“Both policymakers and voters need to know that we still cannot answer the most basic questions about this pandemic with the tools at our disposal,” says Douglass, the vocal Twitter critic of the type of study like this one Published week ago was about Sturgis and co-authored an upcoming article on similar shortcomings in the fast-growing Covid literature. And he’s not alone. Other scientists also fear that the rush to use bad or incomplete data to provide answers, and answers on the impact of large gatherings, will ultimately do little to help end the pandemic. Indeed, they think, it can do more harm than having no answers at all.
“Look, these are really urgent questions we need to clarify,”
In fact, South Dakota officials were quick to reject the researcher’s findings. A spokesman for the state health department told WIRED in an email that the study “does not match the effects we saw in the field.” Governor Kristi Noem, a Republican who avoided mask mandates and other disease prevention measures, went further, telling Fox News that the researchers “did some calculations on the back of the napkin, invented some numbers, and published them.”
This is an alternate universe in itself. The San Diego State University team’s recognized methodology is standard for economists trying to answer questions about the societal impact of a sudden event. But that doesn’t mean they were the right methods of answering This Question, nor that the study is flawless. In particular, Jha questions the lack of so-called “forgery analyzes”. These tests can be used to review your work and make sure they are measuring what you think. One way to do this is to imagine a world where the rally happened three weeks earlier than it actually did and then drive the models again. If the rally (the actual rally) was really the cause of the coronavirus spikes, an analysis with this fictitious Sturgis date should not reveal any changes in the fall rates – because with this analysis the actual rally is still three weeks in the future. When you see a change you know something other than the Sturgis rally is causing the spike. Without these types of controls, it is more difficult to file claims for damages.
The bigger problem, however, Jha says, is that the real world data just doesn’t match the study’s estimates – except in South Dakota. The state health department has so far identified 124 people who attended the rally and who later tested positive for Covid-19. A department spokesman declined to say how many close contacts these people are currently being monitored to see if they get sick. But in the month after the rally, South Dakota’s daily average for new diagnoses more than tripled from 82 to 307. That’s a noticeable spike that coincides pretty well with Sturgis in timing, says Jha.
However, based on cell phone data, 90 percent of rally goers came from outside of the state. However, if you look at the counties identified by the Sturgis study authors as having the highest number of participants, you don’t see any similar fluctuations. According to their model, in the three weeks following the rally, Covid-19 infections rose the most in places like Maricopa County, Arizona; Hennepin County, Minnesota; and three counties around the Denver metropolitan area of Colorado. In these countries, the authors found that the Sturgis rally was linked to a 13.5 percent increase in Covid-19 cases. According to the state health department, Maricopa County actually saw a decrease in the number of cases reported daily – from a 7-day average of 775 at the start of the rally to 266 a month later. In two of the three Colorado counties, the number of new cases declined or decreased after the rally. This trend has also been seen in a few other high-flow countries, including San Diego, Los Angeles, and Clark County, home of Las Vegas.