Ezra Klein: Your Foundation is known for taking a particularly data-driven approach to its work. So how do you know what’s actually working when you’re in failed states with very little data-collection capacity?
Bill Gates: Of all the statistics in health, death is the easiest, because you can go out and ask people, “Hey, have you had any children who died, did your siblings have any children who died?” People don’t forget that. If you say to them, “Did your kids get vaccines or not,” they might have done it and not remember, or they might think, “Oh, this person wants me to say yes, maybe I look bad if I don’t say yes.” Death is something we really understand extremely well.
But you can save a lot of lives. One thing about the childhood death rate is you really can split it into the first 30 days of life versus 30 days to 5 years. Thirty days to 5 years is all vaccine preventable stuff -- it’s diarrhea, respiratory and malaria. The first 30 days, the primary healthcare system really has to engage with the mother pre-birth, and then get the mother to do things like keeping the baby warm, making sure to avoid doing things that break the baby’s skin, breast-feeding, and that’s been harder. We’ve had sites in India where we can cut those deaths down by over 50 percent just by training the mother. But the worker has to engage with the patient, hopefully speak the same language or be of the same caste so that they’re willing to trust the advice that they’re getting.
EK: What’s been the biggest surprise? What has the data shown works, or doesn’t work, that you simply didn’t expect?
BG: I was completely surprised that nobody was funding some of these vaccines. When I first looked at this I thought, well, all the good stuff will have been done. It was mind-blowing me to find things like Rotavirus vaccine were going unfunded. One hundred percent of rich kids were getting it and no poor kids were. So over a quarter million kids a year were dying of Rotavirus-caused diarrhea. You could save those lives for $800 per life. That's like $20 or $30 per year of life. It's just ridiculous that an intervention like that isn't funded.
And I’m really surprised at the variance. Some very poor countries run great vaccination systems and some richer ones run terrible programs. The north of Nigeria has about 30 percent vaccination coverage, and they’re above average in terms of wealth within Africa. You compare that to, say, Somalia, which has absolutely no government at all, and they get about 60 percent vaccine coverage of children. So you have a place literally with no government getting a better vaccine coverage than a place that’s above average wealth.
BG: Well, in Somalia they’ve given up using the government. The money goes through the NGOs. Whereas in Nigeria they’ve designed a system where the federal government buys the vaccines, the state government provides the electricity, and the one level down below that provides the salaries. It’s just a bad design. You know, the north of India has very poor vaccination rates, so we picked a state up there with 80 million people and we drove it from 30 percent to 80 percent. But they had a really good chief health minister and the federal government was providing lots of money and lots of good technocrats, so the skills were there, as long as you employed them in the right kind of system.
EK: This gets into an interesting question about public health, which is that when we think about health-care challenges, we think primarily about technological challenges. We think about cures for cancer and vaccines for AIDs. But in public health, much of the challenge is logistical and organizational -- how you deliver, how you organize, who you actually partner with. And that seems much harder to replicate. If you can invent the pill, then you can probably keep reproducing the pill, but even if you get a good system in the north of India with their good health minister, it isn’t necessarily the case that you can move that over to the south of Nigeria.
BG: It can be replicated, though. Ethiopia is a good example of a country that decided to get serious, train 35,000 health workers and actually put them in the right places. So they did the map, looked at it, got the donor money, it’s a work in progress. It’s doing quite well but there are still a few of the supply chain things that need to be fixed. So, it can be replicated. We do report cards for each country, saying OK, did you have a plan, do you have the money, did you do the personnel right, did you do the supply chain right?
EK: What’s the difference between trying to work in high-income, middle-income, and low-income countries?
BG: The low-income, middle-income and high-income health systems have extremely different problems. You know, in low-income countries, getting to a health post is hard. It’s very expensive. Whereas in rich countries, yes, you can get to your doctor. In low-income countries, the main problems you have is infectious diseases. We’re dealing with countries that in the worst case where kids have death rates of 20 percent and that’s all infectious disease. And nothing else. In the U.S., in terms of kids under 5, other than premature birth, you really don’t have big problems. Kids just don’t die of infectious disease.
Then as you get into the adult phase, in the U.S., what do people die of? From age 5 till age 50, you’ve got suicide, you’ve got traffic accidents. There’s very little cancer and heart disease before age 50. Then once you get past your 50s, the poor countries basically say, “Hey, I hope you don’t get cancer, because if you do get cancer, we just don’t have enough money per person, we’re just not gonna buy chemotherapy drugs. We’re just not going to get engaged in that.”
EK: How do you make these decisions about what is and isn’t worth paying for?
BG: The way that this is talked about is, what’s a year of life worth? They call it a disability-adjusted life year (DALY). When you're running a poor country health-care system, you can’t treat a year of life as being worth more than, say, $200, $300 or else you’ll bankrupt your health system immediately. So, with very few exceptions, you do nothing for cancer. If you get cancer, you're going to die. And so none of the stuff that’s going on in the U.S. about $300,000 a year chemotherapy drugs is relevant.
Even simple things don’t pass the test. We’re on the verge of saying that Africa should do blood pressure medicine because it’s become generic and so cheap and that’s such a common issue in terms of heart attack death, the so-called polypill is so cheap that it’s one of the few non-infectious disease things that meets the dollars per DALY threshold to actually go into a poor healthcare system and say this is worth it given the extremely finite not only financial resources, but personnel resources, that you have.
But here’s the good news for these countries. If you spend the less than 2 percent of what the rich countries spend, but you spend it on vaccinations and antibiotics, you get over half of all that healthcare does to extend life. So you spend 2 percent and you get 50 percent. If you spend another 80 percent you're at over 90 percent.
EK: Your top priority, I’m told, for the next year is the literal eradication of polio. What’s between here and there?
BG: Whenever you can eradicate one of these infectious disease, you get these exponential benefits. Polio’s the extreme example where we’re near the magic number of zero, so the $2 billion that the year spends protecting kids against getting polio, the day you know you're at zero -- you have to really know you’re there -- then you save the $2 billion. And, you know, that happened for smallpox. Nobody spends any money on smallpox unless they worry about a bio-terrorist recreating it. It’s financially the best thing that ever happened because we're saving all that money forever after.
So in 1988, the World Health Organization, through the World Health Assembly, declares they’re going to eradicate polio. It's already been eradicated in North America and South America and most of Europe. Polio is paralyzing 360,000 kids a year when they start. They get it down below 10,000 by the year 2000. Then from 2000-2010 it stays flat. And they lose credibility because they're always saying “Oh, just give us two more years,” and they're just doing the same thing and it's not working. And so in 2010, the polio community got together and said, “Look, are we going to succeed or not?” And so there were a lot of improvements made, those led to finally getting done in India in 2011. And India was expected to be the hardest and the last.
EK: Just a point of clarification, the functional mechanism here is a high enough level of vaccine coverage, right? When we say eradicate, we mean got it to a high enough level of vaccine coverage that the disease died out?
BG: Yeah, eradicate means there's no poliovirus around. The method of doing that is you get to extremely high levels of vaccination -- that is, over 90 percent of the kids have the drops three times, and that protects them and the disease dies out. The number is actually well below 90 percent if you're in a community that either has good sanitation or where the kids don't move around much.
EK: So what did we learn that made eradication possible in India?
BG: The two things that were done super well were social mobilization and mapping where the houses were. When somebody would refuse to take the vaccine, they would mark it down and they would have either a political leader or religious leader come in and convince them. Dealing with refusals is a huge part of this. If your team goes in, maybe they don't speak the dialect, they're not the same caste, the family has heard a rumor that the vaccine is bad, there's many reasons you get refusals, and so you need follow-up for refusals. Usually you'll get 10 to 20 percent refusals. But if there's been a rumor, you get much higher refusals.
EK: A rumor that, say, the vaccine is bad, or it makes you sick?
BG: Yeah or that the U.S. government uses vaccination campaigns to sterilize Muslim women. Vaccination always has problems with rumors. The U.S. doesn't achieve nearly as high a vaccination rate as many countries. Vietnam is 99 percent vaccination, the U.S. is about 95 percent. Because people just hear “Oh, what about autism or something.” But it's particularly bad in poor countries.
EK: The logistics of the operation seem basically impossible. How do you ensure you hit every tiny village in a mountainous, rural, poor country?
BG: We began using satellite maps and we're finding particularly in Nigeria we were missing a lot of settlements, a lot of nomadic people. The thing we were missing the most was a village would be on a border, and one government would say, “Oh, that's on their side,” and the other guy would say, “No, that's on their side.” So your chance of getting polio was super elevated if you happened to live on the border between these local government administrative boundaries.
Then in terms of the teams doing their job, we now put a phone with a GPS sensor in it, every three minutes it says where this team is. It's in the box with the vaccine so when they come in at the end of the day we plug that in and see if they really went where they were supposed to go.
Our biggest problems now are violence, which causes campaigns to be canceled, or people just not to be willing to go into various neighborhoods, and refusals having to do with bad rumors about the vaccine campaign. And these are both serious issues in both Pakistan and Nigeria. Afghanistan is just part of the Pakistan thing, and it's not the big deal. The number of cases there is pretty small and it's just in the areas where there's fighting.
EK: I almost feel bad asking it after this particular discussion, but what has this work made you think about our health-care system's problems, recognizing everything you said about how incredibly, incredibly different they are from, truly poor countries?
BG: It's an important topic and I do care about it. My deep interest in this came somewhat because it's fascinating but also because our big cause in the U.S. is education, and if you look at state budgets, they are moving money from education to health. They have to because the health costs are just exploding. So very quickly say to yourself, gosh, if there's going to be any money left for university education and adequate money for K-12, even to stay flat, you have to figure out health-care costs.
Unfortunately, in rich-world health, innovation is both your friend and your enemy. Innovation is inventing organ replacement, joint replacement. We're inventing ways of doing new things that cost $300,000 and take people in their 70s and, on average, give them an extra, say, two or three years of life. And then you have to say, given finite resources, should we fire two or three teachers to do this operation? And with chemotherapies, we’ve got things where we’ll spend our dollars on treatments where you’re valuing a life here at over $10 to $20 million. Really big, big numbers, which if you were infinitely rich, of course that would be fine.
So most innovations, unfortunately, actually increase the net costs of the healthcare system. There's a few, particularly having to do with chronic diseases, that are an exception. If you could cure Alzheimer's, if you could avoid diabetes -- those are gigantic in terms of saving money. But the incentive regime doesn’t favor them.
EK: You’ve talked a lot so far about this question of DALYs. We’re very uncomfortable putting a value on human life. The way I see our health system is we’ve chosen to pay a huge premium in order to avoid these questions. A prerequisite for the kind of cost-cutting innovations you’re talking about it is being willing to make judgments about what a human life is worth, or even what a few months of a human life are worth. Because if you can't decide that, then of course you just pay for everything. But if you start trying to make those choices, or even get people to think about those choices, people cry “death panels!”
BG. Yes, someone in the society has to deal with the reality that there are finite resources and we're making trade-offs, and be explicit about that. When the car companies were found to have a memo that actually said, “This safety feature costs X and saved Y lives,” the very existence of that memo was considered damning. It was “Oh, you think human life is only a bank account.” Or when you made it reimbursable for a doctor to ask, “Do you want heroic care at the end-of-life,” that was a death panel. No, it wasn't a death panel! It was asking somebody to make a decision.