhio et al. 2005 includes antenatal care in addition to intrapartum and postpartum care, and we don’t know what impacts each part of the intervention had.

Note that New Incentives does not systematically collect data on the type of care women who enroll in its program would have received had they not delivered in a facility, though it has done some limited surveys of traditional birth attendants in the areas it works in.

Non-randomized evaluations of the impact of facility delivery

We have not carefully reviewed these studies, and the studies we identified found mixed effects (including some studies finding higher neonatal mortality in facilities) but we have major questions about these studies’ ability to assess facilities’ causal impact.[1] In particular, women may be more likely to go to a facility for childbirth when they are experiencing complications, which could bias the results.

What is our best guess about New Incentives’ cost-effectiveness?

The most important questions in assessing New Incentives’ cost-effectiveness are (a) the impact its cash transfers have on rates of facility delivery and (b) the impact that increased facility delivery has on neonatal mortality.

New Incentives is conducting an RCT of its impact on (a) and preliminary results indicate that it had a significant impact on facility deliveries: 48% of women in the treatment group (i.e., all those who were offered the opportunity to enroll in the program even if they chose not to do so) delivered in a facility versus 27% in the control group. However, there are differences between the program studied by New Incentives’ RCT and its current program; the RCT only targeted HIV-positive women, so some portion of the impact may be attributable to educating women about the importance of PMTCT. The program studied in the RCT also provided larger cash transfers than New Incentives will provide in its ongoing program: the program originally gave 6,000 naira (approximately 19 US dollars) for enrollment, 20,000 naira for delivery, and 6,000 naira for an HIV test; the program currently gives 1,000 naira for enrollment and 10,000 naira for delivery.

As noted above, we have very limited information to rely on when forming an estimate of the impact of facility delivery on neonatal mortality, and we do not see the evidence from the RCTs described above as particularly relevant or informative.

However, in trying to arrive at our best guess of the impact of the program, we also considered the facts that:

  • The interventions described in Gill et al. 2011 and Jokhio et al. 2005 are relatively low cost and of limited intensity, and they find significant decreases in neonatal mortality. This increases the plausibility that merely referring women to facilities for childbirth could have a similar, significant impact.
  • Our intuition (supported by what appears to be conventional wisdom in the global health community) strongly implies that delivering in a facility (in general, without respect to the specific facilities New Incentives works with in Nigeria) is likely to lead to lower mortality than alternatives.

Philosophical value judgments

Based on the results from the RCTs, we would expect New Incentives’ program to primarily prevent deaths of very young children (largely those within the first days or week of life). In internal, staff discussions about New Incentives, we have asked ourselves how we value the lives of newborn children vs. the lives of those saved by malaria nets (the other life-saving intervention we currently recommend). We have not completed a thorough assessment of the ages at which people die from malaria, but our impression is that the median age of death is approximately 1.[2]

We believe there is no “right” answer to this question, but depending on one’s values, the answer could have a significant impact on the relative cost-effectiveness of New Incentives vs. the Against Malaria Foundation, and by extension our other top charities.

Key considerations include:

  • One could simply sum the number of remaining years of life lost due to a death of a newborn vs a 1-year-old.
  • One could focus solely on lives saved and treat all lives as equivalent.
  • One might say that families and society have invested more in 1-year-olds and that 1-year-olds have more self-awareness and “personhood” than newborns, leading to valuing the 1-year-old more than the newborn.

Primarily for the last reason, the GiveWell staff who participated in these discussions tend to value 1-year-old lives over newborns, though our relative weights vary considerably.

Best guess cost-effectiveness estimate

Ultimately, we don’t have enough information to arrive at a reliable estimate of the impact of facility delivery on neonatal mortality. Our best guess is extremely rough, based primarily on intuitions formed based on limited data, and one that could easily shift significantly. We asked all staff who primarily work on GiveWell research to (a) guess the likely effect of New Incentives’ program on neonatal mortality and (b) enter the philosophical values discussed above. This yielded a median staff estimate that New Incentives was approximately as cost-effective as cash (in GiveDirectly’s program). Our cost-effectiveness model is here (.xlsx).

Is New Incentives transparent?

Yes – extremely. New Incentives has shared all of the information we have requested (and more) in a timely fashion. We feel that it is as good as any other organization we have ever engaged with on this criterion.

Options we’re considering for future support of New Incentives and/or its staff

We have discussed each of the following options with New Incentives and plan to let New Incentives’ preference drive our decision about which one to choose. In considering these options, we took into account (a) the likely direct impact funding would have and (b) the incentives that funding would create for others considering starting a new organization like New Incentives.

  1. Recommend that Good Ventures (a foundation with which we work closely that has provided past funding for our experimental work) provide an “exit grant” of approximately $1.2 million to New Incentives. New Incentives relied heavily on funding we recommended in its scale up, and abruptly stopping funding could cause it significant harm. Our impression is that funders often give grantees exit grants to offer them time to comfortably adjust their plans for fundraising and spending; this has been GiveWell’s experience with support from institutional funders. We would plan to benchmark our recommendation to the level of support New Incentives could have expected from us over the next two years (January 2017 – December 2018) as of the last time Good Ventures made a grant (March 2016). $1.2 million represents half what we would have projected New Incentives spending to be in 2017 and 2018 as of March 2016. (It grew faster than we expected since March 2016, so this is less than 50% of its projected operating expenses.)
  2. Recommend that Good Ventures agree to support some portion of New Incentives’ ongoing operations and a randomized controlled trial of New Incentives’ program’s impact on neonatal mortality. New Incentives’ program doesn’t seem cost-effective enough that we’d be willing to recommend that Good Ventures fully fund an RCT and New Incentives’ ongoing operations, but we’d consider recommending some, significant support (very roughly, we’d cap a recommendation at 50% of the total cost) if New Incentives could raise the rest of the funding elsewhere. This option would provide New Incentives with the opportunity to demonstrate that its program is more effective/cost-effective than we currently expect it to be as long as it is able to convince other funders to provide some support as well.
  3. Provide support to New Incentives/the New Incentives team to do something new. If New Incentives or its staff were interested in starting a new charity aiming to be a GiveWell top charity or significantly changing its program to focus on something more cost-effective, we would recommend that Good Ventures provide support.

We hope to decide soon about which option to pursue.

[Added December 19, 2016: GiveWell’s experimental work is now known as GiveWell Incubation Grants.]

Notes
[1] We identified two relevant meta-analyses. Chinkhumba et al. 2014, a meta-analysis of six prospective cohort studied of perinatal mortality in sub-Saharan Africa found 21% higher perinatal mortality in home deliveries compared to facility deliveries (OR 1.21 [1.02-1.46]) using a fixed-effects model, but this difference was not significant using a random effects model (OR 1.21 [0.79-1.84]).

We are also concerned that studies limited to the perinatal period may not capture longer-term neonatal effects. Tura et al. 2013, a meta-analysis of 19 studies (of various methodology) of the effect of facility delivery on neonatal mortality, found mixed results. Pooled results from low- and middle-income countries showed 29% reduction in risk of neonatal death associated with facility delivery. However, results of the studies were highly heterogeneous. Of the 8 studies in sub-Saharan Africa, 4 found effect near the pooled mean, and the other 4 did not find a statistically significant effect. (Of the four that did not find a significant effect, two studies found a nonsignificant effect close to the pooled mean of all studies, and two found no effect.)

A retrospective study based on the demographic and health surveys in Nigeria found that facility delivery is associated with increased neonatal mortality (adjusted odds ratio 1.28 [1.11-1.47], Fink et al. 2015, Figure 1, Pg. 5).

[2] Here is one paper we found. We have not vetted this paper. The simple average age of death in it is approximately 1.2 years (see Table 1).