The Power of Geospatial Data In Developing Countries

Dr. Victoria M Gammino (Left) Cameron Boozarjomehri (Right)

Interviewer: Cameron Boozarjomehri

Welcome to the latest installment of the Knowledge-Driven Podcast. In this series, Software Systems Engineer Cameron Boozarjomehri interviews technical leaders at MITRE who have made knowledge sharing and collaboration an integral part of their practice.

Many countries in the Global South are not fortunate enough to have the infrastructure or tools that we take for granted. Things as simple as knowing where a community is can mean life or death when battling disease. Fortunately, Dr. Victoria M Gammino is on the case. Join us as she walks us through her work turning geospatial data into the tools and technology needed to keep the world healthy and disease free. (Links to organizations and tools appear in the transcript below.)

Click below to listen to podcast:

Podcast Transcript
Cameron:	00:14	Hello everyone, and welcome to MITRE’s Knowledge-Driven Podcast, a show where I, your host, Cameron Boozarjomehri, have the incredible fortune of interviewing brilliant minds across MITRE. Today, I’ll be talking to Dr. Victoria Gammino, a person who wears many hats, but her current role is Senior Principal Global Health Equity. Victoria, would you like to tell us a little about your background and what you’ve been doing at MITRE?
Victoria:	00:37	Sure. I’ve always actually been interested in global health. I studied medical anthropology in college, and then I went into public health after living overseas for a few years after that, and after I finished my doctorate, I went to CDC and I was an EIS officer, the Epidemic Intelligence Service. It’s most commonly known as sort of the Disease Detective Corps. And I stayed at CDC for about 14 and a half years, and while I was at CDC, I worked on a number of issues, drug-resistant TB. I worked in TB, HIV, and then I got to vaccine-preventable diseases, which was a completely different kettle of fish, and that was really what got me into and very interested in health geography and geospatial data.
Cameron:	01:24	That’s really cool to hear, and I think that’s actually a great segue into what you’re currently working on, and I know this is also a mouthful: leveraging geospatial data to improve emergency response and preparedness in the Global South. I’m actually super jazzed about this topic, but, obviously, Victoria, would you actually like to get us off and running?
Victoria:	01:40	Sure. So I’m actually… I’m working on a number of public health projects that relate to the work I did at CDC, and in a position since CDC in the last few years before joining MITRE, although there’s one I’m super excited about that focuses on geospatial data infrastructure. Specifically, working on a machine learning algorithm to identify health facilities and, eventually, we’re hoping to identify other types of infrastructure in developing countries. And that project is one that is related to work that has been going on for quite a while that’s been funded by the Gates Foundation. They funded Columbia University, a project called the GRID3 project, and Dr. Marc Levy and his team run that project. And, essentially, they have been creating data layers for a number of developing countries that would facilitate the implementation of a lot of the Gates programs there, including vaccination programs, maternal and child health programs, and, for example, malaria programs.
Cameron:	02:45	That sounds like a very invaluable project. That infrastructure is critical to anyone who wants to be able to then take that data or take what’s already going on there, and then using it to track the specific outcomes and what needs to be done, or what’s already being done well.
Victoria:	03:00	That’s exactly right, that’s exactly right. I mean, the core issue is that basic geospatial reference data, like for example, geolocated, labeled infrastructure, like hospitals and schools, or administrative boundaries. An administrative boundary, for example, is literally the boundary for a nation, or a province within a nation, or like in the United States, states and counties—that’s an administrative boundary. So that type of infrastructure, so administrative boundaries, census estimates are another one. So census estimates within those boundaries are commonly out of date in lower- and middle-income countries, predominantly lower-, but also some middle-income countries, and, because of that, these data gaps pose a significant challenge to public health activities.
Victoria:	03:47	For example, in epidemiology, we talk about rates. What is the incident rate of, say, COVID or malaria? We talk about mortality rates per 100,000 or 10,000. These are generally bounded by the geography. So it’s the country, state or county, so the rate per 100,000 in this county. So using a standardized rate allows us to compare across these different geographic regions or units with different population sizes. If, however, the census data aren’t current, then the rates are going to reflect those errors, or, for example, if the boundaries of a province, or a county in a country change because of war or because of political redistricting, and the census estimates aren’t adjusted, this creates errors in the rates.
Victoria:	04:37	There was a really good article about a year and a half or two years ago by Ed Yong in The Atlantic Monthly, and he wrote this great article about how the out-of-date administrative boundaries in the Democratic Republic of Congo, as well as the out-of-date census data, really fouled up the estimation of the rates of Ebola, and being able to track the outbreak effectively in the DRC. It’s really well-written. I recommend it to anybody who’s interested in this topic.
Victoria:	05:07	Anyway, I worked on an outbreak response a few years ago in Sub-Saharan Africa, in a country recovering from a number of years of civil conflict. It wasn’t quite a civil war, but it was a civil conflict, and when we went, it was a vaccine-preventable disease, and the country allocated vaccine to stem the outbreak based on the last census, which had been more than 10 years ago. And later, they realized that when they went out to vaccinate, after the vaccination campaign, they did a survey and they missed about 17,000 children in one province.
Cameron:	05:40	I think that really sets up why this is such a significant problem. Imagine missing that many people in just about anything, let alone just kids—that has implications going on for years for their entire lives. And I think at the risk of this sounding trivial compared to all the incredible information you just gave us, I do want to ask, are there other ways that you’ve found this work being particularly necessary? Obviously, the work that you just described is very beneficial in a health sense, but it sounds like because of this infrastructure, it has a lot of implications beyond just health.
Victoria:	06:13	Yeah. I mean, there’s many aspects. I mean, so for example, the data also play a significant role in routine supply chain planning and disaster response, and, obviously, there’s a health component to disaster response, but if we talk about humanitarian emergencies or natural disasters writ large, having accurate and detailed location data for hospitals and clinics, or, for example, even the roads or transportation routes to reach those, is essential, but it’s not always available. So a country may have what’s called a Master Facility List, which is the names and approximate locations of facilities. However, the precise latitude and longitude might be missing.
Victoria:	06:53	So in what I call the Global North, but basically in the developed world, Google and Apple Maps, Waze, etc, are so much part of the digital air that we breathe, that we take their availability for granted, especially in economies that don’t offer the advertising revenue to Google and Apple to invest in location data. I mean, we see it and use it so often we don’t even think about it, but I think what’s really important to understand is that those data are generated for a profit motive in the United States and in the developed world. I mean, those data are sold to advertisers or businesses who want to direct traffic physically to their locations, whereas in the Global South, you don’t have that profit motive. And so, there’s no incentive to collect, and record, and share those data, so they’re missing. And so essentially in developing countries, crowdsourced data services like OpenStreetMap, are often far more detailed and complete than their dot.com counterparts.
Cameron:	07:58	Yeah. I can see how weighing the commercial benefits versus the just social, and I guess human benefits is something that when we typically think of these problems where you’re expecting, “Oh, this big tech company,” or “This is going to be a government specific issue,” but even governments can be unreliable in the areas you’re describing, and as a result, it does become a community effort in a way, to get this kind of important data out there because obviously, the communities, I’d assume, would be invested in the fact that the people who are going to be leveraging this data, are bringing them some sort of aid. Am I off base with that?
Victoria:	08:34	No, not at all. I mean, that’s a good observation. I think, to some degree, there is the calculus of community investment or country investment in humanitarian aid, but I think more importantly, the bigger incentive really is getting businesses and the private sector in developing countries to recognize the value and the utility of geospatially referenced data for their companies. Speaking of which, there’s an entire podcast dedicated to this exact topic that talks about every single new company coming out, or the new projects and programs coming out in the private sector about geospatial data, and I think the Gates Foundation is moving in this direction. I think that if we can demonstrate the value proposition of these data beyond the humanitarian sphere, then countries and the private sector will recognize the importance of that investment and will continue to maintain the data infrastructure.
Victoria:	09:34	To some degree, there is data infrastructure in Africa; however, one of the bigger problems is that it’s outdated. And so, that lack of continued investment in maintaining the infrastructure is something that can slip through the cracks when you go from administration to administration. So if one government sees the value of it, and then the next government that comes in has different priorities, then it may not be maintained, and so, that’s when it falls back on the humanitarian sector to maintain it. And I think, though, that if we can demonstrate the value proposition of that to the private sector, you’re going to see more investment from the government, they’re going to recognize the value because then if businesses are doing better, it’s going to generate tax revenue. So it’s an entire system, and I think really, this is about jump-starting a mini economic system around this in the developing world.
Cameron:	10:32	Yeah, it sounds like a very powerful resource, especially considering how turbulent things can get, and I imagine there’s plenty of reasons why anyone would want to work on this work, but I was curious what initially drew you to it. What was the thing that set off this curiosity, this stream that got you to working on this work and making this impact?
Victoria:	10:49	So I worked with the Global Polio Eradication Program for about 10 years when I was at CDC, and CDC, along with WHO, supports the, it’s called the GPEI, and it’s complicated, but it’s a very sophisticated program, it’s been going on for many, many years, and we’re almost at the finish line in terms of eradication, but it’s a bit complicated. And so, the way it works is polio vaccine is distributed by vaccinators that walk from house to house. So, in these very large countries where they don’t have the kind of hospitals, and clinic, and private sector medical infrastructure that we have, they rely on vaccinators to literally carry the vaccine in a container from house to house, and then they put two drops of vaccine in a child’s mouth using a dropper, and then they go on their way. So when a child is vaccinated, they mark the child’s pinky finger with a temporary purple dye, because they don’t necessarily have vaccination cards that they can update at every single house, and then after the campaign, they evaluate the coverage using a statistical sampling technique.
Victoria:	12:04	So all of the distribution logistics, from receipt, to storage, to the actual walking routes for each of the vaccination teams, are laid out in what’s called a microplan, it’s like a 20-page document. And in our post-campaign surveys, we saw that we were missing children over, and over, and over again; we called them chronically missed. And when we revisited the microplans and compared what then were hand-drawn maps, and so they literally were only using hand-drawn maps of their neighborhood, or their county, or their ward. When we compared those to satellite imagery, we realized that we were missing entire neighborhoods in the urban areas, and that in the rural areas, we were missing entire hamlets or very small villages. We realized that it was because they didn’t have decent maps, and also because nobody really knew how to read the maps and navigate in the way that we think of it.
Victoria:	12:57	So what we tried to do was create georeferenced maps for them, and we realized at that juncture, the degree to which the digital infrastructure to do that was really lacking. And the Gates Foundation picked up the torch from CDC, we did the very first study and published the first study on this issue, but the Gates Foundation and WHO picked up the torch, and then they went on to make this a core tenet of the Polio Eradication program to build up that infrastructure in the remaining countries. And I developed a real passion for it, and it’s really become my raison d’être, and actually one of the reasons that I came to MITRE was because of its geospatial capabilities, and it’s recognized as a powerhouse in this space. And so, I really wanted to leverage that, to continue doing this type of work.
Cameron:	13:48	I think that’s a really amazing, admirable thing. It’s nice to hear not only that we have such talented folks among us, but that we’re attracting people who are working on what seem to be real passion projects. It’s a significant problem to solve, but not everyone sees, I guess, the value in doing it. There’s not a clear commercial benefit attached to it.
Victoria:	14:06	Yeah, I was surprised and really thrilled at the number of people who have reached out to volunteer their expertise, or to inquire about how they could help with this. I mean, I think you said it earlier, it’s not exactly a commercial enterprise. Humanitarian work is never really a big revenue generator, but it’s, nonetheless, very impactful, and again, that’s something about MITRE that I really appreciate, is there’s more of a focus on impact than there is necessarily on revenue. And so, I think that having that excitement among the people who have been doing this for years, to me, just is really inspiring and makes me want to do more of this work.
Cameron:	14:53	I’m really glad to hear that. Well, unfortunately, I think we’re a little out of time. Thank you so much, Victoria, for talking to us about this incredibly important problem. I’d like to obviously extend a thank you to MITRE and the Knowledge Driven Enterprise for making this podcast possible. And again, just Victoria, thank you so much. I realize that this is something that we don’t really get to hear about much, and I’m really glad that such talented people are working on it.
Victoria:	15:16	Thank you, Cameron. I’m really excited to have had the opportunity to talk about it, and hoping that if anyone’s interested in this, they’ll reach out, so thanks a lot.

Cameron Boozarjomehri is a Software Engineer and a member of MITRE’s Privacy Capability. His passion is exploring the applications and implications of emerging technologies and finding new ways to make those technologies accessible to the public.

MITRE’s mission-driven team is dedicated to solving problems for a safer world. Learn more about MITRE.

An Introduction to Interoperability in Healthcare

Framing Ethical Standards for Consumer Data Use in Health Care

Project Demodocus: Bringing Accessibility to the Masses

MITRE Innovation Program Team Hosts Design Studio on Veteran Mental Health with MITRE Veterans

Jen Choi and Josh LeFevre and the power of “Yes, And”

Turkeys vs Swans, with Imanuel Portalatin

Justin Brunelle: Lessons from MITRE’s Innovation Program

Theodore Wilson: Thinking Like a Turtle

Dan Frisk and Paula Randall on Bringing Innovation to Government

Interview with Julie McEwen on why Privacy is Key

Interview with Awais Sheikh on Deciphering Business Process Innovation