Welcome to Impact Factor, your weekly dose of commentary on a new medical study. I'm Dr F. Perry Wilson of the Yale School of Medicine.
Cast your mind back to March 9, 2020.
You'd heard of this novel coronavirus, but with less than 1000 reported cases in the US, it still seemed remote. Schools and businesses were still open.
On March 9, then-President Trump tweeted:
So last year, 37,000 Americans died from the common Flu. It averages between 27,000 and 70,000 per year. Nothing is shut down, life & the economy go on. At this moment, there are 546 confirmed cases of CoronaVirus, with 22 deaths. Think about that!
I'm not referencing that tweet to dunk on the former POTUS; a lot of people had a lot of bad predictions about COVID in March of 2020.
But I want us to remember that air of uncertainty we had back then. By early March, COVID had been circulating in the US for at least 3 months and possibly more, and yet we were only starting to understand the implications. The World Health Organization (WHO) would only declare COVID-19 a pandemic on March 11, 2020.
Could we have known what was going to happen earlier? Was there a hint that something was going on, long before the press conferences started? A new paper from PLOS Medicine suggests that there was — and it's something that might help us during the next pandemic.
The PLOS paper makes the suggestion that the key to predicting the emergence of a new virus was under our nose the entire time — or maybe in our nose.
This study depends on an ongoing program called FluNet — managed by the WHO — which collates information on testing for influenza-like illnesses across dozens of countries around the world.
The data are pretty broad. Anyone with fever, who gets tested for the flu, has their data entered in FluNet, including the results of that flu test.
That means that sitting there in FluNet are a bunch of people with fever, respiratory symptoms, and yet a negative flu test.
The authors posited that as the coronavirus pandemic hit, there would be a spike in influenza-negative influenza-like illnesses, which might anticipate the recognition of COVID itself.
I think it makes more sense to look at a graph.
What you see here is the count of people in the US who had a fever, got tested for flu, but the flu test itself was negative over time. These spikes then either represent other infections, likely other respiratory viruses, or some inherent false-negative rate of the flu test itself.
Those peaks are seasonal, paralleling flu season, perhaps because we rarely test for flu out of season anyway, or because some respiratory viruses prefer a colder environment.
Through visual inspection, you can get the sense that some peaks are higher than others. In 2018, for example, you see a peak that is particularly prominent; you may recall that was a particularly bad flu season. You see a similar big peak in January of 2020, just as COVID-19 hit our shores.
The key insight the study authors had was to use a statistical tool to automate the process of finding those big peaks, a technique called outlier detection. That means you don't need some subjective human like me to squint at a graph and try to decide if the current peak is just a variation within the realm of chance or something more significant. In fact, those red diamonds were the outliers as detected by the algorithm. In theory, you could have this algorithm running all the time as data come in, and it could potentially sound the alarm that there is likely a new non-flu respiratory virus circulating out there.
Would it have made a difference if we had used this before the coronavirus pandemic? It might have.
Out of 28 countries examined, this technique would flag the presence of a new respiratory virus before the first COVID wave in 16 of them. On average, the alarm would go off around 13 weeks before the first wave peaked.
You can see how this would work in the United States here:
The peak of not-flu influenza-like illness in the US was the week of March 9, 2020. Before schools closed. Before the pandemic was declared by the WHO. Right around that tweet from President Trump.
Some countries would have had even more lead time using this system, like Spain, whose first outlier occurred in January.
And, of course, combining the results from multiple countries would give a clear signal that a pandemic of significant proportions was afoot long before it came to public consciousness.
The point is, these data are already being collected. Infectious disease surveillance systems like FluNet, if set up appropriately, might let us know about the next pandemic before it's too late. We just need to make sure that people with the power to make a difference are listening.
F. Perry Wilson, MD, MSCE, is an associate professor of medicine and director of Yale's Clinical and Translational Research Accelerator. His science communication work can be found in the Huffington Post, on NPR, and here on Medscape. He tweets @fperrywilson and hosts a repository of his communication work at www.methodsman.com.
Image 1: Our World in Data
Image 2: PLOS Medicine
Image 3: PLOS Medicine
Medscape © 2022 WebMD, LLC
Any views expressed above are the author's own and do not necessarily reflect the views of WebMD or Medscape.
Cite this: F. Perry Wilson. An Early Warning System for the Next Pandemic - Medscape - Jul 19, 2022.