Epistemic Status: armchair speculation from a non-expert.

Short version: I expect things to get pretty bad in the Sahel region over the next year in particular. The area is an obvious target for global health interventions even in good times, and impact is presumably higher in bad times. A simple baseline intervention: fill a backpack with antibiotics, fly to the region, and travel around distributing the antibiotics.

What’s The “Sahel” Region?

The Sahel is a semi-arid region along the southern edge of the Sahara desert. Think roughly Mali, Niger, Chad and Sudan.

Bad How?

Based on statistics on the Sahel, it’s one of the few remaining regions on Earth where the population is near Malthusian equilibrium. Fertility is high, contraception is rare; about half the population is under age 16. Infant mortality is around 6-8%, and ~a quarter of children are underweight. (Source: CIA World Factbook entries on Mali, Niger, Chad and Sudan.)

Being near Malthusian equilibrium means that, when there’s an economic downturn, a substantial chunk of the population dies.

Die How?

Traditional wisdom says: war, famine, disease. In this case, I’d expect famine to be the main instigator. Empty bellies then induce both violence and weak immune systems. On priors, I’d expect infectious disease to be the main proximate killer.

The Next Year In Particular?

The global economy has been looking rough, between the war in Ukraine shocking oil and food markets, and continuing post-Covid stagflation. Based on pulling a number out of my ass without looking at any statistics, I’d guess deaths from violence, starvation, and disease in the Sahel region will each be up an order of magnitude this year/next year compared to a good year (e.g. the first-quartile best year in the past decade).

That said, the intervention we’ll talk about is probably decently impactful even in a good year.

So What’s To Be Done?

Just off the top of my head, one obvious baseline plan is:

  • Fill a hiking backpack with antibiotics (buy them somewhere cheap!)
  • Fly to N'Djamena or take a ferry to Timbuktu
  • Obtain a motorbike or boat
  • Travel around giving away antibiotics until you run out
  • Repeat

Note that you could, of course, substitute something else for "antibiotics" - maybe vitamins or antifungals or water purification tablets or iron supplements or some mix of those is higher marginal value.

There are some possibly-nonobvious considerations here. First, we can safely assume that governments in the area are thoroughly corrupt at every level, and presumably the same goes for non-government bureaucracies; trying to route through a local bureaucratic machine is a recipe for failure. Thus, the importance of being physically present and physically distributing things oneself. On the other hand, physical safety is an issue, even more so if local food insecurity induces local violence or civil war. (That said, lots of Westerners these days act like they’ll be immediately assaulted the moment they step into a “bad neighborhood” at night. Remember, folks, the vast majority of the locals are friendly the vast majority of the time, especially if you’re going around obviously helping people. You don’t need to be completely terrified of foreign territory. But, like, don’t be completely naive about it either.)

Also, it is important to explain what antibiotics are for and how to use them, and there will probably be language barriers. Literacy in these regions tends to be below 50%, and presumably the rural regions which most need the antibiotics also have the lowest literacy rates.

How Much Impact?

I’m not going to go all the way to estimating QALYs/$ here, but… according to this source, the antibiotic imports of the entire country of Mali in 2020 amounted to $53k. That’s for a country of 18 million people and change. Now, I certainly wouldn’t take that statistic at face value, but I think we can safely conclude that Mali does not have anywhere near the amount of antibiotics the population could use.

Even if each course of antibiotics bought and distributed has only a 0.1% chance of saving someone’s life, if you can get the antibiotics for $1/course someplace cheap (and not spend too much money travelling around distributing them) that’s still $1000/life - pretty respectable impact/$. And that’s a pretty conservative estimate -  I would guess that you could get closer to $0.10/course with a little shopping around and a large bulk buy, and I’d guess that each course saves more like 1/100 life rather than 1/1000 (assuming the antibiotics aren't completely wasted - starving people are known for eating things, which is another potential issue). So impact could easily be higher by one or two orders of magnitude.

How Would You Measure Impact?

The simple first-pass answer is to not measure impact, and just operate entirely on priors. Not great, but hey, the lack of good feedback loops is often a major cause of low-hanging fruit.

The next better answer would be to do another round a year later, revisit some of the same places, and ask people what they did with the antibiotics.

Why Expect This Fruit To Be Unpicked?

Obvious factors here:

  • Risk tolerance, specifically risk of physical violence. I’d guess that most individuals would be fine with a moderate amount of physical risk, but for a large organization it’s a PR disaster when one person out of a hundred or a thousand gets kidnapped once. So, I expect orgs to systematically underinvest in this sort of thing (and/or partner with local governments, which have massive overhead at best and usually massive corruption too).
  • Just directly physically doing things seems to be high-impact in general

Why Are You Writing This?

That last point is actually the main reason I’m writing this post: just directly physically doing things seems to be high-impact in general. It is not a coincidence that most of GiveWell’s top charities directly distribute physical things. Just ignore the many opportunities to “partner with <bureaucracy>”, or network with Important People, or whatever other social games are fashionable, and go directly solve (some part of) the object-level problem.

It seems to me like a majority of newcomers to EA get lost in social reality, and lose track of physical reality. That manifests, for instance, as a tendency to generate ideas like “try to Gain Influence and use it for X”, and to not generate ideas like “fill a backpack with antibiotics and travel around the Sahel distributing them”. My current best guess is that marginal focus on social influence, rather than thinking about physical reality, is orders of magnitude less valuable across cause areas; this post is meant to be an example. (And yes, this also applies to AI.)

… but hey, maybe I’m wrong about that. I’m a non-expert engaging in armchair reasoning here, after all. So if you think the “backpack full of antibiotics” plan is a dumb idea, feel free to tell me I’m wrong.


New Comment
31 comments, sorted by Click to highlight new comments since: Today at 1:13 PM

If this region is at malthusian equilibrium why not travel with a backpack full of contraceptives instead?

Or more generally, if the region is in Malthusian equilibrium, why expect any one intervention to save anyone? In a perfect equilibrium, you're just shifting deaths around from one cause (infections) to another.

In a Malthusian equilibrium, I expect to see deaths mainly in hard times. Get someone through a hard year, and they're likely to be alright for another few years after. And decade by decade, more areas break out of Malthusian equilibrium, so buying enough time is a plausible strategy.

Though admittedly that's all a qualitative argument, and I don't know what the numbers look like.

Object level comment: Antibiotic resistance is bad, this is likely to make it worse, probably without saving lives. You probably shouldn't self-medicate with antibiotics, you definitely shouldn't give them to others without knowing more about medical diagnosis. If you wanted to help, there are orgs that work on this, which are generally reasonably good at what they do - I don't think supporting MSF is as valuable as giving money to Givewell, but it's far better than independent direct work. And if you want to do direct work, MSF takes volunteers, and will train you. If you want to work in the Sahel, learn Arabic and contact their people there.

Meta comment: I think this style of reasoning is bad. Speculation without asking people who know more seems like a waste of time. Speculation as a call-to-action is even worse. Red-teaming your own ideas a bit would be helpful, and it seems like that is fully missing here. Experts are pretty easily found within the EA/LW community - and asking for them to review your post, or even better, the logic and idea before you write the post, is perfectly reasonable.

Object level comment: Antibiotic resistance is bad, this is likely to make it worse, probably without saving lives. You probably shouldn't self-medicate with antibiotics, you definitely shouldn't give them to others without knowing more about medical diagnosis.

I've certainly heard arguments along those lines before. They seem like obvious bullshit. Evidence: in most of the world, antibiotics are readily available over-the-counter, and yet I don't hear about most of the world's human-infecting bacteria becoming antibiotic resistant. Most of the world continues to use antibiotics, as a self-medication, and year after year they keep mostly working.

It seems to me like a very strong analogue to Oregon and New Jersey's laws about pumping your own gas. Both of those states don't allow it, which the rest of us know is completely stupid, but there's still somehow a debate about it because lots of people make up reasons why it would be very dangerous to allow people to pump their own gas.

"yet I don't hear about most of the world's human-infecting bacteria becoming antibiotic resistant"

You're not paying attention, or looking for the evidence - it's not covered by the news, because of course they ignore gradual threats that pose minor inconveniences. But there is a lot of academic work on this. Even on short time scales, changing resistance is observable. The problem isn't small. According to that last link, at this point, every year more people are killed due to antibiotic resistance than are killed by Malaria. And unlike Malaria, the trends are getting worse. I'm sure you can argue about the exact numbers, but unlike pumping gas, it's a classic collective action problem were individual acts harm themselves too little to be noticed, and erodes a global commons that we can't easily replenish or replace.

The "most" was doing key work in that sentence you quoted.

I totally buy that antiobiotic resistance is a large and growing problem. The part which seems like obvious bullshit is the claim that the cost outweighs the benefit, or is even remotely on the same order of magnitude, especially when we're talking about an area like sub-Saharan Africa. Do any of those studies have a cost-benefit analysis?

(Also, side note: antibiotic resistance is totally in the news regularly. Here's one from yesterday.)

Cost-benefit analysis is a very weak tool here, since costs are very hard to assess, long term, and uncertain, and in every individual case, it's worth it because it's a collective action problem and others are doing it wrong already.

There are estimates of the cost of antibiotic resistance, for example, almost $5b/year in the US alone. So from a collective action standpoint, if you assume that all agents are going to follow a policy, you at the very least only want to prescribe specific antibiotics when they are clinically useful - and even if you're not running tests, etc. you need to know a really significant amount to know which antibiotics to use for which set of symptoms, and you should only prescribe them if there's a pretty significant chance of full compliance. Hence the DOTS regime for TB - WHO guidelines require observing the patient taking each dose, not just prescribing it.

As a general rule of thumb it's correct. Antibiotic resistance is a growing problem - it's your basic red queen situation. We're currently far ahead, but bacteria are catching up. This film is a pretty way of showing it happen (sort of) live. MRSA is causing more and more problems, especially as it tends to pop up in places where vulnerable people are (since that's where it gets its resistance in the first place). It helps that there are different families of antibiotics, so if a strain is resistant to one of them, you can switch to a different kind. Though this paper suggests that each class works for ~50 years (I'm not a microbiologist and don't know how true that is, though it fits with my priors). So it's a valid issue, especially in the context of livestock, where it's common to mix in antibiotics with the food (as each day ill is a day not growing).

That being said, this is mainly a problem in developed countries with ample access to antibiotics, which is a totally different case from the one you're making. You could even say that it bolsters your idea, as why not send the weakened versions somewhere where they still have a chance of working?

I'm not sure that availability of antibiotics OTC in most of the world is good evidence that they don't pose a problem. The main danger of gasoline is that it could catch on fire. It's easy to see cause and effect if that were happening. If OTC antibiotic use was causing problems, that would require epidemiologists to figure out, and a political solution to fix. I'm not confident that most of the world is well-positioned to catch and fix any issues that may be arising from OTC antibiotic sales.

There's reason to ask about the relative contributions to drug resistance of self-medication at home, hospital antibiotic use, and antibiotic use on factory farms. We hear about drug resistance mostly in hospitals, but I'd want to check whether this is reporting bias before believing that hospitals are the main drivers of antibiotic resistance.

I'd also question the degree to which antibiotic resistant bacteria can be "trafficked" from place to place. If antibiotic resistant bacteria develop inside a hospital, does that make the bacteria that people pick up outside of the hospital more likely to be drug resistant?

I wouldn't be too surprised if antibiotic resistance is a local, hospital-centric phenomenon, with home antibiotic use posing a negligible threat of causing users to contract antibiotic-resistant bacteria. But I do not have data to back that up. It would be an interesting research project.

I'd be more concerned that some people would harm their gut bacteria or kidneys by overconsuming antibiotics. But I'm not confident about this. It's just the most obvious direct consequence of expanded and unregulated access to antibiotics, and it's where I'd start if I were to research possible negative consequences of parachuting into Chad with a backpack full of amoxycillin.

We can stop speculating about these questions - the answers exist and are relatively easy to check.

https://academic.oup.com/cid/article/27/Supplement_1/S12/459194 (Horizontal transfer is where the resistance is "trafficked" between different pathogens.)

https://academic.oup.com/cid/article-abstract/33/3/364/277722 (Geographic spread is very common, but you need better tracking to see exactly where and what the routes are.)

This isn't my field, but there are tons of people who could give concrete and specific answers to all of these questions, and so it seems silly to continue speculation.

Thanks for finding these!

To be clear, I don't think the claim that self-medicated antibiotic use causes more antibiotic resistance is obvious bullshit. Maybe the effect size is close to zero outside of hospitals, maybe it's not, but the claim isn't obvious bullshit either way.

The "obvious bullshit" part is the (implicit) claim that the cost outweighs the benefit, or is even remotely on the same order of magnitude, especially when we're talking about an area where the alternative is usually "don't use antibiotics at all".

Yeah, it seems in the neighborhood of worrying that giving parachutes to people being pushed out of planes might lead to manufacture of low-quality parachutes to meet the demand, resulting in a net increase in deaths from plummeting to the ground.

Strongly upvoted for identifying and working around the issue of "social reality". While I doubt that this "direct action" approach is cost-effective on average, since there is no leverage and no scale advantage, it is nearly guaranteed to achieve the stated goal of preventing a number of deaths from infectious diseases. They might still die from starvation and violence, but that is a separate issue you can't hope to solve on your own.

The scale advantage is that you become the world expert on on-the-ground antibiotics distribution in the Sahel region. I can easily imagine that you can onboard locals to do better distribution, build a pipeline for sourcing and shipping in antibiotics, get grants to fund operations at a larger scale, identify even more pressing problems, etc.

That is definitely a possibility, but not something OP seems to have in mind.

Don't you think you should first talk to the organizations with that experience, instead of trying to learn from your own experience, without even looking at what is already happening?

My first comment on this post recommended starting by contacting the WHO regional office for the country in question.

I'd have gone with doctors without borders, which I linked to, which does far more on the ground work and would know about the ability to help better, but I think we agree here.

Also, it is important to explain what antibiotics are for and how to use them, and there will probably be language barriers. Literacy in these regions tends to be below 50%, and presumably the rural regions which most need the antibiotics also have the lowest literacy rates...

Even if each course of antibiotics bought and distributed has only a 0.1% chance of saving someone’s life, if you can get the antibiotics for $1/course someplace cheap (and not spend too much money travelling around distributing them) that’s still $1000/life - pretty respectable impact/$. And that’s a pretty conservative estimate...

The simple first-pass answer is to not measure impact, and just operate entirely on priors.

A very important question to ask here is: Is giving random people random antibiotics even positive EV? Just off the top of my head, here are some things that might happen:

  • People might steal them from one another, including at the risk of violence.
  • People might not trust the foreigner giving out random packs of stuff and throw them away.
  • People definitely won't understand the foreigner giving out random packs of stuff, due to the language barrier.
  • People might take the antibiotics at the first opportunity, irrespective of whether they're ill. That would likely result in negative EV (depending on which fraction of the population has antibiotics-treatable infections), e.g. antibiotics side effects. E.g. I imagine digestion problems are not good for people with malnutrition. If you save 1/1000 people and give the rest a week of antibiotics side effects, that can very easily sum to negative EV.
  • People definitely wouldn't consistently take antibiotics as they're prescribed by doctors, i.e. "take X tablets per day for 1-2 weeks". How does that affect the EV? Suppose you're infected and take antibiotics intermittently, or only for a few days. Is that even positive EV? What if people share the antibiotics and everyone only gets one tablet? (EDIT: And how bad is it to overdose on antibiotics?)
  • Do random antibiotics even work against random infections? <-> How do doctors decide what antibiotics to prescribe, and when?

And so on.

(And separately, as mentioned elsewhere: In a Malthusian equilibrium, why would you expect to save anyone by healing their infection?)

I'm skeptical. I understand that driving around dispensing antibiotics yourself cuts out various (possibly corrupt) middlemen. But wouldn't that just reduce the benefit somewhat, rather than make it useless? If you instead of doing it yourself just dropped off a backpack full of antibiotics with the first reasonable-seeming person you met, and then they went and started a business selling it to various people who sold it to various people who sold it to users... wouldn't all the antibiotics still make it out to people who needed them? Arguably with an allocation more skewed to people who actually need it as opposed to people who happened to be in your random path. The downside is that said people would need to pay instead of getting it for free, but still. Say they get a bad bargain and pay 2/3rds of the total surplus, gaining only 1/3rd for themselves. Well, then doesn't that mean that your charitable act of dropping off a backpack full of antibiotics was 1/3rd as good for the recipients as driving around finding them yourself and giving them antibiotics for free? A haircut-to-impact like that is actually small potatoes I'd imagine, easily outweighed by the other benefits.

Yup, I think that's one of the right questions to ask. I expect that there's a real market inefficiency somewhere in there, but it's definitely the sort of thing I'd recommend actively investigating while on the ground. If it turns out there's not an inefficiency in distribution (or end-use), then yeah, just drop the load on someone. Ideally, one figures out exactly where the inefficiency is and can then pivot to a much higher-impact strategy.

It also could be that the market is efficient & the people of the Sahel have better things to spend their money on than antibiotics; i.e. you'd be better off giving them money than antibiotics. Or whatever. I'd be interested to see more argument that they need that stuff in particular, rather than just money.

Will you fly to the Sahel with a backpack full of antibiotics?

I imagine you suggesting this, a bunch of people nodding along in agreement, then no one doing it because of personal safety and because it's weird.

I do not currently have plans to do it myself. It does seem like the kind of vacation I'd enjoy a lot, but only if I were looking to take several months off of alignment work.

One way to start might be to reach out to the WHO's Chad office. Note that while Chad has many problems, it's also got a public health infrastructure. They vaccinated 3 million people against polio in 2020.

Or just visit to get information. Don't choose antibiotics vs vitamins based on estimated value delivered, but diversify to learn about them all, to learn what it takes to deliver them. But the most valuable information will probably be unrelated to what you bring.

This is sticking your finger in the dike when the dike is already overflowing. Besides which, IANAD, but handing out antibiotics like sweeties sounds like a recipe for antibiotic resistance. Are you a doctor, or has a doctor commented on this plan?

If someone were concerned about personal risk,  they could fly into the major cities and then distribute the antibiotics with pictograms via drones and parachutes. This might also reach more people, assuming the drones could operate autonomously via GPS or something?

Walking around a starvation-stricken area with the equivalent of months to years' worth of income in a backpack isn't a good idea. The mostly likely outcome is that you get robbed, possibly held for ransom, and the antibiotics get sold on the Nigerian market since the locals aren't rich enough to afford them.

New to LessWrong?