Belgian researchers chose to bypass all standard science publishing protocols to publish research that has been overhyped and isn’t well understood.
In the last 24 hours, a computer simulation by a team of Belgian engineers that tracks the “spread droplets” and “slipstream” of the exhalations, coughs, and sneezes of people who are running, walking, or cycling has gone viral. Perhaps you have seen this gif on Twitter, Facebook, or NextDoor. Or, as some people on our staff have seen, perhaps write-ups of it have been texted to you by concerned friends or family:
Though this was not the specific goal of the simulation, it is currently being used on neighborhood groups and social media as scientific evidence that people who are jogging and biking are putting others at risk. If you are getting “droplets” or “globules” on you, the thinking goes, you are at risk of contracting coronavirus.
“People should read and not misread my tweets and texts,” Bert Blocken of Eindhoven University of Technology, the lead researcher on the simulation, wrote in an email to Motherboard. “I have never and nowhere discouraged people from walking, running, or cycling. Rather the opposite. Maybe people should read more, and react less.”
Blocken has yet to publish a peer-reviewed paper about the simulation. In fact, he hasn’t even published a non-peer-reviewed study. Instead, he spoke to a reporter in Belgium about it, who wrote a news article, which has now been aggregated and shared widely by many publications. Given what Blocken has put into the world, taken at face value, some people are understandably concluding that it is impossible to run or cycle safely in many cities; he recommends a distance of 65 feet between bikers and other people, something that is impossible to do in cities. The issue with Blocken’s suggestion that we “read more, and react less” is that there is almost nothing to read, and there is no study to critique.
Blocken’s team took the extraordinary step of speaking to the media about his research before publishing anything about it. There is no written study to read or interpret. We do not know the specifics about how the study was done or how the simulation was run because the research team has not shared that information.
On Twitter, Blocken said the “crisis is urgent, so exceptionally we turned order upside down: (1) media, (2) today I submitted the proposal for funding (3) peer review article later. Public cannot wait months for peer review. I have a short text, I will post it on Linked In within the next hour.”
A day later, that LinkedIn post has not been published. What the team has published is something that it’s calling a “white paper,” but which is actually a Google-translated version of the Belgian newspaper article that was not written by Blocken or his team, but which quotes him. Ansys, the company that did the simulation in concert with Blocken, has also published a short but vague press release. In the meantime, this simulation has gone viral.
Studies like this are “not really useful. Not to epidemiologists anyway. The amount of transmission from this route even if it is possible will be dwarfed by that from others.”
A Medium post written by Jurgen Thoelen, who describes himself as an “entrepreneur, building clouds in all forms and shapes and life-long athlete” has been shared thousands of times. On Facebook and Twitter, the article is being shared in neighborhood groups and is being used to spur a battle between pedestrians and runners and cyclists. A typical comment is something like this, shared in an Iowa City “Quarantine Survival” Facebook page: “Omg people keep doing this. Runners and bikers with zero regard for fellow pedestrians 🤬🤬” The simulation has also been written up by the Daily Mail, while gifs, stills, and memes of the simulation, shared with little or no context, have spread on their own.
This is all to say that we are unsure of the specifics of this study, what it actually shows, what its limitations might be, and how it was done. What it’s suggesting could be accurate and useful, but we have no way of knowing that at the moment. And yet, this research is already being used to ask people to change their behavior and held up as definitive evidence that running and cycling are irresponsible during the pandemic.
Blocken said in an email that this was not his intention.
“Choice made in agreement with all the researchers involved and both university media agencies. The crisis is worldwide and the situation is urgent,” Blocken wrote. “We did not want to keep results behind closed doors until we have found the time to write down the full story. If I would have done the opposite, we would receive criticism about that. Never possible to do right for everybody. Given all the fuss I notice now, I will do an extra other late night effort and post the full story on Linked In tonight.”
“By the way this is aerodynamics work, not virology. Good luck with speeding up procedures in engineering journals,” Blocken added. “COVID-19 will not wait months or until our paper is published.”
Blocken is right: We face an urgent situation, and it’s important to get rigorous science out as quickly as possible. But hundreds of other scientists have managed to publish peer reviewed research about coronavirus in the last few weeks, on an expedited time scale. Thousands of others have published studies that are not peer reviewed, but that are at least studies in the way we usually think of them: Their methods and findings are explained in a rigorous way that can be critiqued. Even though this is a dire situation, scientific publishing safeguards exist for a reason, and we’ve already seen during this pandemic that a rushed process has led to bad, inaccurate science being published (that’s not to suggest that Blocken’s research is bad or inaccurate science, we simply have no way of knowing based on what’s been published.)
Even if the simulations hold water and are accurate, virologists and experts should be the ones making public health recommendations, not random “entrepreneurs” on Medium, which is what has happened because these simulations were not published with the specifics of how they were done or what they mean. This type of research is of course important and should be done, but it should be released in a responsible way, with the caveats, limitations, and unknowns explained clearly. Then the research should be used by virologists and public health officials to make concrete recommendations to people.
I showed Blocken’s research to William Hanage, an epidemiologist at Harvard’s Center for Communicable Disease Dynamics. He said that the virality of Blocken’s research is harmful, and that Blocken’s suggestion in the white paper that this research is a “modest contribution” toward the fight against Covid-19 “makes my blood boil.”
“Where the droplets are is much less relevant than the amount of transmission that occurs via this route”
Crucially, scientists are still unsure how well the coronavirus spreads in the air, and many have cautiously speculated that the overall risk of transmission appears to be less outdoors. Globules and droplets do likely carry the virus, but that doesn’t mean that anyone who gets a droplet on them from someone’s breath is going to be infected. Transmission depends on a host of factors; scientists believe an important one of these is “viral load,” which is a measure of how much of the virus is present.
“On the epidemiology side—where the droplets are is much less relevant than the amount of transmission that occurs via this route,” Hanage said. “Advice on physical distancing is really about *reducing* the risk of transmission rather than eliminating it altogether.”
He said studies like this are “not really useful. Not to epidemiologists anyway. The amount of transmission from this route even if it is possible will be dwarfed by that from others.” He added “it’s concerning” how fast the study has traveled … especially “when you consider I have had to write this email rather than putting the finishing touches to a model of nosocomial transmission [in hospitals].”
In a footnote on the white paper, Blocken admits “currently the subject of intensive debates between scientists world‐wide—is to what extent the residue of micro‐droplets with the virus, after evaporation, still carries an infection risk. Further virology research should shed more light on this issue.” Last week, the Atlantic‘s Ed Yong spoke to many virologists about this, and there currently is no consensus about how dangerous it is to exercise or be outside, but there is much research suggesting that the mental health benefits of exercising outdoors are important and should be taken seriously.
The issue of viral load and transmission is not addressed or mentioned in the Medium article nor in the Belgian newspaper Blocken spoke to.
When I asked if he was concerned about the fact his work had gone viral, especially in write-ups by non-experts, he said, “I am surprised by this question. You with your expertise should know that one can control the first line of media attention, and then people write stories of the stories, and it is impossible to control,” Blocken wrote. “That would have happened equally if the full paper had already been published. This is not my first big media coverage, so I have been there, done that. There is free press.”
Hanage said it’s probably OK to exercise outside as long as you “apply common sense.”
“I think there is a balance,” he said, “other than in the situation where there are really really high rates of community transmission.”