Why Twitter was so valuable for pandemic science
Reflecting on the positives among many negatives
Social media is a bit of a mess currently. Increasingly scattered people on increasingly scattered platforms. A stuttering Twitter exodus, if only there was a clear destination.
Yet despite its many deservedly-documented flaws, Twitter did have some real value for science during COVID. So as we wait for what – if anything – comes next, it’s worth taking a moment to look back at some of the positives. Because it would be a shame if these benefits disappear for good.
In a situation too fast for traditional journals, Twitter provided de facto peer review for pre-prints
Back in the old days, scientists would hear about preliminary work at conferences, then later see it in detail in peer-reviewed academic journals. The problem is that this process requires regular travel and lots of time, both of which are in limited supply during a pandemic.
There had already been a shift in biomedical research pre-COVID, with teams getting more work onto pre-print servers like bioRxiv and medRxiv before submitting to academic journals to undergo review. In a pandemic where sharing new information rapidly was crucial, pre-prints became all the more important. (In some cases, even pre-print servers were too slow; in early 2020 our team created a dedicated repository to make our work available instantly, rather than waiting several days for a pre-print server to process it.)
Having high quality research shared rapidly is a good thing. But pre-prints also mean low quality research can be shared rapidly. Among a deluge of new COVID research, how could we tell which results were more robust? For many, this is where Twitter came in. Not only was it a source for hearing about new research; it was possible to get crowd-sourced reviews in real-time from leading researchers. Often when a surprising or odd result came out, we’d only need to wait a matter of hours to get a nice accessible breakdown of the findings from multiple specialist sources. The process wasn’t perfect, but it was very useful overall.
In a world of reduced interactions, Twitter sparked collaborations
In June 2020, Rupert Beale at the Crick Institute shared a pre-print on the SAFER study, which had regularly tested healthcare workers at UCLH for SARS-CoV-2 by PCR, as well as measuring antibodies. At the time, we’d been wanting to know how easy it was to detect virus soon after infection, because this would be crucial for evaluating different testing and contact tracing strategies. I replied to Rupert’s Twitter post, and soon a collaborative effort had generated the estimates we needed.
It turned out we weren’t the only ones who needed such information. Our estimates would end up being reused by scientists working on the COVID response in a range of countries. That Twitter conversation would also be the start of an ongoing collaboration with the Crick – including, more recently, a new research consortium to examine heterogeneity in SARS-CoV-2 immunity globally.
In a world of new evidence, Twitter spurred creativity in communication
Pandemics are littered with counter-intuitive quirks, from non-linear dynamics to delayed outcomes. Effective science communication therefore becomes vital.
One common conceptual stumbling block was exponential growth, particularly in the early stages of a wave – or during the emergence of a new variant. In late 2020, as Alpha spread across Europe, I pointed out that because transmission had an exponential effect, but disease severity only a linear one, an increase in transmission could be a much bigger problem when dealing with a new variant. The accompanying (somewhat technical) thread would become my most retweeted post of the pandemic:
Thankfully the idea soon got into the hands of people who are more creative than me, which helped amplify the warning. It was shared widely, from Vox to The Atlantic; I particularly liked the visualiations it inspired in the below Instagram post by Mona Chalabi.
Twitter became the go-to venue for new COVID insights. In the UK alone, you could get breakdowns of variant analysis from UKHSA epidemiologists like Meaghan Kall, crucial data reporting from journalists like John Burn-Murdoch, and the benefits of having fresh eyes on the problem from statisticians like Oliver Johnson. (Not to mention the many, many others who provided valuable perspectives in real time.)
Where next?
I don’t know exactly. But I do hope that we can find a similar way in future to have a massive real-time review system for pre-prints. And a way for disparate research groups – and junior researchers especially – to build new collaborations outside of expensive, distant conferences. And a way to regularly hear about new ideas and research, with creative communication and visualisation.
Perhaps Substack (which lots of people I respect are moving towards) will help with some of this. Maybe Twitter will get its act together (although I’m increasingly a fan of being able to consolidate longer posts in a more accessible way). Or maybe an entirely new hub for wider scientific discussion will emerge.
Here’s hoping.
Good read, lets hope we can keep the benefits alive somehow. I deeply appreciated your work during the pandemic and often used your posts to inspire colleagues as well.