New tools in the fight against COVID-19: Sema Sgaier talks data modelling around global vaccinations
Interview with Dr. Sema Sgaier, Co-Founder and CEO Surgo Ventures
Originally I was a neuroscientist, but I got into global health about 15 years ago, working on pandemics. I started with the HIV/AIDS pandemic in India, then in southern and eastern Africa. I’m just driven by the power of data and working with communities to make sure people have access to health. Data reveals patterns and helps you zoom in on what matters. Communities hold the ground intelligence that have the power to turn these insights into change. I find working in pandemics extremely exciting. It’s really the challenge of our times. I co-founded Surgo Ventures with Mala Gaonkar — we’re a non-profit, based in the US and UK, but really we’re a global organisation, and essentially our mission is to leverage behavioral science, data science, and AI to bring greater precision to how public health and development programs are designed.
When the pandemic started, I think — like everyone else — we were thinking this was just going to be a few weeks, or maybe a month or so. But then also, as an organisation that’s been working in global health for a number of years on lots of different issues, we felt an obligation to do something about COVID-19. As an organisation, we took a break for two weeks and said, “What can we do? How can we add value?” And where we started was in the United States, by building this very granular COVID-19 Community Vulnerability Index which could tell policymakers and people on the front lines which areas were going to be impacted the most. That Index is now being used by a number of different organisations, and was the start of the journey which led to us building many, many different data tools that provide many different insights, helping organisations at the front line.
Essentially, what we’re trying to do here is help policymakers and organisations use their limited resources to tackle the COVID-19 pandemic. What that means is answering questions like: What are the geographies, countries, or areas within a country that they should be focusing on? And what should they be doing in those areas to ensure that the health, economic, and social impacts of COVID-19 are mitigated as much as possible?
On a day-to-day basis, what we’re doing is leveraging as much data as possible to be able to give as much granular information as possible to policymakers. We tell them where the pandemic is moving, where it’s going to go next, and what it’s going to do once it gets there. Once it reaches that city or region, is it going to be leading to a lot of deaths — more than expected? Is it going to have devastating economic outcomes? Is it going to lead to systemic fragility and insecurity? We’re using data so that policymakers can be a step ahead of the virus, rather than trying to catch up with it, on an ongoing basis.
We took that tool and built the Africa COVID-19 Community Vulnerability Index, which assesses 751 regions in Africa across 48 countries. For each of those regions it can answer key questions: How are they going to fare? What’s their resilience when it comes to COVID-19? You can go within a country, look at regions within that country, and ask, “Is region A going to experience a much more devastating impact than region B?” And as a tool, it’s been used in many different ways.
In Nigeria, for example, the National Bureau of Statistics has incorporated our Index into their COVID-19 data hub — and so it’s now being made available to an even wider range of organizations when it comes to figuring out how they should be using their limited resources in the COVID-19 response. If an organisation wants to offer food assistance, they know which regions are most vulnerable to the impact of COVID-19 when it comes to food resilience, and prioritise those efforts. If the government wants to introduce policies like social distancing — should that be all across Nigeria, or should it be in only a few regions? This tool allows you to be really precise in modelling your response.
One surprise in Nigeria was that, when the Index showed us which communities were more vulnerable than others, the communities that were more vulnerable as per the Index weren’t actually the communities first hit by COVID-19. That makes a lot of sense in retrospect — those were the urban, densely populated capital cities of Africa, which had a lot of incoming people because they’re connected to the rest of the world. But what’s interesting about that is that areas that have dense populations aren’t the same as the areas that have underlying epidemiological factors — where people also have chronic diseases at higher levels than other areas. So while those urban regions were getting impacted first, the most vulnerable communities are the ones that get hit later, because they’re more fragile, with weaker health systems. The key thing we discovered is that, while we can get stuck on where we’re being impacted the most, we need to also look into the future. What we’ve seen from the US, UK, and Europe is that, ultimately, every area gets impacted by COVID-19. That was surprising to us.
The Index is also being used right now to develop a system around vaccine allocation. Nigeria just received its first doses of the COVID-19 vaccine, and one of the big questions is how those vaccines should be distributed across the country. So, with the COVID-19 Vulnerability Index and additional data, we’re trying to build a tool that helps governments operationalise their vaccine rollout, and figure out where they should prioritise the first batches of the vaccine. And as part of the upcoming Vaccine Data CoLab, we’re excited to actually leverage the tools that we build through the vaccine rollout across Africa. Africa is going to have only limited doses of the COVID-19 vaccine — initially, it’s only about 20 percent of the population that will be covered — and so the question of how those doses should be prioritized is key:which groups, which kinds of people within the country, andalso which regions. We’re going to leverage another tool we built for the United States — a vaccine allocation planner — so government can know the answer to, “OK, how many healthcare workers do I have in the different regions of my country, and I need to prioritise one region over another, what criteria should I be using?”
We were really fortunate to get support through the COVIDaction data challenge, which was originally funded by DFID. It’s been incredible in not only giving us monetary support to be able to build this tool and distribute it, but also really link us to various partners and organisations across the continent so that this tool can get into the hands of the right people, to actually use it for their pandemic response.
We built this tool with a long-term view. The initial impact of COVID-19 is on the health side, but ultimately COVID-19 is actually going to have a much larger (and longer-term) impact on the economic and social aspects of community life. So this tool is really useful, firstly, in terms of thinking ahead of time which communities are going to have worse economic outcomes, and how to plan for that. The second important point is that we also built this tool in a way that it can also be used for future pandemics. We built it with a modular structure, so essentially you can pull out modules and replace them with other modules that are specific to the nature of those future pandemics. It’s built so that it’s available immediately for COVID-21, or another SARS, or some other pandemic. Hopefully not — but that is the plan.
One of the most positive aspects of the pandemic is how countries in Africa and Asia — so-called “low resource” countries — were able to respond better than some of the higher-income countries. And to me, that really says a lot about the decades of work that these countries have been doing in pandemic response, and global health in general, in building their public health infrastructure. A lot of innovation has come out of these countries, which is encouraging.
And in terms of our partnership with DFID, and COVIDaction, it’s been so great — and we see it growing, not only in terms of the tool we built, but in other tools that we’re building together, both for COVID-19 vaccines and the wider future. But most important is the support that they’ve given us in connecting us with other organisations, and really expanding the network that can come together on this. We do this better as a community — not as a single organisation on our own.
