A Stanford University study released this week calling restaurants potential “superspreader” sites is deeply flawed, the National Restaurant Association said in a strongly worded statement Thursday.
Researchers tracked the movements of 98 million Americans in 10 of the country’s largest cities to create predictive modeling of infection risk based on where people went, how long they stayed and how crowded those places were.
Using modeling, rather than actual contact tracing, is not definitive enough to point the finger at restaurants as a major source of coronavirus spread, the Association said.
“Absent contact tracing and public health data determining who among the groups actually contract the virus and without a higher level of specificity to where they contracted it, using a modeling exercise with anonymized location data pulled from mobile devices is not an acceptable way to determine that restaurants are a likely cause for virus transmission,” the industry advocacy group said.
Among the Association’s other issues with the Stanford research:
- It did not consider safety protocols put in place at a wide number of restaurants, including mask wearing, social distancing, air flow upgrades, outdoor dining and surface disinfecting.
- Its sample size of 10 major metropolitan areas did not reflect the entire country.
- The research took place from March through May, a time in which dine-in operations were suspended across much of the U.S. The study likely recorded data from people picking up takeout orders and not people sitting for a full meal.
- Stanford’s research did not differentiate between different types of restaurants. “Not all restaurants carry equal risk,” the Association said.
- The study used “secondhand” datasets, with death counts and confirmed cases coming from New York Times media reports, rather than directly from public health agencies.
- Mask wearing was not part of the analysis.
“Due to the complexities of the real world, predictive models are inherently fraught with error even under the best of circumstances,” the Association noted. “In this case, the researchers' model seems to fit their data, so they conclude that it is reasonably accurate. However, this is not enough to assert that restaurants are a significant source of risk across the entire United States.”
The Texas Restaurant Association, in a statement, expressed similar concerns about the study’s computer modeling and general scope.
“We’re eight months into this pandemic, and the stakes are simply too high for our public health and for our economy to continue to let sensational news headlines—rather than health data—drive our collective efforts to combat COVID-19,” the Texas Restaurant Association said. “Restaurants will continue to implement all of the evidence-based strategies doctors advise so we can provide a safe alternative to the unregulated, in-home gatherings that we know are driving the spread of COVID-19.”