Day 144: The Real Surge

A new paper in Science Translational Medicine uses influenza-like illness (ILI) surveillance data to document an outsized epidemic surge in March in the US. [Paragraph breaks added for legibility.]

We identified excess ILI cases by first subtracting cases due to influenza and then subtracting the seasonal signal of non-influenza ILI (Fig. 1). Our approach identified known outbreaks of respiratory disease, including the recent outbreak of Respiratory Syncytial Virus that occurred in Washington state in December 2019. Starting in March of 2020, many states, including Washington, New York, Oregon, Pennsylvania, Maryland, Colorado, New Jersey, and Louisiana, showed a surge in number of non-influenza ILI cases in excess of seasonal norms.

For example, in the fourth week of March, 2020, New York State saw approximately 2 times higher non-influenza ILI than it had ever seen since the inception of the ILINet surveillance system within the US. We found that 10.2% of all outpatient visits in New York State during this time were for ILI that could not be explained by either influenza or the normal seasonal variation of respiratory pathogens (8.0% to 11.2% credible set).

As the seasonal surge of endemic non-influenza respiratory pathogens declined toward the later weeks in March, this excess ILI correlated more strongly with state-level patterns of newly confirmed COVID-19 cases, suggesting that this surge is a reflection of ILI due to SARS-CoV-2 (Pearson ρ>0.35 and p<0.05 for the last three weeks; fig. S2).

The US-wide ILI surge appeared to peak during the week starting on March 15 and subsequently decreased in numerous states the following week; notable exceptions are New York and New Jersey, two of the states that were the hardest hit by the epidemic, which had not started a decline by the week ending March 28.

Because they use all hospital visits as part of their ILI surge estimates, they go into some detail about how care-seeking and care-avoidance behavior (due to the growing fear of COVID-19, which would keep non-coronavirus-suspecting patients away but encourage coronavirus-suspecting patients to seek care) could have significantly affected their surge estimate. However, they conclude that non-care-seeking behavior for both mild ILI and non-ILIs cancelled each other out, leaving their original surge estimate intact.

Next up is consideration of mild and asymptomatic cases that would not have appeared in their surge estimates at all, leading them to the following conclusion:

Together these additional contributions from sub-clinical cases correspond to a mean clinical rate of 32% (the overall rate at which SARS-CoV-2 cases seek medical care) and a lower bound of 8.7 million SARS-CoV-2 infections between March 8th and March 28th (95% credible set 8.0 million to 9.4 million). Prevalence estimates for each state within this time-period are shown in fig. S4.

The figures, including the (tiny) state-by-state ILI graphs and that prevalence chart for the country, are in a separate PDF. For Massachusetts, they estimate prevalence at around 4%, making us #6 after NY, NJ, Louisiana, CT, and Maryland. (While all their data is state-based, it's not all labeled by state.)

Using deaths, they estimate a doubling rate in the last three weeks of March of 3 days, similar to the doubling rate of 2.65 days in Italy. They discuss various factors involved that could push the doubling time as high as 4 days, or lower it, including the lag time between the experience of symptoms and seeking care (and thus getting into the ILI surveillance system—4 days) and from seeking care to death (for their death estimates—11 days).

It's sort of a bright side that, in their estimation, the pandemic peaked for the US in mid to late March, and all the alleged peaking later was just about testing coming online as the disease itself petered out, either because of social distancing, advancing spring, COVID-19 having already consumed a large proportion of the susceptible population, or a combination of such factors.

On a different topic, PlagueBlog must note the loss of a popular science blog, Slate Star Codex, due to the New York Times' threat to doxx the author, who is a psychiatrist with various professional and personal reasons to want to maintain his anonymity:

When I expressed these fears to the reporter, he said that it was New York Times policy to include real names, and he couldn’t change that. After considering my options, I decided on the one you see now. If there’s no blog, there’s no story. Or at least the story will have to include some discussion of NYT’s strategy of doxxing random bloggers for clicks.

I want to make it clear that I’m not saying I believe I’m above news coverage, or that people shouldn’t be allowed to express their opinion of my blog. If someone wants to write a hit piece about me, whatever, that’s life. If someone thinks I am so egregious that I don’t deserve the mask of anonymity, then I guess they have to name me, the same way they name criminals and terrorists. This wasn’t that. By all indications, this was just going to be a nice piece saying I got some things about coronavirus right early on. Getting punished for my crimes would at least be predictable, but I am not willing to be punished for my virtues.

PlagueBlog will be replacing our previous SSC citations with archive.org links. Here's an archive link to all the coronavirus data and discussion that has been lost to the NYT's click-greed.

P.S. Massachusetts cases are up a fifth of a percentage point today. Also, the National Review reports on the Scott Alexander situation.