Preparing for future pandemics requires us to reuse and share data quickly in order to repurpose that information for improved diagnostics and treatments, proposed data experts and scientists at a high-level discussion at the Geneva Health Forum Wednesday afternoon.

“It’s the value of what’s in data that can be reused, repurposed, to create new knowledge to provide insights into guidelines, and even lead towards better and improved treatments,” said Rob Terry, TDR, the Special Programme for Research and Training in Tropical Diseases.

The panel, ‘Data sharing in the time of COVID-19: What works and what we need’, focused on raising awareness of the lack of sharing of de-identified individual level data from COVID-19 clinical studies to sharing and identifying strategies for sharing data during the pandemic.  

All panelists noted that with sharing data globally, privacy and confidentiality still remains a concern, with many pointing to a need for data protection alongside data sharing.

“There’s a strong need to rationalize data protection laws and then contextualize them and make sure they’re not working against the public that they’re meant to be protecting,” said Naomi Waithira, Mahidol Oxford Research Unit Data Sharing Working Group, COVID-19 Clinical Research Coalition.

Three E’s of data sharing 

In an effort to maximize the benefits of reusing data, Terry brought up the three E’s of data sharing: efficient, ethical, and equitable.

“We must be equitable about sharing – when data is shared, it must not disadvantage those who are sharing the data,” Terry said.

He points to how researchers from low- and middle-income countries often find themselves exploited and exported for the benefit of others.

Data sharing also needs to be fair, so that it’s findable, accessible, and able to be used with other data sets. 

While there are hurdles to overcome with data sharing, shifting to a culture where data sharing is the norm is key, said Terry. 

“There are a lot of issues with data sharing, and they tend to be political, ethnical, technical, legal, and social. And we have solutions for many of those.”

“We need to create a culture where sharing is the norm. The benefits we’re gaining are so much better when the data is available and available for reuse rather than locked away in a drawer or in these days, a hard drive.”

‘Blanket’ privacy laws do more harm for data reuse 

However, while data privacy laws exist worldwide, to enable data sharing for scientific research, there needs to be more than just “blanket privacy regulation’, but rather regulation that addresses the actual risks of disclosure within its details,” Waithira proposed. 

In the past, there has been the General Data Protection Regulation (GDPR) in the European Union, the Protection of Personal Information Act in South Africa, the Personal Information Protection Law in China, and others that aim to ensure responsible data collection. 

 With data anonymization and de-identification from these laws used as ways to address confidentiality, this reduces the scientific usefulness of the data that could be used for datasets in the future. 

“When anonymizing data, you remove parameters which could potentially identify the person, but the complexity of data reuse [needs this information] because we actually do not know what in the future these datasets could be used for.” 

“We need to understand the data protection rules. What exactly is meant by the GDPR, what does it cover, what doesn’t it cover?” added Nathalie Strub-Wourgaft, of COVID-19 Response & Pandemic Preparedness, Drugs for Neglected Diseases initiative (DNDi). 

Waithira brings up a need to engage with the public so they fully understand their needs and what is actually considered risks with data sharing. 

“We need to have more engagement with the public to understand their concerns and what they perceive as risks over data sharing and reuse, and then use that data to rationalize data protection laws,” she said.  

Harmonizing and pooling data during times of peace 

Others addressed how data must be pooled and harmonized across the board in order to quickly combat future pandemics and the spread of  infectious diseases. 

“We need to have information quickly to make sense of the data as quickly as we can,” said Strub-Wourgaft.

Pooling data together, stated Strub-Wourgaft, allows for secondary research to take place, which is especially important for neglected tropical diseases (NTDs).

NTDs typically have little research, few guidelines and little data to work with. 

Using sleeping sickness as an example, DNDi’s consideration of data from diverse studies together allowed regulators to define a certain drug for treatment of sleeping sickness as a benchmark, she said. This also paved the way for other treatments to be developed and to further understand the severity of the illness.

“How do you make sense of clinical severity symptoms in relation to white blood count? You cannot do that without having more coherence by looking at all the data, putting them together and making a robust analysis,” she pointed out.  

Pooling data also provides researchers with more robust indicators about specific sub-population responses, as well as defining factors like: safety signals to consider in  treatment and baseline characteristics to predict severity progression of an illness, as has been the  case with COVID-19. 

Had this harmonization occurred before, we would have prevented many missed opportunities on the potential therapeutics of certain drugs, added Philippe Guerin of the Infectious Diseases Data Observatory.  

The infectious Diseases Data Observation (IDDO) is a platform managed by University of Oxford which is providing researchers with a quality-assured means of sharing data with other experts exploring the same disease or treatment – it’s the kind of platform that Guerin and other panelists say should be expanded now in preparation for the next pandemic – to enable more robust studies of treatments more rapidly. 

“We can’t organize things in times of crisis,”  Guerin said. “Things have to be organized and set up in times of peace. Without this organization in place, we will eventually fail again as we did for Ebola, as we did for COVID. We will fail again to have an environment where we can share and amalgamate data of value for a pandemic like COVID.” 

Existing data repositories such as IDDO were previously recommended as a way to support data sharing effectively, ethically, and equitably. (See related Health Policy Watch story)

Image Credits: Internet Archive Book Image/Flickr, GHF.