To fight COVID-19 we need data, says WHO's chief scientist

Only 4 out of 10 deaths worldwide are reported — and that can have a critical consequence on the response to the pandemic, said Dr. Soumya Swaminathan, chief scientist at the World Health Organization. Designating COVID-19 deaths as such has been a challenge, and the absence of data on deaths exacerbates the problem.

“Our own estimates are that the true death toll [of COVID-19] is at least two to three times higher,” Swaminathan told Devex.

WHO is planning a data governance summit for the end of June that will get all the member states and stakeholders to discuss the importance of collecting high-quality data to record deaths and also “to share it, use it, be transparent about it, and WHO would like to play a role in that area,” she said.

Swaminathan, a pediatrician from India and a researcher on tuberculosis and HIV, is WHO’s first chief scientist and leads the newly created science division. She started in her role a few months before the pandemic hit, and while she started working on the division’s priorities in March 2019, “each of them was tested by the COVID pandemic,” she said.

In a conversation with Devex, Swaminathan laid out the work that still needs to be done in the months to come in the areas of vaccine equity; new research and development on vaccines and therapeutics; data transparency across countries; and the challenge of tackling new variants of the coronavirus.

This conversation has been edited for length and clarity.

From your vantage point, what does vaccine equity look like on the ground, and what needs to happen in the next few months to achieve that?

Our first priority is really to protect those who are the most vulnerable. We've seen now in the last 18 months, we’ve lost an incredible amount of people: 3-and-a-half million people officially. But our own estimates are that the true death toll is at least two to three times higher, so you can imagine the amount of impact it has in terms of lives lost.

We knew from the beginning that we would have limited vaccine stocks and supplies.

The right thing to do now is to be able to share the vaccine doses more equitably, and then I think in the second half of 2021 and the early part of 2022 will have lots of increased supplies to start coverage very, very rapidly.

It's only now emerging what the actual impact of the pandemic is going to be in terms of deaths. What needs to be done at a country level to strengthen this system of data collection?

So it's interesting that you asked that question because data is the other key area that WHO transformation is aiming to strengthen, and [WHO Director-General Dr. Tedros Adhanom Ghebreyesus] when he announced the science division also announced the new division of data delivery and analytics.

And the idea of that really is to address the issues and the gaps that you're seeing. Only something like 4 out of 10 deaths worldwide are even reported and often only a small proportion have the cause of death that's described properly — which means that many countries do not have any idea of how many people are dying and what they are dying of [due to poor data collection mechanisms]. It's what we call vital statistics.

You need to know what those people are dying of — whether it’s infectious diseases, noncommunicable diseases, suicides, road traffic accidents — unless you know that, a country's health system cannot plan properly for the future.

So a huge amount of emphasis now needs to be placed on strengthening the CRVS, which is the civil registration and vital statistics system in every country. In some countries, we have siloed data systems that do not talk to each other, and therefore they cannot be really merged to answer the kind of questions that you want.

So one thing is collecting the data and … that data then also needs to be analyzed and used appropriately for policymaking, and there needs to be transparency around that.

And what I also feel strongly [abou] is that this data needs to go back to the people who actually helped collect it in the first place, in order to help them to inform them, and improve their performance. And what happens is data flows usually one way: it goes up up up.

You need to make sure that the data that's being collected is being used to inform policies and programs, which are then transmitted back to the people at the ground level so that they can improve.

So I think there's a lot of work that many countries have to do. The whole digital ecosystem has got a big push from the pandemic, this is the time and for countries that do not have good health data systems, this is the time to really develop a very well thought out integrated, data system and data platform which doesn't silo but rather brings together different data streams to make it easy for policymakers.

At WHO we’re planning a data governance summit towards the end of June in order to get all stakeholders and member states but also others to come together and talk about why it's so important to not only collect high-quality data, but then to make it, share it, use it, be transparent about it, and WHO would like to play a role in that area.

Peru recently revised its COVID-19 death count, and now it's three times what they had previously announced. So is that something that is part of your work and part of WHO’s work — to encourage that culture of transparency?

Yes, absolutely that's exactly what we want to encourage because you see it's in the country's own interest to have credible data and make it transparent, because only then can you act on it.

So the first thing is to really accept the ground situation and the challenge. It could be good news or bad news, but one has to accept it. And then you act on it.

It's not a comparison between countries — it's not something that should be really used for ranking or for making countries feel good or bad about the data. It's for every country to really use their own data to improve their health systems.

I think there is a lot to be said, now, in this day and age for pooling data to get much more out of it, than you could otherwise. We know that artificial intelligence algorithms work well when you have big large data sets, which are very representative and heterogeneous.

So you could imagine that if you pool data from many, many countries together and you wanted to ask some questions about a particular disease or a particular risk factor, it would work much better if you had those kinds of large, representative, heterogeneous data sets. So, in order to harness the new tools that we have like artificial intelligence and big data analytics, we also need to get into the culture of sharing and working collaboratively.

What can we expect from WHO, in terms of new research in COVID-19 vaccines and therapeutics going forward?

Therapeutics is an area where we have not been as successful as vaccines because there have not been too many antivirals that have been developed. There are now some in development, and those need to go into large phase three trials, and hopefully, we can have a drug that's [a] good antiviral that can be given early in the course of infection to prevent the disease from progressing.

We do have some monoclonal antibody combinations now that can be given again early in the course of [the] disease, particularly to those who have some risk factors. If these monoclonal antibodies are given, it's been shown that they can reduce progression to severe disease and the need for hospitalization.

But, again, we need more studies coming from low-resource settings and then of course, we know that dexamethasone is helpful for severe and [in] critical illness antithrombotics are helpful. Oxygen, you know, has proven to be the essential commodity, which every country in the world has found that there is a lack of when there is a surge.

On vaccines, I think we have a huge number of vaccines in development. We have some very interesting and exciting candidates like nasal vaccines, oral vaccines, [and] single-dose vaccines. So there's a lot of questions also about how vaccines can best be used, will boosters be needed, can you mix and match vaccines, vaccines and children, the elderly.

Lots of questions still to be answered, despite the fact that we already have so many successful candidates.

As countries such as India prepare for a third wave, what kind of planning should be done going forward?

This is a kind of difficult situation that many countries have [been] faced with, not just India, but even very high-income countries have faced the same situations where some decisions were made, it turned out the health systems got overwhelmed and so it's taken everyone by surprise. I think now is a time when we know enough about this virus and how it behaves and also what measures work — public health measures, prevention measures, and treatments.

We know much more about this than we knew last year, in April or May  — so I think the right thing to do now is to really have policies that are based on the available information and evidence and have very clear strategies for how you respond to changes in epidemiology.

Again we've seen that this is constantly in flux: there are waves, it goes up, comes down. In India, for example, we saw for many months this virus was circulating at very low levels. And when many other countries were having their second and third waves, India was relatively very quiet. But then again the second wave came, all of a sudden, and then it was associated with this variant.

So I think we know now that sequencing is very important as part of surveillance. You need the traditional methods of surveillance also. That is where you're monitoring the prevalence of the SARS-CoV-2 virus amongst patients with pneumonia and influenza-like illness, but you have to add to that now genomic sequencing.

We need to keep a very close watch on what's happening at the state level. It's not enough to just have aggregated data, but we have to look at [the] district, or even below that level to see how many cases are being diagnosed, how many tests are being performed per million population, what's the test positivity rate what, what does the bed occupancy look like, especially for oxygen and critical care, and this needs to be tracked very closely, on a week by week basis.

So the moment you start seeing trends going in the wrong direction you take early action to stop this from becoming out of control.