As the COVID-19 pandemic has raced across the globe and killed millions worldwide, people continue to ask: Could this have gone differently? The early months of 2020 marked a turning point where public health efforts shifted from containing the virus’ spread to mitigating its damage once that was no longer possible. But what if we had known about future surges in infection rates earlier, or had a better sense of the peaks?
A team of researchers from the University of Washington set out to determine whether or not the surveillance of influenza-like illnesses would have clued countries in on viral peaks before they occurred. In a paper published Tuesday in PLOS Medicine, they lay out the case that an automated method of tracking flu infections “could have informed national and global responses to SARS-CoV-2 during the rapid spread of this novel pathogen in early 2020.”
The new study is centered around data from FluNet, an online tool for global virus surveillance that includes illnesses presenting with flu-like symptoms that test negative for influenza. The researchers built a program that alerted them to dates when more people than usual seemed to have flu-like illnesses, then matched up the timelines with recorded COVID-19 peaks. Across 16 different countries, they found that their outlier method alerted them to possible surges weeks before COVID peaks were first identified.
ADVERTISEMENT
In other words, co-opting flu surveillance systems like FluNet could have been a possible tool for public health officials to use before the full scope of an illness’ spread is known. Improving these existing networks and developing new ones is “essential” to future pandemic preparedness, the authors wrote in the study.
But investing in such illness surveillance systems and blindly trusting them may not be a silver bullet solution—in part because not everyone agrees that the paper’s findings stand up to scrutiny.
Justin Silverman, a statistician at Penn State University, said that parts of the group’s methodology should raise questions. The pandemic affected U.S. states unevenly and in waves. The same think happened abroad, suggesting that an approach at the country level is too broad and ineffective.
Additionally, any alert tool must balance recognizing minute trends with minimizing the number of false alarms. If implemented in the years before the pandemic, the program used in the study would have sent out annual alerts for some countries, which could have given decision-makers “alert fatigue,” Silverman said.
“My personal opinion is that there was no early warning indicator at the country-wide level that would possibly have given us a leading edge on intervention,” Silverman told The Daily Beast.
Silverman, who also researched the utility of applying influenza surveillance networks to COVID-19, said that estimating infection prevalence might be a better use of these tools than trying to predict surges before they occur. Knowing that COVID-19 was widespread in the U.S. before hospitalizations and antigen tests indicated as much could have prevented unnecessary restrictions and directed resources where they were most needed, he added.
With that in mind, surveillance networks are a cost- and resource-efficient way to inform crucial public health decisions—even if they aren’t true crystal balls.
“From a policy perspective, I think it's a no-brainer to monitor influenza-like illness and watch for outliers,” Silverman said.
