I’m here in Washington, DC, at the second annual mHealth Summit, where I gave a talk and conducted a keynote interview with Bill Gates. During a wide-ranging discussion, Bill stressed that improved health is the key to controlling many of the world’s most challenging problems. In keeping with the mHealth theme, Bill discussed the role of the mobile technology—especially the cell phone—as a tool for improving medical services in various populations.
Here are just a few of the highlights of the Q&A:
Spotlight: On-demand video
AI Explainer: Foundation models and the next era of AI
Explore how the transformer architecture, larger models and more data, and in-context learning have helped advance AI from perception to creation.
Kristin: You’ve stated that the mobile phone, like other computing platforms before it, is not the panacea for healthcare. Given the cell phone’s ubiquity, what are your thoughts on the possible advantages over the preceding technologies?
Bill: Computing technology has been great for healthcare, but primarily on the research side. Although it would be difficult to measure, some combination of the Internet, digital database, and collaboration tools really have changed medical research. In the case of the cell phone, there’s a chance to go beyond that and actually be there with the patient, there in the clinic, which might not be staffed with a fully trained doctor. So there’s a lot of opportunity, but we have to approach these things with some humility. There’s not always Internet and data connections out there, and when people are sick they are often too late for some cell phone type application to do something for them. There’s absolutely a role, but we have to hold ourselves to some pretty tough metrics.
Kristin: So what, then, do you see as the main applications for mobile health, and what specifically will be most impactful?
Bill: I think it is always valuable to go back and ask what the key metric is you’re trying to improve. Maybe the simplest one is the number of kids who die each year under the age of five, which is about 8.5 million. If you go back to 1960, that was about 20 million, so there’s been a pretty dramatic reduction. About a third of that’s been achieved by an increase in income, which gives you better nutrition and better living conditions, but the majority of it’s been done through vaccines. Smallpox was killing 2 million a year; now it kills zero a year. Measles was killing a million and a half; now it’s killing about 300,000 a year.
So, now we say to ourselves, what can we do with this 8.5 million? That actually breaks into two parts: There’s the first 30 days of life, which is about 3.5 million deaths, and then there’s the 30 days to 5 years, which is about 5 million deaths. There again, vaccines are going to be the key—malaria vaccines, respiratory, diarrheal—but the question is, how will cell phones fit in? If you could register every birth on the cell phone, get the fingerprints and get a location, then you could go around and make sure the immunization happens. Vaccination coverage in some poorest areas, like the north of Nigeria and the north of India, are below 50 percent. So you could get a huge improvement if you could just take the vaccines we have today and get those delivered.
So when I think about the biggest impacts [for mHealth], I think of patient reminders, improvements in the supply chain for medical goods—making sure there’s no counterfeits and no stock-outs—and finally, this sort of digital patient record, where the high payoff is vaccination coverage.
Kristin: In regards to the Gates Foundation Grand Challenges, you’ve had two calls now and are about to announce a third, can you tell me what type of projects have been funded?
Bill: We’re in a phase now where we are doing these Exploration Grants, which are kind of small, $100,000 each, trying to catch a set of wild ideas and people who wouldn’t ordinarily apply. We’ve done some in mobile health … If you want to do cheap diagnostics, the idea that your communication screen comes from the mobile phone and all you have to do is sample preparation and signal detection and then that whole backend is just sitting there for free, it’s clearly a huge win. Some of these diagnostic things, in malaria or TB, will probably be the first areas where we’ll be able to say without this mobile phone application people would have died.
Kristin: You’ve said that R&D can prevent future disasters and that research investments are important to mitigate some of the problems facing the world. These problems are vast and diverse, and often dependent on location. Do you believe generalizations can be made, and if so, what are they?
Bill: In general, the world underfunds research because the person who takes the risk of doing the research doesn’t capture the full benefit of having done it. Capitalism does amazing things, but it has this systemic problem in terms of research not doing enough, and it has another systemic problem in that the needs of the poorest won’t be prioritized the way they would if you put a more human-values driven system in. Of course government steps in for things that the market doesn’t work well on, such as funding for basic research. The Gates Foundation tries to take the diseases of the poorest and fulfill that role that the market-driven signals aren’t going to.
How do we get more investment in research? Well, the Internet is a great platform. The world as a whole is getting more science and engineering graduates. And so whether it’s energy or biology or robots or software, people underestimate the amount of innovation that’s going on. Part of the reason people have such bleak views is that they assume technology stays the same.
Kristin: How does education lead to better healthcare outcomes?
Both health and education are great examples of behaviors where you have to put in effort now for a benefit that comes a lot later. It’s like when kids drop out of high school—did nobody tell them that the fun jobs are the ones that pay really well and you’re not going to get one? And likewise with health, there are some things that are pretty straightforward. We had an experience in South Africa where we saw that awareness of AIDS was very low among teenagers, which is where the disease acquisition was taking place. So we said, let’s raise awareness. And we did. We were the biggest billboard buyer in the country, and we drove awareness up to 97 percent. We’d ask people “How do you get AIDS?” and they could explain it. “How do you stop yourself from getting AIDS?” and they could explain it. And we saw no behavior change. When we interviewed people, we found that the time horizon is such that a disease which is seven or eight years out when compared to the here and now activity is just not that big. If AIDS killed immediately it would be better, because you’d see these piles of bodies outside of bars or brothels and people would think “I don’t know if I want to go in there or not. It looks suspicious.”
Kristin: What about the one-off nature of mHealth projects? How do we break down silos?
Bill: Look for problems with gigantic scale, where several hundred thousand deaths a year could be avoided if it was solved well … things like vaccine coverage, TB drug follow-up, and TB diagnosis. The lens of did-it-work has to be a very tough one … because it’s possible to see things that worked in the microcosm but then, when you scale it up, it’s not going to work that well.
Kristin: So, we’re saving these people and letting them have a better quality of life, but is that really true? Are we just letting them live on to face harder challenges to come, particularly in slum areas like Mumbai, where population growth is actually outstripping the city’s population growth.
Bill: Most of the world’s population growth is in urban slums. The key thing in this question—the most important fact that people should know and make sure other people know—is that as you improve health and save children under five, that reduces population growth. That sounds paradoxical, but within a decade of improving health outcomes, parents decide to have fewer children. There is no such thing as a healthy, high-population growth country. No matter what problem you care about, environment, schools, nutrition, unrest, the key thing is that those problems are insoluble at a 3 percent per year population growth rate.
In these developing countries, by age five, over a third of the kids have had some insult in terms of cerebral malaria, birth asphyxiation, parasitemia, not enough nutrition because of diarrhea. [Consequently] their brain is not developed properly and so they’re never going to be a fully productive member of society. The same thing that gets rid of the death gets rid of the sickness, the morbidity, and so it’s a huge payoff.
Kristin: What’s the next big thing in technology?
Bill: Robots. If you don’t want to go to a convention, just send a robot. Seriously, though, there are a lot of medical procedures that can be done by robots. When we look at something like infant mortality, there’s a certain level you can’t get below if you can’t do C-sections. C-sections require a sterile environment, but the procedure itself is fairly routine and could be done by a robot.
Computers are learning to see, learning to talk, learning to listen, and learning to move around. This is where robotics is headed. The dexterity things are maybe five years behind, but they’ll come. And once a robot learns a task, it doesn’t forget how to do it. It can do it 24 hours a day.
As a founding sponsor of the mHealth Summit, Microsoft External Research is committed to using mobile technology to promote healthcare for underserved populations around the world, whether they are found in the jungles of the Brazil, the slums of Mumbai, or rural areas of the United States. Through our Cell Phone as a Platform for Healthcare Initiative, Microsoft External Research is collaborating with academics and researchers around the world, providing them with the tools, technologies, and resources needed to develop creative and novel healthcare solutions that are accessible, affordable, and easy to use, helping to mitigate the growing healthcare concerns in underdeveloped nations around the globe. Through collaborations with the academic research community, Microsoft Research aims to tackle healthcare problems through mobile technology and ultimately further the understanding and treatment of diseases affecting millions worldwide.
—Kristin M. Tolle, director of Natural User Interactions, the External Research division of Microsoft Research
Learn More
- mHealth Summit 2010 (opens in new tab)
- Video: Bill Gates and Kristin Tolle’s discussion (opens in new tab)
- Video: Kristin Tolle talks about mHealth (opens in new tab)
- Video: Interview with Kristin Tolle (opens in new tab)