Scalable Strategies to Address NCDs in Developing Countries

Experts discuss scalable and practical strategies developing countries can use to address the global health challenges of noncommunicable diseases.

This session is part two of the two session meeting, Noncommunicable Diseases and the New Global Health.

THOMAS BOLLYKY: Well, terrific. Good afternoon, and welcome back. For those of you that have made the trek upstairs to the beautiful David Rockefeller Room, thank you. You're in for a real treat. My name is Tom Bollyky. I'm the senior fellow here at the council for global health, economics and development.

We have -- this will be the second session of the meeting on noncommunicable diseases and the new global health. In the first session, Tom Frieden spoke very powerfully and persuasively about the burden -- increasing burden of noncommunicable diseases in low- and middle-income countries and their economic and even foreign policy implications.

This session is going to drill down on the question of what is there to be done about it. This is not, as I'm sure everyone here knows, an auspicious time for new international or U.S. initiatives on complex global health development challenges. Each new global summit on these topics are dealing with systemic challenges of global health, development or environmental matters seems to produce only a less encouraging result than its predecessor. There seems to be less consensus than ever before on how the costs and responsibilities of major power should be divided on these systemic issues. I think many were disappointed that the U.N.'s high-level meeting on noncommunicable diseases and the processes that have followed at the World Health Organization haven't produced more in terms of tangible action.

Within the U.S., of course, we're in the midst of bitter political gridlock, a still slow economic recovery and a looming potential budget sequestration that seems to limit the potential for the U.S. doing very much on global health issues of any stripe, let alone noncommunicable diseases.

What we have -- with all that good news, what we have at this panel, though, is a series of speakers who can talk persuasively about what can be done, scalable, effective strategies for making progress on noncommunicable diseases in developing countries in these austere and divided times.

And we have three tremendous speakers to discuss this issue. To my right we have Matt Myers. You have their biographies in your pamphlet, so I'm not going to go through them in depth, particularly because we have three speakers, and I want to make sure to give them all enough time to speak. But Matt is the president for the Campaign for Tobacco-Free Kids and has done tremendous work both domestically on this issue as well as abroad and is a real leader in his field, and we're honored to have him.

Bernhard Weigl is the CEO for Diagnostics for Global Health at PATH and has some exciting programs to talk about in terms of the extension of approaches pioneered in addressing infectious and epidemic -- infectious and emerging diseases and how they might be applied in this context.

And last, we have James Habyarimana, who is an associate professor at Georgetown and is going to discuss about the use of social cues to address road safety, which I can assure you is a topic I cannot imagine has been discussed at the council before. And we're thrilled to have someone here to talk about the international challenges around road safety.

So let me start with Matt and say, you know, Matt obviously -- Tom was talking about -- there's been tremendous progress on tobacco control domestically. The family smoking prevention act was passed in 2009. We've seen tobacco use rates really decline in the U.S. over the last several decades. You can have a meal in New York City without someone blowing smoke in your face. Why aren't you happier? What is -- what is going on internationally that diverges from this pattern?

MATTHEW MYERS: Well, first you probably need to know that I'm a bit of an optimist, so "happier" is a funny phrase. You know, I think there's a couple of critical points to be made. First, Tom, despite what Dr. Frieden said this morning, the problem of tobacco use is even greater than most people realize. We've gotten used to it, and yet it is the number one killer, you know, globally, other than hypertension, kills more people than AIDS, tuberculosis and malaria combined. And what people don't realize is the burden from tobacco is rapidly moving from the developed world to low- and middle-income countries that can least afford it.

But then you get -- you do get to the good news part of it, and that is tobacco is the one field where we do know what works. Tom referred to it repeatedly, what he calls the MPOWER package. We do know that if you raise taxes, provide strong graphic warning labels, provide people environments where people don't smoke indoors and you're not exposed to secondhand smoke, conduct powerful mass media campaigns, that each of these solutions have been proven to reduce tobacco use significantly, now across cultures, across political systems, across continents, north and south, east and west, in low- and middle-income countries as well as high-income countries.

The other very important things to know is that these policies and programs, as Tom talked about earlier, work on a population base. We know how to scale them up. They automatically scale up when done right. They're affordable, they're sustainable, and the critical issue with regard to each and every one of them is whether or not we have the political will to do them, because tobacco is the one issue where you actually have an opponent, somebody who makes money off of being the vector of the disease, which adds an additional layer on it.

But let me just give you a couple of examples, because I think it's -- I think it's really important to understand that, as Tom said, we know what works because we're measuring it, and we know what works because we have documentable, credible studies that cut across geographic regions, countries and continents to demonstrate that each of these policies is shown to be effective. And so let me just pick out a couple of quick examples for you.

Ukraine, a country that saw dramatic rises in tobacco use after the fall of the Soviet Union, between 2008 and 2010, increased the tobacco taxes enough to create a 117 percent increase in the price of retail -- retail prices of tobacco products. In 2009 they saw more than a 13 percent decline in tobacco use; 2010 they saw a 15 percent decline in tobacco use. And as a result, in the last year and a half, the Ukrainian parliament has adopted three other of the MPOWER packages, strong graphic warning labels, a hundred percent protection against secondhand smoke and a virtual ban on all tobacco use.

Neighboring, a country you wouldn't think of as a -- as a place where we're likely to have success because of the long history of tobacco use, Turkey, is perhaps a perfect example of a scalable and effective program. In Turkey, what we saw between 1990, '99, among the largest increases in tobacco use around the world. Turkey then ratified the Framework Convention on Tobacco Control, the world's global tobacco treaty, and began to implement it.

We just have the most recent data from the CDC Bloomberg-funded Global Adult Tobacco Survey in Turkey, and what does it show? What it shows is that for -- over the last three and a half years, we've seen a 13.7 percent decline in tobacco use, even more importantly, a 15 percent decline in tobacco use among males. What we've also seen is over a 40 percent decline in tobacco use among physicians. And we often don't think about that, but physicians are often literally the leading edge for change around the globe. In countries where you have high usage of tobacco by physicians, you have very -- you have much less success in reducing tobacco use. Absolutely critical distinction that we're seeing there, and as a result, we're beginning to see a change in disease risk.

Tom talked about Uruguay. Uruguay is a great example, because not only was it among the early adopters of the Framework Convention on Tobacco Control, but Uruguay has now been doing these things for long enough that we have begun to see a real and significant change. So what do we see in Uruguay? We see a 25 percent decline in tobacco use since 2008 alone. What has that resulted in? It's resulted in a 22 percent decline in hospital admissions for serious cardiovascular disease events. Critically important, because in Uruguay, 60 (percent) to 70 percent of all deaths are directly related to noncommunicable diseases. What that means in terms of health care in Uruguay is just dramatic.

Tom also talked about Brazil. And Brazil really is a good example, because Brazil's one of the major tobacco-growing countries in the whole world, another country with a truly dramatic history of tobacco use.

Brazil is another one of the countries that's been doing it just long enough now that we can begin to see the impact. Now, why is that important? Well, because when you look at Brazil, what you see is -- when you look at their cancer rates, their heart disease rates, they're astronomical as a result of a long history there. And so what we have found -- and let me just give you some quick data -- between '89 and 2008, smoking rates in Brazil were cut literally in half, from over 35 percent down -- using the GATS survey -- down to 17.5 percent. What is that already documented change in terms of health? A 30 percent decline in mortality from cardiovascular disease. Think about that. For a country the size of Brazil, a 30 percent decline in mortality from cardiovascular disease and chronic respiratory disease -- absolutely critical.

When you think of the numbers -- Tom said it correctly. Today in China, there are over 300,000,000 male adult smokers. But because the tobacco epidemic in China hasn't yet reached its long-term effect, there is only approximately a million deaths. Now, if you just look at the standard death rate on that, that number will triple without an increase in tobacco use in a very short period of time.

We've seen it in the U.S. Twenty years after the start of male smoking, lung cancer rates skyrocketed. Twenty years after -- and women started smoking in large numbers; the lung cancer rates skyrocketed.

Russia faces one of the biggest crises of any country in the world. When the Soviet Union fell, 5 percent of Russian women smoke. Today, over 22 percent of all Russian women smoke. Many of them are young. What we'll see in Russia in 20 years -- now about 10 years -- is an explosion of lung cancer rates that will be among the greatest in the globe has ever seen, unless we do something about that right now.

So the good news is that with the assistance of WHO, the CDC and Bloomberg funding, we have seen over the last couple of years the most dramatic change in the adoption of these policies than we've ever seen before. Today, as a -- since 2007, there are more than one billion people in the globe who are now protected by strong graphic warning labels. Since 2007, 22 countries have adopted strong smoke-free policies around the world. What we have also seen at the same time is over 40 countries now have graphic warning labels. We have a dramatic increase in mass media campaigns, with the assistance of the World Lung Foundation to assist them.

So we're on the path to progress. The critical issue is whether the countries that haven't -- not yet adopted have the political will to do so. So there is both reason for optimism, but this is one of those issues that is going to be a long, tough struggle.

BOLLYKY: Great. Thanks, Matt. That's very -- that's very helpful. Now, to turn to Bernhard --

(Audio break.)

Right, thank you, Lori (sp). Applying high-impact,low-cost technological solutions to emerging and epidemic diseases like malaria, like meningitis around the world.

And with that introduction, Bernhard, why is -- why is PATH turning to diabetes now? Why is this an area of focus for them?

BERNHARD WEIGL: So thanks, Tom. Yeah, I mean, clearly, as Tom Frieden earlier mentioned, diabetes is among the big four, and is -- the scale of the diabetes problem -- and I'm going to talk specifically about diabetes, although we are doing other NCD work as well as at PATH at this point. But diabetes rates are now approaching, you know, 10 (percent) to 20 percent worldwide in urban areas, certainly, even in developing countries. Many cities such as -- big cities in India such as Chennai or also some sub-Saharan African cities have, essentially, rates that are pretty similar to what we have in the U.S.

And what that practically translates to is that you have -- in India and in China, you have about a hundred million each of people who have either diabetes or pre-diabetes, gestational diabetes -- that's diabetes during pregnancy, which is also a predictor for a type 2 diabetes later in life; is also in the 20 percent range. So one out of five women in many places does have gestational diabetes.

So this is an enormous scale, and at PATH we are a pretty large NGO with a long history of innovation in technical areas, primarily, but also behavior change and health system strengthening. And we did -- we have offices in many countries in the developed -- developing world, and we did a survey recently as part of our chronic disease work to ask about diabetes care practices in each of the countries, and the needs and so on. And I think we've never gotten sort of a -- such a -- as tremendous a feedback and interest from our country staff as with diabetes. It was really that almost everyone told us, oh, yeah, I've got somebody in my family, I've got somebody -- oh, I myself have diabetes. I mean, it's -- there is a lot of interest and a lot of, clearly, need around this. And so there is -- there is probably -- the staff or the countries are really the -- developing countries really are the drivers for this right now, and that's, I think, why we are doing this.

So why those increases? The enormous increases are part in due to lifestyle changes, just as here. People are -- lead more sedentary lives in urban areas, clearly less exercise, less walking, more food. But there's also a second component to that, and that is malnutrition in early childhood or during pregnancy of the mother strongly predicts diabetes later in life. So essentially what it means is that -- I mean, a simplified explanation that I've been given -- I'm not an MD, but is that, essentially, malnutrition leads to defects in the pancreas, essentially, and that eventually has impact on diabetes.

But the -- what you see, practically speaking, is that the typical diabetic -- while you do have, you know, overweight diabetics in developing countries as well, you also have plenty of skinny ones. In several countries where you have known periods of famine, at periods about, you know, 30, 40 years later, you have these outbreaks of diabetes that are clearly correlated to those earlier famine periods. So there's that effect, too.

It is -- in short, the scale is enormous, the health impact is enormous, the economic impact of diabetes is enormous. One of the reasons for that is people get diabetes when -- and get the health impact of diabetes when they are the primary breadwinners, when they finally make enough money to, you know, support a family. One person might support many people, you know, more than here. One person -- one working person with a salary might support 10, 15 people or so. So those people would then, you know, if they can't work because they have, you know, foot -- nonhealing wounds, foot problems, kidney problems and so on. That has an impact on a lot of people.

So the task facing chronic disease care, compared to, say, many infectious diseases seems daunting in many ways. In chronic disease, there are generally no silver bullets such as with vaccines. And, you know, with polio, you can essentially get rid of the problem. Diabetes we'll probably never be able to get rid of the problem, but we can -- so care is essentially needed for life, and it can be also very fairly complex. It usually requires a lot of different things. It requires drugs, it requires lifestyle change, social support, health policy approaches to -- for prevention and so on. So lots of different things.

But at the same time, the good news is that there are proven models out there for diabetes. Not as many as for tobacco because we are essentially 20 or 30 years behind the curve, here, I think, with diabetes prevention, but they are emerging. We're working with several of them.

There's a very exciting NGO in Cambodia that we're working with. They are able to provide diabetes care for relatively advanced diabetes patients for about $5 per person per month, and that includes insulin, it includes a degree of testing, it includes social support. And the way they do this is by, essentially, task shifting. They're giving the patient more responsibility for their care. They're giving peers -- other diabetics -- more responsibility for helping, and by doing all these things, you can essentially provide care at a much lower cost, and also benefit from this whole social support around this, and that's actually a critical outcome and also a very positive outcome. Health -- in chronic disease -- health, especially chronic disease kinds of issues, can only be addressed if the patient plays a significant role as part of his or her own care. And that's something that -- actually, with, I think, diabetes care, this might actually -- these models might actually lead to allow people to essentially take more responsibility and also have more tools available for them to take care of other diseases as well.

BOLLYKY: Thanks, Bernhard, for that.

James, to turn to you, why -- for many people in the room, I imagine it's news that road safety is a global health issue. Why should the international community be concerned about road safety in low- and middle-income countries?

JAMES HABYARIMANA: Well, that's a kind of more difficult question than I -- I'm not sure I can answer it in whole, but let me give -- let me give you a sense of what the numbers are and why I think the international community should care.

So it's not as important as diabetes or tobacco right now. It's not any of the big four. But if you look at the trends, it is -- it will be essentially in the next 20 years certainly one of the big five. Right now road traffic incidents kill about 1.3 million people every year. The numbers, of course, are I would say fairly sketchy. There is a lot of modeling that has to be done to actually sort of generate these numbers. And so as Tom was saying earlier, the fact that a lot of people essentially sort of are born and never actually sort of properly documented makes, you know, some of the challenges around actually understanding these trends a bit more acute.

Ninety percent of the deaths from road traffic incidents essentially occur in middle-income and low-income countries. And there isn't actually a huge gradient between low-income countries and middle-income countries, so between -- amongst those countries there's actually sort of about an even split in fatalities.

In addition kind of the fatalities, you know, there were nearly about -- you know, between 20 and 50 million individuals are injured by road traffic incidents with huge social costs. So people sort of drop out of the labor force and medical -- you know, households essentially have to run down savings to actually sort of look after people who are disabled as a result of road traffic incidents.

And one of the things that I'm not -- you know, since this is the first time this is being discussed in this context, I am -- I am sure not many of you know that the leading cause of death amongst 15- to 29-year olds around the world is actually road traffic incidents. It's not -- you know, it's not HIV; it's not -- it's not tobacco-related. So essentially (kind of ?) this is a disease condition that kills young people. Amongst 30- to 44-year-olds, road traffic incidents are essentially (kind of ?) a third-leading cause of death, OK? So it is a disease that -- or a condition that affects -- certainly affects most productive sort of sectors and groups of individuals.

And I think -- I think for that reason, for the fact that as countries get richer, people essentially sort of are going to be using cars a lot more and, unfortunately, kind of the drivers of -- (the basics of ?) poor road safety, you know, will probably sort of not be addressed as quickly as people are going to be sort of getting into cars and -- you know, so I should also say that, you know, about half of the people who die from road traffic incidents are people who are inside cars. The other half are actually sort of a vulnerable population of other road users, so people on scooters, people on bicycles and pedestrians.

And one of the challenges of addressing this issue has to do with the fact that, you know, we could, if every country sort of becomes like Sweden, we have well-structured sort of spaces for people to walk, you know, traffic regulations that are essentially sort of well-enforced, then we -- you know, I suddenly should not appear on any panel to talk about road safety.

The problem with that, of course, it takes a really long time for a lot of countries to get to where Sweden is. You know, building good roads is very expensive. Having -- you know, forcing people to sort of drive safer cars, very expensive. Lots of governments around the world have been trying repeatedly to sort of get unsafe vehicles off the road -- very, very difficult to do. I mean, so regulation is an issue. Enforcement is certainly (sort of a big issue ?). And ultimately sort of, you know, what people can afford is a huge issue. And so I think, you know, those things will take a long time.

But I think there may be opportunities to actually focus on things we can do now. You know, in some of the work I've been doing in Kenya, we actually sort of talk to a few owners of minibuses which sort of transport a lot of people to -- you know, within the city and between cities. And, you know, she was sort of quite forthright in saying, look, vehicles -- you know, the roads don't kill people, the weather does not kill people -- and these are clearly issues that actually sort of -- sort of factors in making roads very unsafe -- you know, drivers and individuals, essentially, sort of are responsible for a lot of the incidents. And I think, you know, we can -- we can certainly sort of target strategies at these individuals and trying to sort of get them to behave better, and we could potentially sort of make a dent in this -- in this scourge.

BOLLYKY: Great.

Now, each of you are involved in initiatives that are potentially scalable in this area. I would be grateful if each of you could describe a specific context in which it works.

Matt, you talked about the political will, if needed, to implement the effective tobacco control policies that are making a different in Uruguay, Turkey and the Ukraine. What has the organization you lead, Campaign, done with its partners to try to enable that to happen?

Bernard, in terms of the support you're giving through diagnostic technologies that can be used in low-resource settings, what convinces you that what's happening in Cambodia is something that could be scaled up?

And James, what is it about the work you're doing with behavioral cues with minibus taxi drivers in Kenya? Tell us a little bit about the effect that's having and why you think that behavior cues might be a successful strategy for addressing a broader range of issues in other settings as well.

MYERS: Those are big questions. But I think there is actually a common set of themes with regard to them, certainly for tobacco and I think ultimately for others.

I think it starts with what Tom said this morning, which is documenting the problem. To a stunning degree, public officials are often unaware of the magnitude of the problem, of the severity of the problem. So the Global Adult Tobacco Survey, youth surveys, are extraordinarily helpful.

Second, I think education is absolutely critical. We all assume that everybody understands these issues, but when you go to China and you see 60 percent of doctors smoking, you realize that either the level of knowledge is far less than you thought, or the defensive mechanisms are far higher than you thought -- and therefore doing that.

Second, you have to create a sense of urgency. And I think that does -- in fact, our organization is called the Campaign for Tobacco-Free Kids -- I think that does involve a comprehensive campaign, one that involves educating the media, engaging civil society and working closely with government altogether, not in opposition, to create a real of urgency. Politicians only act on things -- in every political system, I should say -- where they see both a political benefit to themselves to acting and a -- and a political detriment to themselves for not acting -- so I think creating a sense of urgency by working with the media and civil society organizations and creating a drum beat of activity afterwards.

And then lastly, I think you have to have a concrete set of what Tom called the technical package to find solutions that you can demonstrate work, are affordable, sustainable and, while we all like to think of ourselves as public health people, politically popular. And I think you need to bring that package together with them as well.

And then last, I think you need to recognize that the change for NCDs is it's not a sprint; indeed, it's not even a marathon. As somebody else said -- I heard say the other day, it's a relay. It's a long-term run with lots of partners who need to be involved to make it work. It has to be a comprehensive effort.

But what we have seen is, in political systems as disparate as Russia, Ukraine, Uruguay, Brazil and a lot of other countries, when you bring those factors together in a manner that's culturally sensitive to their political system, you can get a response, not uniformly, but almost always so that the progress that we've made over the last four or five years in one sense is discouraging -- you could hear Tom say, no individual is covered by all of the MPOWER package -- and in the other sense is remarkable if you think about the number of countries that have adopted these proven solutions: tax increases that are the single most important thing that one could do, warning labels that sometimes catch people unaware, true understanding of the effects of secondhand smoke -- 600,000 people globally died last year from breathing somebody else's tobacco smoke. That's a stunning number. But the -- 22 countries now are protecting people about it, in a way, and that's a scalable thing that doesn't cost a lot of money, because it actually doesn't hurt business.

That's what we've learned, is that it helps business, just as it did in New York. They all told Mayor Bloomberg that when he adopted a smoke-free law in New York City, that tourists wouldn't come from Europe, that the restaurants would go bankrupt, that New York would become a desert, that people would all sit home -- or whatever it is. And in fact, the restaurant business went up. Restaurateurs loved it, tourists loved it, and New York has now become a role model.

So I think there are some core lessons in terms of how you go about these activities, to create a sense of political urgency, political priorities, and to begin to change the social norm, because you have to do both.

BOLLYKY: OK, thanks, Matt.

Now, Bernard, what about adapting existing treatments? Is that a scalable strategy to you?

WEIGL: Certainly to a -- to a degree. I think we, for diabetes, we -- may be the biggest issue to start out with, and that's where we thought we could maybe have some impact. Also our path was the fact that most people who have either diabetes or pre-diabetes don't actually know that they have it. So the vast majority, or the significant majority of people -- and that's true -- that's true everywhere, but it's more true, of course, in developing countries. In the developed world, there is no -- there are no generally formal programs for diabetes screening. But eventually, somebody is going to find a high glucose value, and then somebody will follow up, because we have this tradition and this model of going to preventative care visits or to some kind of random care visit.

That sort of -- that sort of model really doesn't exist in developing countries. So there, you really need -- you need to have tools for screening for diabetes that are simple, low cost, and can be performed just about anywhere. And, practically speaking, the way you screen for diabetes now is you need -- you need to fast, and then you need to do some blood testing, and then after that, you might have to do a glucose challenge, especially if you're tested for gestational diabetes, and what that means is that it's a relatively complex, sort of multi-step process.

And so -- we know that our methods out there, or at least under development, that don't require at least fasting. And that -- instantly, what that allows you to do is that -- pretty much anybody who walks through the door for health care setting, if you have a strip test that tells you if somebody has diabetes or gestational diabetes, then that will be really, really helpful, because then you can start intervening, or at least start counseling, and so on.

So part of what we do is, we're evaluating a bunch of new diabetes technologies. Some of them are screening technologies, some of them are non-invasive. They essentially measure something called advanced glycation end products through the skin, so that's a pretty exciting method. But there's other -- there's other methods as well.

Another -- that's for the technological area. The less technological portion of this is to seek out populations that will particularly benefit from screening, and some of those will be populations that have a pre-existing comorbidity, a disease such as, for example, TB. That was mentioned earlier today. People who suffer from TB are much more likely to have diabetes, and people who have diabetes are also much more likely to contract TB, also. So cross-screening among those two conditions, in populations who are, who are already identified as having TB, and using that subgroup and doing the TB -- diabetes screening there is very useful. We're doing this in Mexico, and we're going to do it in Tanzania also, soon.

Conversely, if you ask among identified diabetes populations for persistent cough over the last few weeks, you're much more likely to find people who have TB than if you were to do that in the general population. So that's a very useful thing to do, and we're embedding that in existing programs, and we're learning from these existing programs there.

And the same could be done, we believe, also for HIV populations. HIV have all these -- have the -- HIV patients have the syndrome of accelerated aging conditions, which includes, you know, heart disease and so on, but also diabetes. So there's value in doing accelerated screening in HIV populations.

So there's a lot of these different kinds of interventions that could be done, but none of these are -- these are not useful by themselves, because all they do is, they tell you if somebody has diabetes, so now you got to do -- so the big question is now, so what are you going to do with those people once you have them?

And this is really the big effort that we have with a Cambodian NGO where -- that I mentioned before -- that has figured how to -- how to at least treat advanced diabetics for $5 per month per patient. And really, this is a model we're looking at very carefully right now; we're doing evaluations. We're also -- we will, we will do some tweaks to the systems, working with that, and then hopefully, ultimately scale it to other places as well. But we have done a pretty wide survey worldwide, and we have not found very many places where that sort of level of care can be provided. And really, the trick is to have -- we always knew that peer support is very important for diabetes care and chronic disease care, because you have to keep motivating people to get care, and we know this also in part from HIV care, where this has been successful there.

But what we didn't appreciate, I think, sufficiently is the fact that peers can actually be used, to some degree, especially for the conditions like diabetes, which, here, in this country, in the developed world, are treated, in many ways, by the patients themselves. I mean, patients themselves have their own glucose meters, they do their own injections, they do go to the doctor once every three months or so, but for the most part, their care -- they become experts at their care. They're truly -- I mean, there are diabetics out there that are -- that know every bit as much about diabetes as your average primary care doctor.

And you can use that, and there's no reason why people in developing countries could not learn as much about this either, and take on a greater degree of responsibility for diabetes and, I think -- we also think -- for other -- with the advent of more -- things like more point-of-care diagnostics and better kinds of treatments regimens, maybe also for other chronic diseases.

BOLLYKY: Great. Thanks, Bernhard.

So we've heard about using evidence and monitoring evidence on technical packages that work, leveraging existing treatment platforms and technologies and using peer support to advance treatment. James, tell us about behavioral prompts and social cues and the role that they can play.

HABYARIMANA: Yeah. So I think the work -- I should mention that the work that I'm discussing is actually sort of a collaboration with Billy Jack (sp), also at Georgetown -- was motivated by, kind of, two things. So one was if you travel in most -- you know, so I do a lot of my research in Africa -- if you travel a lot along the road, you know, every five miles or so there is a signpost that says, drive safely, or something like, speed kills. I mean, something to that effect. And, you know, so we were actually, sort of, both in Kenya, actually working on a well-funded sort of condition, you know, HIV, and sort of thought about a lot of the hard, sort of, messaging that is used in other sort of conditions, and saying, you know, what if we brought hard messaging to road safety?

So, "drive safely" is sort of fairly passive; it's -- you know, it's really not, sort of, engaging whatever cognitive processes, you know, drive behavior change. So what if we plastered, essentially, kind of, the highway from Mombasa to the western border with Uganda with essentially, kind of, pretty aggressive messages? And that's essentially, kind of, you know, that was the idea, and we went around and sort of talked to lots of advertising people, and they came back with a cost, and we basically sort of gasped because, you know, we work with budgets of, you know, $50,000 dollars for a research project, and then we're talking about, sort of, millions of dollars. And so we thought, OK, this -- and in addition, of course to the measurement challenge, this would be very hard to figure out whether it was working or not, and how, but we basically sort of went back to the drawing board and sort of thought, OK, what if we take these messages inside?

So messages on the highway would essentially, sort of, affect both the driver and the passenger. What if we took these messages inside the vehicle in the form of stickers, and basically, sort of, motivated passengers, who have some sort of key advantages over, kind of, the late regulatory system or even the person who actually sort of owns and operates the vehicle, because the passengers are always with the driver, and they exactly -- they can observe what he's doing. Right? So, you know, my sense of what might be dangerous may be different from yours, but if there is something in the vehicle that legitimizes me speaking up and saying, you know, slow down, or, you know, don't be reckless, that maybe that kind of social pressure may actually sort of affect the driver's behavior.

I don't -- you know, so we -- so Don (sp) wrote a proposal on it, and I can't tell you how many people laughed at the -- essentially, kind of the fact that this would not work. And if anything, this actually sort of may be very dangerous, and I think we -- you know, we pretty -- we recognize the fact that that actually was -- it was possible, right?

So lots of people shouting in the back may actually sort of be another form of distraction -- (laughter) -- and, you know, so -- and this is kind of -- you know, cell phone usage in Kenya had not sort of taken off in the same way, so this could have been essentially kind of an important social distraction.

So we -- we tried to sort of tailor these messages so that only the passengers saw them. And I know there's kind of big questions -- sort of, what do you put on the sticker? And you know, so we -- we're economists, we know nothing about how to persuade people to buy Nike shoes or Coca-Cola or whatever it is, so we thought, you know, let's see if we can actually with an advertising agency in Kenya to come up with a bunch of messages. And of course, we sat down with them and thought we knew the kinds of messages that would work -- very subtle messages like, you know, missing child in a picture at a graduation or a wedding, you know? And we also had very, sort of, strong messages because we also figured that actually sort of may work.

And then we did a little survey -- about a -- you know, 200 passengers -- and said, you know, which of these stickers would essentially sort of make you -- you know, sort of -- get you to speak up and say something to the driver? And it turns out that, you know, the stickers that had severed limbs -- you know, that had several -- or you know, a lot of, you know, disfigured faces actually did -- were a lot better than our little subtle messages about sort of, you know, you don't want to miss her graduation, that sort of thing.

And so we actually sort of did a pilot of about 2,500 of these vehicles in -- that operate this -- in sort of -- in and around Nairobi. And in a thousand of them, we put essentially, kind of, four stickers -- kind of two stickers with some pretty strong messages; two stickers with just essentially kind of text -- but you know, very, very well designed by, you know, these -- I wish I could show you some of the pictures of the stickers we used, but they are available online -- and found that in the stickers we -- in the vehicles we put the stickers in that accident rates had actually sort of fallen by a half.

So you know, when I first saw these numbers I thought, OK, there's something wrong with the data I'm using. So we were also using very, very good data. We went to the insurance companies -- there's four insurance companies that actually sort of cover this entire sector -- and they were, and surprisingly, sort of willing to actually share, you know, whatever data we wanted to get from them, OK.

And so they didn't know which vehicles had been treated by our sort of intervention, so it's not like that, you know, they could have cooked the books to get this to work. And in fact, most of them didn't believe that this intervention would actually work, right? So the drivers of a -- powerful individuals, passengers, once they've paid their fare, they're effectively captured, right? They -- you know, what else, what other bargaining power do they have? If they make noise, I can just tell them to get off the bus.

So we looked at these number again and again. And essentially, it kind of was very, very clear. It was a randomized controlled trial, so you think, you know, we're taking care of a lot of essentially, potential -- (inaudible).

And you know, we -- then we took these results to a few insurance companies. You know, three of them basically said, I don't believe -- I don't think, you know, Kenyans are, essentially, kind of individuals that don't, you know, speak up -- you know, and drivers are powerful individuals. One insurance company, you know, basically said, I'm not sure I believe these results, but let's see if we can do something together and, you know -- you know, let's do this on a much larger scale, because if you're right I could make a lot of money. And so we've been working over the last two years with one insurance company, covering about 12,000 vehicles -- so roughly one-third of the fleet in Kenya.

And in addition to that, we sort of decided to extend, sort of, the intervention to actually radio. So we have a complementary sort of radio campaign, which of course reaches more people -- and certainly, exceedingly more scalable than sort of putting stickers on vehicles. One of the things about the stickers is -- you know, unpleasant images while you're sort of on a two and a half hour journey between Nairobi and something may lead either passengers or the drivers to actually sort of pull them out.

So one part of our intervention had to be to bribe the drivers to keep the stickers in. So every week we run a lottery, and the lottery -- you know, and the -- lotteries can be very cost-effective ways of actually, sort of, delivering incentives. So I think -- I -- you -- even though lotteries in other contexts are sort of potentially questionable, I think in this context, actually, sort of a lottery's actually a good way to do this. And if a vehicle has all four -- all four, five stickers, then they're eligible to sort of win something like two week's worth of wages for the drivers.

So that's actually sort of kept the stickers, and that's -- you know, it does raise the cost of this. But you know, these stickers cost about a dollar to produce, they seem to last about, sort of, at least -- you know, from the data we have, up to six months -- so you know, pretty cheap, if they produce the results that we actually saw in the pilot.

BOLLYKY: Well, you all have been enormously patient. So thank you for that, those are great examples.

If you'd like to speak, please just raise your hands and I'll go through. When you speak, please identify yourselves and your affiliation.

QUESTIONER: Hi there. I'm Ella Gudwin, with AmeriCares. Thank you so much for the comments.

I can just -- a lot of comments resonated; one in particular. We were doing a cancer intervention in Armenia, and the head of the National Institute for Cancer was a chain smoker. So I have seen that.

One of the things that we're looking at currently, in the slums of Mumbai, is treating diabetes and hypertension, and looking at them -- co-managing them, because of the overlap. And we found that -- we have mobile medical units, and we're trying to extend care and provide a primary care home to those who wouldn't have access to a fixed clinic by rotating the vans on a fixed cycle of 15 days. And we've found that when -- in the first couple of months we're mostly seeing acute issues, and the chronic patients really start to show up after about three months, and then they start to expect the van and they start to come. And so it's really interesting to think about how to tie the treatment and -- with the other interventions like TB and HIV.

So picking up on that theme, in Malawi we're seeing the same thing, and I wanted to get a sense for how you piggyback the diabetes and hypertension intervention into an HIV patient outreach infrastructure -- because that is so strong, particularly reaching out into rural areas as well, and I think there's a lot to learn there. And I'm wondering what success you've seen on that.

WEIGL: So yeah -- I mean, generally piggybacking, I think, is certainly -- we're looking for this. I mean, clearly -- and that's some of the work that we're doing with the TB programs and also the HIV programs. There is, though -- a bit of caution here I think is appropriate because both TB and HIV programs are fundamentally not self-funded; I mean, they are fundamentally driven by donors. I think the idea with chronic disease programs in -- at least in the long term is that they will ultimately have to be self-funded because the scale is so enormous, there's no way that the donor community can cover chronic disease care for everybody in the world; that's just not happening.

So it's -- the big -- you -- I mean, as exciting as HIV progress -- you know, the PEPFAR program -- and so, yes -- I mean, it's clearly -- it has shown what can be done given appropriate funding. But the thing is, I mean, we -- that sort of funding will not be forthcoming. We know that, for -- certainly for that reason that -- setting up programs is one thing, but continuing them in the long term is probably not going to happen.

So I think what -- the models that you showed here are actually exciting because those are very low-cost models -- I mean, mobile clinics, for example. If you can do that, and if they sort of support an on-the-ground kind of peer support model, we think that sort of thing is exciting. We've looked for some of these models.

There's a few in India -- Bandhan is trying to do this in India. There's also -- there is -- much of the work right now, on the ground -- there's actually still charity care, even by people -- even local charity care. So local hospitals in India, for example, they do charity care. So they have -- I mean, we work closely with Dr. Mohan in Chennai; he is one of the main -- the biggest care people in India. And what he does is he has these outreach where he sort of provides close-to-free care once a week or once a month or twice a month or something like that. Cancer hospitals do the same thing. Tata Memorial in Mumbai provides cancer care in outreach clinics -- for example, one in Barshi; that's a rural clinic. So that's sort of the -- kind of the model.

And that works, you know, to a degree, but I think it doesn't give you the kind of reach that you ultimately need. So I think we need this truly low-cost model on the ground, and I think you just have to look at the specific cost factors in there, I think.

MYERS: Can I just add a couple of things to that, because I think that's one of the reasons why, if we're going to talk about scalability, we have to talk about political will. Because we can create models of treatment, but unless we create self-sustaining mechanisms and government buy-in to them, then that's what they'll remain, is models of treatment.

And in some respects, while there's lots of reasons to argue about lumping various NCDs together about it, the most effect way to reduce tobacco use is also a potential mechanism for funding a number of these programs, and that is raising taxes on tobacco products. It is the single most effective way to reduce tobacco use. You know, public health people, health people don't think about tax policy as health policy, but this is a case where it is health policy. And raising the tax on tobacco has been shown not only to increase revenue and decrease the number of tobacco users, it's a potential core source of funding for health programs. And there's even a moral and ethical argument that it has to be used for that, because if you're going to tax people to get them to change their behavior, particularly poor people -- and increasingly, that's what we're talking about here -- then there's a very strong argument that that money should go back into the community to help them deal with their health problems.

And as Tom said, you know, the evidence on tobacco use -- and the same is true with alcohol -- is that the disease burden falls disproportionally on the poor -- on the poor, and they're most responsive to changes in price.

So you know, in a -- and places that are looking to solve multiple NCD problems, this is one of those places where if people came together, the potential for coming up with a sustainable model, rather than just models that will work for a short period of time, grows exponentially and -- but that also requires, then, people come together and engage in what some people aren't comfortable with, which is the political activity to create the political will to adopt that political -- that model to go forward. But it's hard to see how you're going to long-term fund the kind of activity you're talking about there. And there's just a powerful argument for it. I mean, WHO and CDC says that literally half of all NCDs are preventable -- and some of that includes diabetes -- by changes in tobacco, you know, diet, nutrition and the abuse of alcohol. So there is a marriage there, if we see it in a broader frame.

BOLLYKY: We have --

QUESTIONER: Just -- sorry -- just one last comment: The other option that we're looking at is fee for service, because we have an experience in El Salvador where 60 percent of our revenues in a rural clinic are generated by the patients themselves. And it's very poor population, and so we're looking at that for India as well.

BOLLYKY: We have about eight minutes left, so I want to make sure we get as many questions as possible.

Charlotte.

QUESTIONER: (Audio break) -- where earlier customers subsidize cheaper care for diabetics. And have you worked with anyone in the public sector, is there any government that you've worked with or public hospital system in any of the countries where you're operating that might provide a model for other health ministries?

WEIGL: So we have not yet formally worked with any governments yet. We intentionally looked for sort of smaller scalable programs, also primarily because we -- you know, we didn't see among the -- among the government efforts what we were looking for in the sense of providing truly sustainable care.

I think there's a few exceptions. The Bangladeshi program is actually generally seen as one of the leaders that -- they're relatively good in providing care.

But for the most part, what government health care programs were able to do in developing countries for diabetes is tertiary -- a little of tertiary prevention, essentially care for symptoms of diabetes, such as eye care, foot care, that sort of thing. That's the kind of care you're getting. But you don't channel -- you might get -- in many countries you actually do get free insulin, but you don't get free testing, for the most part, or certainly not daily testing level or several times daily testing that we -- that you would expect here.

So that level is simply not attainable, it seems like, from government programs at this stage, and we were looking for programs that did do at least a degree of that, that went beyond that level of sort of very, very basic care.

So we haven't -- I think the short answer is, I don't think there are any government programs out there that really do provide comprehensive care.

BOLLYKY: OK. Thank you.

Other questions. Laurie?

QUESTIONER: You -- Matt used the phrase we -- what's holding us up is political will. There's political will targeting host countries but also political will that could target the donors, and I don't know why it's not been used.

For example, we now -- as of the international AIDS meeting in July, we know that there has been a sudden surge in sudden-onset cardiac arrest among people on long-term antiretroviral therapy. It would seem, therefore, that it's medical malpractice to distribute antiretrovirals without doing cuffing for hypertension.

Similarly, we know that if you have tuberculosis, your likelihood of surviving treatment if you're a smoker is markedly lower than if you're a smoker. So it would again seem to be medical malpractice not to tie donor funding for tuberculosis treatment to tobacco intervention.

And then, you know, we have -- I don't know how much more evidence we could possibly need to demonstrate that maternal health programs that fail to deal with maternal alcoholism or smoking will doom the baby that does survive that pregnancy to suffer all the things that we know are the results.

So it seems like you have obvious targets -- I've just come up with three off the top of my head -- that -- where the pressure could go to the donor to say, you don't have a right to give out this money and give out these pills and do these stovepipe programs unless you tie it to these other specific interventions.

MYERS: I'm not sure I'd use the term "malpractice" with regard to it, but I think there's a powerful case to be made that donors who are looking at narrow stovepipes that don't do that are missing the boat. Since I think there's -- it's fair reason to assume that most donors actually want to do good and they want to be effective, and you know, as Tom talked about this morning, increasingly, I think, we have to push donors to be coming up with measurement tools to ensure that what they're doing works, even those of us who are the recipients of those donations, because if the goal is to solve the problem, then I think what we need to be talking to donors about is we need to have tools that are measuring them. In each of the examples you gave, a donor would fairly rapidly discover that they were being far less effective as a direct result of the narrowness of their viewpoint, and either need to bring in others, need to change their criteria.

So that's the way I would go about it. It's one of those things that we need to ask donors to be measuring carefully the results of what they get. And it's hard for me to believe that we're going to have a lot of donors out there who will want to continue to operate on ideology, rather than data.

BOLLYKY: We have two questions left and about three minutes to do them, so I'm going to take them both and then let the panel respond.

Wang Zho (ph).

QUESTIONER: I am Zho Wang (ph) from the council. In the -- this is to Matt, actually -- (chuckles) -- in the speech, you talk about the importance of political will. Actually, this is just to follow up what Laurie's question is. We know that China is the largest consumer, also the producer of tobacco, right? But if you look at -- actually, this is from the report, right? -- none of the Politburo Standing Committee members in China actually are smokers. So I'm curious. Maybe from your experience -- (chuckles) -- you know, the Campaign for Tobacco-Free Kids in China, you know -- what accounts for this lack of political will in China?

QUESTIONER: Hi. I'm Dara Erck from the Gates -- Bill and Melinda Gates Foundation, so from the donor perspective. Do you know what the ask is -- you know, for neglected tropical diseases, there's been tons of modeling on what it costs to develop a vaccine or, you know, a TB or an HIV vaccine. We know those numbers, tons of modeling. Do we know what -- you know, do you know what the ask is for diabetes or for tobacco-free?

BOLLYKY: And to add to the speakers' burden, we have really just a couple minutes left, so if you'd like to make some general remarks in the 30 to 45 seconds you have, that would be wonderful as well.

MYERS: I think China you can answer in a -- overly simplistically, in -- one, the government owns the tobacco company and historically has made between 5 and 8 percent of its revenue -- and matter of fact, there's one province that makes 70 percent of its revenue from tobacco. So you have to convince the government that the up -- that risk-benefit ratio, the return on investment, overcomes that for them.

Two, you have a government that is extraordinary (sic) fearful of disrupting the status quo, given their power relationship, and therefore you have to convince them that taking strong on -- action on tobacco is not only beneficial from a health standpoint but will, in the long run, be received in a positive manner, so it doesn't threaten their power base, which I think is incredibly important.

And three, when you look at the government's priorities, actually health has never been -- has not been a very high priority for the Chinese government. And again it relates to what they see as China's future as an economic power and the political power of those, which is the reason why we have to change that political dynamic and that debate if we're going to succeed which is, you know, again, a reason a lot of us have been doing a lot of work with the media in education, to create a real drum beat within China to demonstrate that that's not the case in that way.

I think it's doable, but it's not doable in a 25 minute -- it's not a simple vaccine. I think we've already seen a change in media coverage, debate and sensitivity. We have a long way to go, but I think we've begun walking up those steps.

BOLLYKY: Great. Bernhard and James on targets, asks in this space?

WEIGL: Sure. I mean, I don't have a -- (inaudible) -- number here, but I have a very narrow number, and that is for -- I know for diagnostics for global health the amount that goes towards diagnostics for chronic diseases for global health is about 2 percent of the overall. And 2 percent is obvioulsy not corresponding to the 50 (percent) or 60 percent morbidity and mortality that comes from chronic disease. So I think the ask will be gets somewhere closer to that 50 percent level. I think that's, you know, probably the short answer.

BOLLYKY: Great. James.

HABYARIMANA: So -- yeah, in terms of the ask, I'm for -- (inaudible) -- at least in the context that I'm working. I actually think that this is maybe one area where, you know, this can be financed sort of almost entirely by insurance companies that essentially have a lot to gain. I think, you know, leaving government out of some areas is not a bad thing. I think there's clearly some places where government can do great things.

I think in this context -- and sort of -- you know, political will is -- I think is a great thing. I think, you know, and a lot of the work that's being done at the high level, WHO, you know, this decade for -- (inaudible) -- safety or some of the work that the Bloomberg Foundation is funding, sort of very high-level, institutional intiatives, will build essentially kind of the the political will.

I'm not convinced that over the next five years that political will will translate to a lot of action on the ground. I actually think that we have to find ways to essentially sort of motivate local actors who can experiment and generally sort of produce that, I think it was the technical package that Tom Frieden sort of talked about, that essentially will the basis. I think political will without strategies will not lead to any sort of useful action.

BOLLYKY: Well, that is a great message to end on. I've kept you here three minutes over and it's our custom at the council to end on time, so I apologize for that. But please join me in thanking these three excellent speakers. (Applause.)