Iodine deficiency has been linked with child mortality as well as permanent cognitive debilitation. Some references are on page 25 of the 2004 Copenhagen Consensus report on malnutrition; a WHO report paints a more vivid qualitative picture (page 6):

In areas of iodine deficiency, where thyroid hormone levels are low, brain development is impaired. In its most extreme form, this results in cretinism, but of much greater public health importance are the more subtle degrees of brain damage and reduced cognitive capacity … the mental ability of ostensibly normal children and adults living in areas of iodine deficiency is reduced … there is little chance of achievement and underdevelopment is perpetuated. Indeed, in an iodine-deficient population, everybody may seem to be slow and rather sleepy.

How do you and I avoid these disorders? The answer isn’t as simple as you might think. Within the last century, iodine deficiency was common in the U.S., and it has been combated using an explicit and sustained public health effort. The following is taken from the Micronutrient initiative’s global progress report:

The element iodine was discovered in 1811, but almost a century passed before it was established that lack of iodine caused the swelling of the thyroid gland commonly known as goitre … In the United States, the alarm was first raised in Michigan in 1918 when it was revealed that over 30% of men medically examined for war service had been found to have an enlarged thyroid. Many were declared unfit for service. By 1923 an Iodised Salt Committee had been formed, including physicians and representatives of the Salt Producers Association …. Later that same year, the Morton salt company began marketing iodised table salt nation-wide … by 1932 iodised salt accounted for 90% to 95% of all sales.

Iodine deficiency disorder is now extremely rare in the U.S., but it is still common in many other parts of the world (see the WHO 2004 report on iodine status worldwide). The effects on cognition and economic growth are widely unknown but potentially disastrous - and this is a problem that isn’t necessarily going away by itself, or going away as soon as wealth increases. (Especially if there’s a direct link between iodine deficiency and productivity.)

Iodizing salt may not have the same visceral appeal as water-related programs, but it’s a proven way to solve a truly debilitating problem and reduce both death and poverty. And it’s an area where there’s arguably no substitute for large-scale public health programs. Note that the Copenhagen Consensus rates it as the 3rd most cost-effective use of funds.

We were referred to the Against Malaria Foundation today. It’s the most transparent developing-world aid organization we’ve seen to date.

AMF grants funds to NGOs that distribute insecticide-treated nets (ITNs) to fight malaria. ITN distribution is a well-established method for saving lives and lowering the burden of malaria (see the “Insecticide-treated nets” section of the malaria chapter in the Disease Control Priorities Project).

But what sets AMF apart from other organizations carrying out similar programs is that they publish all the details for all of their projects.

This map lists all programs, by country, date, partners, and status (whether the nets are en route, have arrived, are being distributed, etc.) Click on an individual program and you can see the full proposal that was submitted and funded (and the proposals are quite concrete and specific), and, when applicable, other materials - including the post-project report and even shipping records. Pictures and videos are also available. Here’s an example of a project in progress; here’s a completed project.

Finally, you can see which donors’ gifts were used to fund the project. There’s been a lot of excitement over organizations such as Kiva and DonorsChoose because of their ability to link donors’ money directly with people served. This organization does something very similar, but it’s distinguished in that the types of projects it’s linking donors up with (bednet distribution projects) are larger-scale projects serving many people at once, and are well documented (see link above) in terms of their impact on human lives.

We’re going to have to spend more time with all of this information to check how the actual cost-effectiveness is living up to the top-line numbers. What we know now is that if all charities were as transparent as AMF, our job would be a lot easier.

We spoke at Google (Mountain View office) on Wednesday. Elie discussed our project and its history, and I presented our chief research findings from year 1.

We’re on a pledge-raising trip in CA, and our schedule has been tighter than anticipated. We’ll be back on the blog Monday.

In addition to choosing straightforward, cost-effective program-based interventions, we’re also planning to identify two regions to study in detail.

I’ve focused on finding priority region 1: an area that has both high and drastic disease burdens and relatively stable, low-corruption environments in which aid can likely be productive. This post details the criteria I used to narrow the field to a few finalists. (Note: We recently spoke to a private philanthropic advisor about this, and our methodology is quite similar to theirs.)

Corruption and poor governance

I used the latest versions of the Failed State Index (from the Fund for Peace and Foreign Policy Magazine) and Corruptions Perception Index (from Transparency International) to exclude countries where governance would be a large issue. In both cases, I removed countries with the worst rating on either index. (The FSI ranks 177 countries; 31 received the worst rank. The CPI ranks 180 countries; 38 received the worst rank.)

Disease burden and poor development

I used the 2007-08 United Nations Human Development Index, considering only countries that are marked “Low human development” (received by 22 of the 177 countries rated). Of those 22, only 9 passed the corruption/governance screen.

The UNHDI considers life expectancy (to assess health burden) but doesn’t consider morbidity. So, I also looked at DALY burden per 100,000 people (data from the WHO’s Burden of Disease Project) and include countries in the top decile of health burdens. This criterion excluded Senegal, Benin, and Eritrea.

That left me with 6 countries:

• Zambia
• Mozambique
• Mali
• Rwanda
• Tanzania
• Burkina Faso

Charitable activity in each country

We need to find a region that not only has both high and drastic disease burdens and relatively stable, low-corruption environments in which aid can likely be productive but also a region with a significant amount of aid being given (since our role is to identify outstanding charities rather than to create new ones, we’d rather work somewhere where we’ll have more options). One proxy for this idea is the total aid given by OECD countries to each of our six finalists (data from OECD). This measure is far from perfect since it reflects government aid, not private philanthropy. We also conducted a Guidestar search for the name of each country, and use the number of results as an approximation to the number of US-registered NGOs working there. The data is below.

Notes:

1. Population data comes from Wikipedia’s List of Countries by Population page.
2. OECD aid is the average annual aid over the last 5 years

These criteria would seem to indicate Zambia, Mozambique and Rwanda as our top contenders. We are also strongly considering Tanzania, as there may be superior availability of data on health and living conditions through the Living Standards Measurement Study.

Our plan now is to talk with our advisors for a) their thoughts on our process thus far and b) their ideas for picking a focus region from this list. We’re also planning to examine the Poverty Strategy Reduction Paper for each of these countries to get a better sense of the major challenges and available information there).

(Note: for the intrepid, you can access an Excel file with all the data used for this post here.)

Below is our basic outline for identifying the best options for donors within developing-world health. (Economic empowerment and possibly other developing-world causes will come after we have completed our report on direct health interventions.)

Focus on two priority regions and several priority interventions

As this post explains, we are simultaneously taking two approaches: a region-based approach (studying a particular region in depth, and aiming to fund interventions targeted to the specific needs of the region) and a program-based approach (identifying particularly promising interventions and funding their scale-up throughout the world).

Priority interventions: focus on the most proven, scalable, and cost-effective

Our general criteria for selecting program-based interventions are discussed here. To start, we are prioritizing what we call straightforward interventions, or interventions that can be carried out infrequently and for which the burden of monitoring and evaluation is relatively low. As discussed here, these interventions are also rated among the most cost-effective interventions by existing studies of cost-effectiveness. They are:

1. Vitamin supplementation programs, providing nutrients such as vitamin A (which reduces the risks of infant mortality, development of blindness, and deaths in childbirth - reference here).
2. Mass drug administration programs, particularly those aiming to treat school-age children who suffer from intestinal parasites. (Such interventions have been shown to improve school attendance and likely nutritional status.)
3. Vaccination campaigns to cost-effectively save lives.
4. Vitamin fortification programs such as the iodization of salt.
5. Surgical programs to correct deformities such as cleft palate and fistula.

We may add other priority interventions, particularly if we see particularly strong donor demand for interventions with fundamentally different goals.

Priority regions: help those in great need and those with great potential

Our region-based research will be less focused on strict definitions of “cost-effectiveness” and more on finding programs that address a region’s needs broadly and holistically, somewhat along the lines of Partners in Health (and more in line with the vision of the WHO commission on macroeconomics and health than that of the Copenhagen Consensus). The first step will be picking two priority regions.

Priority region 1: one priority region will reflect a goal of helping those in the most need. We are currently in the process of identifying countries that have both high and drastic disease burdens and relatively stable, low-corruption environments in which aid can likely be productive. Preliminarily, our strongest candidates are:

• Zambia
• Mozambique
• Mali
• Rwanda
• Tanzania
• Burkina Faso

The process for identifying these six countries is detailed here.

Priority region 2: the other priority region will reflect more of a triage approach: identifying an area where health conditions are poor and help is needed but strong economic opportunities exist for those at a reasonable level of personal health and productivity. Preliminarily, we are guessing that our work in this area will likely focus on somewhere in India, but we have not yet completed this analysis.

We know of two large-scale, systematic projects devoted to rating the cost-effectiveness of different health interventions. Both use an approach centered on disability-adjusted life-years (DALYs), which is not the only measure of cost-effectiveness we intend to use, but they’re a start.

One is the Disease Control Priorites report. We are currently in the process of constructing a full summary of the interventions and cost-effectiveness estimates discussed in the report, as the summary provided by the report itself appears incomplete. From looking at the average cost-per-DALY-averted of the existing list, however, it appears that three of the five most cost-effective interventions (including the top two) fall under what we call straightforward interventions: vaccination campaigns and mass drug administration (deworming).

The other is the 2008 Copenhagen Consensus, which explicitly ranks interventions by the desirability of funding them. Four of the top six (including #1) fit within our straightforward interventions: vitamin supplementation, vitamin fortification, vaccination campaigns and mass drug administration (deworming).

From our scan of the Disease Control Priorites Report, we also believe that our straightforward interventions will prove top candidates for cost effectiveness by other criteria. In particular, we noted vaccination campaigns as having the lowest “cost per death averted” (#4) in the report, which likely corresponds to the lowest “cost per life-year saved” (#3) as well since most deaths averted are infant deaths. We also feel that surgery may be the strongest intervention we’ve seen in terms of averting extreme suffering.

We therefore consider the “straightforward interventions” to be priority interventions, and are planning on looking into them actively. They will not necessarily be the only priority interventions we name, but they stand out given what we know.

Again, these interventions are:

1. Vaccination campaigns.
2. Mass drug administration programs, including albendazole for treating helminths (i.e., deworming).
3. Vitamin supplementation programs, with nutrients that need be taken only infrequently (particularly vitamin A).
4. Vitamin fortification programs, such as the iodization of salt.
5. One-time surgery programs, along the lines of Interplast’s surgical team trips.

As discussed previously, we are looking for (program-based) interventions that are both proven and scalable. The Disease Control Priorities report lists many interventions that are “proven,” in the sense that one or more studies have been done indicating that the program has improved health outcomes in the region. However, showing that a program has worked once - or even several times - is far from showing that it will work again. Not only could the context, circumstances, and clients change, but the people running the program (and the scrutiny they’re under) most likely do change from a studied experiment to a less thoroughly tracked replication.

In examining the case for different interventions (see the list of interventions here), we have noted that some types of programs are inherently more “straightforward” than others, in the sense that a lower burden of proof is necessary to extrapolate from their past effects to their future effects. For example:

• Many programs focus on educating clients, and their success thus depends on sustained and difficult-to-track behavior change from locals. Even if a sanitation program has successfully reduced diarrhea incidence rates in one region, bringing it to a new region means dealing with a new group of people, who may respond completely differently to the same education techniques. Without thorough and ongoing monitoring and evaluation, we feel it is difficult to be confident that a program along these lines is continuing to work.
• Other programs, such as DOTS for tuberculosis, focus on improving medical care. They may depend on sustained behavior change from local medical professionals, but not from the population in general. The success of such programs will depend heavily on the existing health care infrastructure, qualifications of health professionals, and quality of training. We feel it is difficult to have confidence in such programs without continually tracking patient outcomes.
• Other programs, such as vaccinations, do not rely on sustained behavior change from anyone in the local population. A vaccination campaign in a given region can be carried out fairly infrequently (i.e., once a year) as long as the vaccines administered are legitimate and a large number of people can be gathered and treated. Both of these factors are relatively straightforward to track and report.

Below are all the interventions we’re aware of that fit in the last category: interventions that can be carried out relatively infrequently (i.e., every 6-12 months), and for which all necessary behavior change can be directly observed and reported by the people carrying them out. For each intervention, we briefly characterize what information would need to be reported in order to reasonably extrapolate from the intervention’s past, studied effects on health outcomes.

1. Intervention: vaccination campaigns. Information needed: number and age of people vaccinated; disease incidence/prevalence estimates for the region.
2. Intervention: Mass drug administration programs, including albendazole for treating helminths and ivermectin for treating onchocerciasis and lymphatic filariasis (more on these conditions in future posts). Information needed: number and age of people treated; disease incidence/prevalence estimates for the region.
3. Intervention: Vitamin supplementation programs, with nutrients that need be taken only infrequently (particularly vitamin A). Information needed: number and age of people treated; estimates of vitamin A deficiency prevalence for the region.
4. Intervention: Vitamin fortification programs, such as the iodization of salt. Information needed: amount and circulation of food fortified; estimates of iodine and other deficiency prevalence for the region.
5. Intervention: One-time surgery programs, along the lines of Interplast’s surgical team trips. Information needed: Condition treated for each client; completeness of surgery (before-and-after pictures would capture this information).

In labeling interventions “straightforward,” we are not claiming that they are easy to carry out or that they will always work. However, all else equal, we find them more promising for our purposes than other program-based interventions. This is largely because our experience to date with developing-world aid has shown that thorough, high-cost monitoring and evaluation (of the sort that could track sustained behavior change, for example) is relatively rare. We believe we are most likely to be able to confidently recommend interventions along the lines of those listed above, for which the necessary burden of monitoring and evaluation is lower.

In our analysis of bednet distribution programs, we considered the likelihood that a distributed bednet was ultimately used for its intended purpose: to protect against malaria.

The Malaria Matters blog recently posted links to numerous examples of nets used for other purposes:

Net stories include use for fishing in Zambia, as bridal veils in Zambia and other countries and trapping edible ants in Uganda. These problems arise when LLIN distribution programs focus on the wrong numbers. It is not enough to say how many hundreds of nets have been distributed in a community. The real concern is whether they are used correctly and for the intended purpose.

The post also notes that:

The three most popular reasons for using bednets to dry fish were: fish dry faster on these nets, they don’t stick and not surprisingly, these nets are cheaper.

On the one hand, it’s important for donors to know what their donations accomplish, and if donors’ aim is to prevent malaria, the above stories provide evidence of the types of problems that can occur with unmonitored aid. On the other hand, if people are buying nets and using them in other ways that improve their lives - even if it’s not malaria prevention - it’s not clear to me that that impact isn’t worth considering as well.

The Disease Control Priorities report has a summary section (pg 60-85) listing interventions, along with cost-effectiveness estimates (in disability-adjusted life-years per US$) and some other basic info (target population, required infrastructure, etc.) We’ve created an Excel version of the list that we will be referring to in future posts:

List of interventions from Disease Control Priorities projiect (XLS)

This list is incomplete, in the sense that it does not list all of the interventions (even the recommended interventions) in the report. We aren’t sure why this is (and neither is the only DCP author we’ve spoken to so far). We will be using our notes on the report to add all interventions in over time.

Most numbers below from this table (2000 data).

• Malaria kills about 1.1 million people per year in developing countries.
• ~65% are 4 years old or younger. (This particular figure appears to contradict the data from the Global Burden of Disease report pg 126-7, which implies a proportion closer to 90%).
• The burden of malaria goes far beyond mortality, as the vast majority of number of cases are not fatal. Cases per year are estimated at ~200 million, lasting an average of ~4 days each.
• Malaria both exacerbates and is exacerbated by malnutrition (see pgs 415-417 of the Disease Control Priorities report).
• Malaria can, but usually does not, lead to permanent non-fatal debilitation including partial paralysis, quadriparesis, hearing and visual impairment, behavioral difficulties, language deficits, and epilepsy. Estimates for the numbers of these conditions caused by malaria total 13,000-15,000 cases worldwide per year.

Broadly, I would say that fighting malaria will reduce infant mortality and lower the overall burden on the local economy, health care system, and day-to-day quality of life, though it will not have much direct effect on adult mortality/morbidity. It’s therefore most relevant to goals 1-4 of this list.

Using Global Burden of Disease data, I put together a quick look at mortality in lower- and middle-income countries (LMICs) by age group. This is particularly important when seeking interventions that focus on adult mortality, one of the goals from this list.

Burden of mortality in LMICs by age group

All the way on the right of the table is the proportion of deaths that different conditions cause in each age group. (Row 4 gives each age group’s mortality as a proportion of total LMIC mortality.) Yellow coloring means that the condition accounts for 5%-10% of all the mortality in that age group; orange means 10-20%; red, greater than 20%. My notes (chapter and page references are to the Disease Control Priorities report):

• More than 20% of all LMIC deaths happen before the age of five (also see the pie chart in our developing world summary). Of these deaths, a total of 75% come from one of the following:
  • Perinatal conditions account for more than 20%. Better maternal care, as well as micronutrient supplementation for expectant mothers, could substantially reduce this burden (Chapter 26).
  • Lower respiratory infections (including pneumonia and influenza) account for close to 20%, even though vaccines can be highly effective against these diseases (pg 485-6). Other vaccine-preventable diseases account for an additional 10%.
  • Diarrhea accounts for another 15% of these deaths. Even rudimentary medical care (such as the use of oral rehydration therapy) can prevent such deaths (pg 378).
  • Malaria accounts for another 10%.
• Mortality between the ages of 5 and 14 is far less common. The biggest causes are accidents (25%), childhood-cluster (generally vaccine-preventable) diseases (15%), respiratory infections (~10%), and HIV/AIDS (7%).
• People between 15 and 44 - relatively close to the age range I would call “adult” - are at much higher risk than children from tuberculosis (accounting for nearly 10% of deaths in this range), HIV/AIDS (~20%), and maternal mortality (~6% of all deaths in this range; ~15% of female deaths in this range). Cancer (~8%), cardiovascular disease (~10%), and accidents (~15%) are also major causes of death in this age range.
• People between 45 and 59 face similar mortality risks from tuberculosis and accidents; lower (but still high) mortality risks from HIV/AIDS; and higher mortality risks from cancer, cardiovascular disease, and pulmonary obstructive disease. These three conditions are also the predominant causes of death in people over 60.

We previously performed similar analysis here, with a slightly less detailed breakdown of conditions and more focus on the developing-vs.-developed world contrast.