5. How deworming became the darling of the aid community

I review the results of a major deworming effort, 20 years later

^ Child receiving deworming pill / Courtesy of EvidenceAction

In the 1990s, two economists saw an opportunity to study a really simple public health intervention: deworming. Intestinal worms were (and continue to be) a widespread problem: 1 — 1.5 billion people are infected with worms around the world, and most of these people are children: one study I read said that almost all of the children of the “bottom billion” (the roughly billion people who live in or near extreme poverty) have one or more intestinal parasites (from 2010, so numbers are lower now).  

But worms are particularly bad for kids: children are more likely to be infected (they  worms live in the small intestine and compete with the child for nutrients. As a result, children with worms are often more malnourished, underweight, or anemic, and more likely to be absent from school. Incredibly cheap and effective treatment exists, with deworming pills costing less than a dollar per person per year.  

At the time, hardly anyone who wasn’t working in global health knew about deworming. Now, deworming has become a darling of the international aid world, the health intervention par excellence. In this issue of Giving to Strangers, I want to tell the story of deworming: what set off a massive international effort to support deworming? And, importantly, why do we believe it works?

Some wormy background

Worms can enter the body through different pathways. One way is through the feet: kids who run around barefoot might step on larvae that’s been deposited in the soil. The eggs then burrow into the feet and travel trough the bloodstream. The worms might up in the soil through defecation — children who are infected and defecate outside instead of a latrine can spread worms to other children via soil. Kids can also be infected if they swim in nearby Lake Victoria, where the worms are carried by currents around the lake, even infecting kids who live some distance away. Crucially, for the study, these patterns of spread also means that the fewer people have them, the less likely it is that others will get them. 

How it all got started

Michael Kremer and Edward Miguel (Kremer would go on to share the Nobel Prize in Economics in 2019) were some of the first economists to do fieldwork in Western Kenya. In 1998, Michael Kremer and Edward Miguel learned that an NGO there was going to deworm over 32,000 students. In a stroke of brilliant opportunism, they convinced the NGO to randomly pick which schools to give deworming treatments to first. The result was that all students were dewormed, just some started 2 — 3 years later than others (or, put equivalently, students in “treatment” received extra years of deworming). This small difference in the design enabled one of the most impactful, long-lasting, and sometimes controversial research projects in development economics.    

Miguel and Kremer found that deworming had potentially life-changing effects: Students who received extra years of deworming attended substantially more school (absenteeism fell by one quarter over the first two years of the program). But deworming didn’t just benefit the students who received it — there were major benefits (what economists call “spillover“ effects) on students who hadn’t received deworming: they started attending school more, too! Treating some students decreased the overall transmission of worms in the community, benefitting everybody. 

The importance of long term surveys 

Since that original study began 20 years ago, the authors have been continuously surveying the dewormed children (now adults), and releasing their major findings roughly every 5 years, to show how deworming impacts people’s lives as they grow up. This has often meant great feats of data collection — a large team of over 60 people is tasked with following up with the individuals, a job that sometimes involves tracking people down when they move from their remote village to Nairobi, or even neighboring Uganda. The team of data collectors has even surveyed individuals by phone in Europe and Dubai. These logistical efforts pay off — 20 years later, the research team is able to reach a staggering 84% of the original sample!

As an aside, for the social science curious: Perhaps because of how logistically difficult it is, this kind of long-term data collection over the course of a subject’s life is very unusual in development economics. The majority of studies I read collect data 6 or 12 months, or several years after the intervention (I once saw a 7 year follow-up). Following up twenty years later is unheard of, yet often crucial. Why? Say you wanted to see if deworming boosts income later in life, so you measure a child’s income 10 years they were dewormed, at age 18 (children are dewormed in elementary school, you can do the math). If deworming affects income by increasing school attendance, by age 18, it’s not unreasonable that more of these students might pursue additional years of education. But this means they would be temporarily earning less than someone who quit school to work full-time. It would look like deworming decreases income, when of course, additional education (+ its downstream benefits) is exactly what we want! Though this is a bit of a toy example, it illustrates how the decisions that researchers (and the people who fund them) make about data collection might matter for the kinds of conclusions they draw.

20 years later: How does receiving deworming as a child impact adult life?  

The paper presenting the impact of deworming twenty years later was released last month by Edward Miguel, Michael Kremer, Michael Walker, Sarah Baird, and Joan Hamory. (For Miguel’s fascinating, really accessible presentation explaining the paper, click here, he’s a phenomenal public speaker) 

The authors find that twenty years later, people who received more deworming as children earn 13% more per hour, are more likely to work outside of agriculture (and therefore in potentially higher paying office or business jobs), and more likely to live in cities. This is an incredible return on investment for the Kenyan government: given how cheap deworming is, the authors calculate the rate of return to be 37%.

The takeaway? Small things that governments do to improve children’s health can pay off even into adulthood. 

Why are there few effects of deworming on income for women? 

When the paper came out, I talked to Michelle Layvant, a researcher on the project and my coworker at the Center for Effective Global Action. She kindly explained one of the puzzling findings of this latest paper, that most of the benefits accrue to men and older people in the sample:

“We see a huge gap in earnings and consumptions effects for men versus women. [The effects are] big for men, and insignificant or close to zero for women. One reasons might be that there’s not as much opportunity in the labor market for women. But another reason could be that the effects for women are coming out in different ways.” 

Here’s my interpretation: Say deworming means you’re able to attend more school as a kid. More education translates into better and higher paying jobs as an adult, but only if your society is structured in such a way that you can access those jobs. We might not be seeing job benefits for women because of the barriers they face in the labor market in Kenya. Instead, the effects of the intervention on adult women might show up as different gender or political views, religious beliefs, or choices in family structure. Michelle says there’s some evidence of this: women in the treatment group are having fewer kids, but again, not all of them have potentially finished bearing children — the team is going to keep collecting data to really see how deworming has impacted women in the long run.  

Who should pay for deworming?

The final part of this story is a fascinating lesson about when the government should pay for preventative healthcare. As part of the original deworming program, they asked some parents if they were willing to pay for part of the deworming costs for their children. Most parents said no. 

Michelle on the takeaways: 

“Even if an intervention is super important, it doesn’t mean that the individual can value it correctly. Sometimes it’s the government’s responsibility to do it. [Deworming] isn’t a self sustaining program, it’s something that needs to be supported by a government or a public institution.”

The authors of the original paper went even further: “Moreover, internalizing these externalities would likely require not only fully subsidizing deworming, but actually paying people to receive treatment.”

Paying people to accept something that’s really good for them. It may seem counterintuitive, but it’s not uncommon. And it shouldn’t prevent us from doing it. 

Which strangers might you give to? 

That 2004 deworming study was important fodder for a global effort to fund deworming programs. 

The publication of the study fortuitously coincided with the growth of the effective altruist movement in the mid-2000s. People who cared about not just donating their money, but donating effectively, started reading economics literature and realizing deworming was great value for money.

Organizations like GiveWell, for example, turned into a great boon for the charities that implemented deworming interventions. GiveWell is a charity evaluator: it finds the organizations that save or improve the most lives per dollar donated, and then wires money from donors directly to those truly outstanding charities. Since 2011, NGOs that do deworming have featured heavily on GiveWell’s top 10 charity list, and between 2011 — 2018, GiveWell directed over $112 million to deworming interventions.

While this may seem like a huge number, these charities remain on GiveWell’s list because they’re tractable and have extra capacity — there’s still need to deworm, and these charities have capacity to actually carry it out. That means you can be confident that your dollar will improve children’s lives in the ways I described above. Donate here.  

If you found this post valuable or at least mildly apocalyptically entertaining, it would make my day if you shared Giving to Strangers with friends so they can subscribe:

Share Giving to Strangers